Testing World Magazine

You are here:: Home >; Testing World Magazine
<Back to Home

Testing World Magazine

* Required Field

* First Name

* Last Name

* Company

* Email address

* Job Role

* Country

Phone

Send me updates on downloads, events, and promotions

Cancel

TESTING WORLD Jan Issue,2011 A Spirent magazine © 2010 Spirent Communications, Inc. All of the company names and/or brand names and/or product names referred to in this document, in particular the name “Spirent” and its logo device, are either registered trademarks or trademarks pending registration in accordance with relevant national laws. All rights reserved. Specifications subject to change without notice. www.spirent.com Spirent Communications 1325 Borregas Avenue Sunnyvale, CA 94089 USA SALES AND INFORMATION sales-spirent@spirent.com www.spirent.com Asia Paciﬁc T: +852 2511.3822 Europe, Middle East, Africa T: +33 1 6137.2250 Americas T: +1 800.SPIRENT +818 676.2683 Opinion Securing the Cloud an Impossible Feat? Think Again New Product Spirent TestCenter 3.60 Spirent Avalanche 3.60 Spirent Virtual Drive Test (VDT)–Conversion Tool Services Spotlight on Wireless Services Q&A on testing Why Test Your New Cloud? 10 Things You Need to Know Mobile Backhaul Testing Ethernet has been a key player in the Local Area Network(LAN) and Metro Area Network (MAN) for some time. It is fast becoming the technology of choice for mobile backhaul expansion because it’s a workhorse that offers reliable packet-switched, high bandwidth characteristics that will carry the IP bandwidth load required by expanding services. www.spirent.com 04 06 26 29 31 32 34 36 14 09 15 17 22 20 Content News & Analysis Opinion Securing the Cloud an Impossible Feat? Think Again New Product Spirent TestCenter 3.60 Spirent Avalanche 3.60 Spirent Virtual Drive Test (VDT) –Conversion Tool Services Spotlight on Wireless Services Q&A on testing Why Test Your New Cloud? 10 Things You Need to Know Spirent Update Ethernet is fast becoming the technology of choice for mobile backhaul expansion. You will get more details about mobile backhaul testing from this cover story, including: Ethernet Mobile Backhaul and Ethernet Service Veriﬁ cation Mobile Backhaul Ethernet Equipment Market is Booming Effectively Testing Mobile Backhaul with STC STC Live Mobile Backhaul Solution Testing Next-generation Mobile Networks Testing Methodology for Mobile Backhaul 03 Content Editor Email: janet.peng@spirent.com Messaging With the forthcoming New Year, on behalf of Spirent Communications, I’d like to express my sincerest regards and best wishes to friends who have supported us. I wish all of you a prosperous and happy New Year! In the meantime, I’d like to express my sincere thanks for your constant support to Spirent Communications. Great changes in telecommunication industry and rapid development of communications in APAC has posed higher standards for Test System. With profound understanding of customer requirements and market demands, Spirent has provided advanced technologies / products/ test solutions and ﬁ rst-class services, thereby becomes a well-recognized leader in T&M market. Looking forward to 2011, Spirent will continue to increase investments in APAC and pay key attention to the following test requirements : 1. Cloud Service is getting prevailing across the industry. Spirent continues to promote its PASS (Performance, Availability, Security and Scalability) test solution, and lead Cloud Computing test market. 2. Data Center’s scale continues developing, Bandwidth thirsty never stops – all result in strong demands for higher density, higher speed network device to be developed and deployed. Ultra high density 10GE and 40GE/100GE requirements start to soar in 2011. Spirent makes itself ready for the challenge with plan of releasing series of new products in 2011. 3. Mobile broadband application is driving a lot investments by Mobile SPs on Mobile Backhaul and Mobile Packet Core area, where a lot new technologies are emerging and evolving. Spirent provides comprehensive test solutions in both areas, including IEEE-1588v2, SyncE, MPLS-TP, ERP, GTPv2, Security Gateway, ePC, etc. to address biggest concerns from both Mobile SPs and NEMs. 4. With depletion of IPv4 address space, IPv6 commercial deployment will enter a strong growth stage in next two years. Spirent had well setup itself as a leader in IPv6 conformance, functional and performance testing for years. Looking forward, Spirent commits to optimize existing solution as well as develop new test features to meet latest IPv6 technology evolution. 5. In the area of mobile terminal, Spirent will enhance the performance testing of 3G UMTS & CDMA mobile terminal testing and provide end to end solution. 6. Based on MIMO and beamforming technology, Spirent continuously improves the accuracy, scalability and easy to use for high-end antenna testing, and meets evolving customer requirement . 7. With the development and deployment of LTE, Spirent will provide TDD/FDD-LTE testing solution on single platform 8. Spirent will continue to improve testing efﬁ ciency by delivering the industry’s most comprehensive end-to-end automation solutions, covering the entire breadth of test automation requirements. In the coming year, we will continue to provide state-of-the-art solutions and world-class services, and help customers win in both their products and services in the shortest time, and therefore to become the best partner of their customers. Once again we thank you for your support and wish everybody a happy and a prosperous New Year! Global Vice President & General Manager, APAC Gene Zhang Highlights of 2011 Mobile Backhaul Testing Cover Story 04 05 News & Analysis News & Analysis 1901™ Broadband Powerline Standard for 500 Mbps Communications Approved New Smart Grid Standard Provides Common Data Formats The World In 2010 Source: ITU World Telecommunication/ICT Indicators database New Ethernet Amendment Addresses Resource Reservation New 802.11™ Amendment Covers DLS Recently IEEE announced the ratiﬁ cation of the IEEE 1901™ Broadband over Power Line (BPL) standard. Sponsored by the IEEE Communications Society, this globally recognized BPL standard will be a key enabling technology for a wide range of applications including smart energy, transportation and Local Area Networks (LANs) in both the home and the enterprise. Networking products that fully comply with IEEE 1901 will deliver data rates in excess of 500 Mbps in LAN applications. In ﬁ rst- mile/last-mile applications, IEEE 1901-compliant devices will achieve ranges of up to 1500 m. The technology speciﬁ ed by IEEE 1901 uses sophisticated modulation techniques to transmit data over standard AC power lines of any voltage at transmission frequencies of less than 100 MHz. Although IEEE 1901 has been widely recognized as the standard that will enable universal communications in Smart Grid applications, it will also have signiﬁ cant impact in other applications. In the transportation sector, for example, the standard's data rates and range make it possible to deliver A/V entertainment to the seats of airplanes, trains and other mass transit vehicles. Electric vehicles can download a new entertainment playlist to the A/V system while the car in charging overnight. In the home, PLC will complement wireless LANs by providing a link through walls and other RF impediments as well as over distances beyond the normal range of wireless networks. It will complement wireless networks in hotels and other multistory buildings by carrying multimedia data over the longer distances and allowing wireless to complete the communication link over the last few meters. IEEE 1901 will also beneﬁ t utilities, service providers, and consumer electronics companies - anyone with a stake in smart grid technologies - as well as smart-meter providers and home appliance manufacturers. The IEEE has approved and published its latest standard for smart grids, IEEE C37.239™, "Standard for Common Format for Event Data Exchange (COMFEDE) for Power Systems." The standard deﬁ nes a comon format for data ﬁ les used for the interchange of various types of event data collected from electrical power systems or power system models. It deﬁ nes an XML schema and also includes a sample ﬁ le. The rise of 3G The latest IEEE 802™ amendment is available: IEEE 802.1Qat™, "IEEE Standard for Local and Metropolitan Area Networks-Virtual Bridged Local Area Networks Amendment 14: Stream Reservation Protocol (SRP)." This enhancement to the Ethernet protocol allows resource reservation at the Ethernet layer. This amendment speciﬁ es protocols, procedures and managed objects usable by existing higher layer mechanisms, which allow network resources to be reserved for speciﬁ c trafﬁ c streams traversing a bridged local area network. Before this version, 802.1Qau-2010,IEEE Standard for Local and Metropolitan Area Networks - Virtual Bridged Local Area Networks Amendment 13: Congestion Notiﬁ cation. This amendment speciﬁ es protocols, procedures and managed objects that support congestion management of long-lived data ﬂ ows within network domains of limited bandwidth-delay product. This is achieved by enabling bridges to signal congestion to end stations capable of transmission rate limiting to avoid frame loss. The IEEE has published an amendment to the IEEE 802.11™ standard covering extensions to direct- link setup (DLS). The amendment creates a new DLS mechanism which will be increasingly important as IEEE 802.11 is applied to new market applications such as video streaming. The new DLS mechanism does not require access point upgrades, supports power save mode, and continues to allow operation of DLS in the presence of existing DLS capable access points. The full name of the new amendment is IEEE 802.11z™, "IEEE Standard for Information Technology— Telecommunications and information exchange between systems—Local and metropolitan area networks—Speciﬁ c requirements Part 11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) speciﬁ cations Amendment 7: Extensions to Direct-Link Setup (DLS)." This amendment provides direct-link setup enhancements to the IEEE 802.11 MAC and PHY, extending direct-link setup to be independent of the access point (AP), and adding power save capabilities. The direct-link setup is made independent of the AP by tunneling the protocol messages inside data frames. New Standard is Enabling Technology for Smart Grid, Transportation and LAN Applications By the end of 2010, there will be an estimated 5.3 billion mobile cellular subscriptions worldwide,including 940 million subscriptions to 3G services. Access to mobile networks is now available to 90% of the world population and 80% of the population living in rural areas. People are moving rapidly from 2G to 3G platforms, in both developed and developing countries.In 2010, 143 countries were offering 3G services commercially, compared to 95 in 2007. Towards 4G: a number of countries have started to off er services at even higher broadbandspeeds, moving to next generation wireless platforms – they include Sweden, Norway, Ukraineand the United States. Connecting homes While in developing countries 72.4% of households have a TV, only 22.5% have a computer andonly 15.8% have Internet access (compared to 98%, 71% and 65.6% respectively in developedcountries). At the end of 2010, half a billion households worldwide (or 29.5%) will have access to the Internet. In some countries, including the Republic of Korea, Netherlands and Sweden, more than 80% ofhouseholds have Internet access, almost all of them through a broadband connection. The number of people having access to the Internet at home has increased from 1.4 billion in2009 to almost 1.6 billion in 2010. Figu res ormat f er syst ms or po m mode r (PH ndme Opinion Opinion 06 Securing the Cloud an Impossible Feat? Think Again Security typically sits at the border of the LAN and WAN, protecting the data center infrastructure from threats. A ﬁ rewall inspects all incoming and outgoing trafﬁ c, passes through legitimate trafﬁ c and blocks malicious trafﬁ c from the outside. In addition, a ﬁ rewall can sit at the top-of-rack or end-of- row, monitoring trafﬁ c on the LAN to detect and contain inter-server threats from spreading through the LAN. These could be attacks that somehow got past the ﬁ rewall or threats introduced internally, either unconsciously by uploading an infected ﬁ le or intentionally through sabotage. In the typical scenario, it is not feasible to deploy an IPS in front of every server. The best that can be done is to have an IPS per row or per rack and attempt to contain inter-server threats within a small segment of the data center. In addition, nothing sits inside a server, detecting and stopping an intra-server threat, whether it is a hacked hypervisor or a rogue VM attacking and infecting other VMs in the same server. For example, a compromised VM could send counterfeit transactions, destroying the integrity of back-end databases. Since all the trafﬁ c that leaves the physical server appears legitimate, traditional security systems can't detect and stop this breach. Infra/Inter/Intra Vulnerabilities Traditional data centers have inter-server and infrastructure vulnerabilities, such as the possibility of performance and security weaknesses internally between servers, externally at the gateway, and in the end-to-end network. Virtualization intensiﬁ es these potential threats and adds another level of vulnerability, intra-server, i.e., threats between VMs inside a single physical server. Infrastructure Traditional end-to-end testing validates the performance of an entire system. System testing is even more important in the era of virtualization. With dozens of VMs per physical server, the amount of trafﬁ c one box can generate increases dramatically, easily ﬁ lling a 10 Gigabit Ethernet link. The cloud can be composed of hundreds or thousands of physical servers. Inter-server Device testing evaluates the performance of a device interacting with other devices. For example, testing a security appliance involves sending legitimate trafﬁ c mixed with malicious trafﬁ c to the appliance and evaluating its ability to deﬂ ect threats while forwarding legitimate trafﬁ c at acceptable levels. The increase in utilization due to virtualization means an increase in trafﬁ c, placing more demands on the performance of the security appliance. Intra-server Now that we have multiple applications running in separate VMs on a single server, we have the possibility of security threats residing completely inside a physical server. Intra-server trafﬁ c never sees the network, so traditional methods of implementing 07 The rapid rise of cloud computing has delivered cost and productivity beneﬁ ts to thousands of organizations as over 200 cloud providers have emerged in the last decade. But questions of cloud security reveal that the growth of the networking and computing capabilities has outstripped the development of technologies to protect the cloud from cyber attacks. Greg Day, security analyst at McAfee, told ComputerWeekly.com, "As cloud computing gains popularity, cyber-criminals are likely to target these services to steal information for ﬁ nancial gain." At the heart of the issue is virtualization, the ability to run multiple server instances inside virtual machines (VMs) on a single physical server. This basic element is both the foundation of cloud computing and the source of new vulnerabilities that are already being exploited. Traditional vs Virtual Security When implemented and conﬁ gured correctly, current cyber security solutions do a good job of detecting and blocking a wide range of malicious trafﬁ c from outside and even inside the data center. This is true because mature technology underlies security applications like IDS, IPSand DPI. Validation is the essential element in the technology cycle that drives maturity. Current security technology reached maturity through the iterative development of test methodologies that assessed and validated speciﬁ c implementations. As we shall see, cloud-aware test methodologies are the key to bringing security to cloud computing. Some may assume that existing security solutions are adequate to protect the cloud. After all, the virtual servers reside on physical servers that are behind the ﬁ rewall. To see why this is not the case, we must look at the relationship between virtualization and security, more speciﬁ cally, where security is traditionally implemented in a data center. and testing security are completely ineffective for intra- server threats. If a rogue application is spawned in a VM and launches a DOS attack on other VMs on the server, a software appliance in the DMZ will never know. Virtual Security for Virtual Machines Traditional security approaches are inadequate to protect the cloud because they can't detect and deﬂ ect intra-server threats. Virtual machines require virtual ﬁ rewalls. Figure 2: Today's virtualized environment network diagram A virtual IPS performs the same functions as a physical IPS. The difference is where it is located. In the case of a virtual appliance, it resides in a service VM on the physical server along with the application VMs. A redirect policy allows a virtual controller to inspect and control VM-to-VM communications and direct the trafﬁ c to the appropriate appliance, whether physical or virtual. This arrangement places a virtual IPS in front of every connection to allow the trafﬁ c to and from every VM to be inspected. A cyber security system that combines physical IPS appliances with virtual IPS appliances has end-to-end visibility of the data center network, from the DMZ at the demarcation point to every VM in every server, and all devices of interest in between. Metrics of Virtual Service: PASS Here is where cloud-aware test methodologies come into play. Like the traditional data center, the virtualized data center has fundamental and critical network attributes - performance, availability, security, and scalability (PASS). Established test methodologies answer the critical questions related to the PASS attributes. However, virtualization fundamentally changes the environment that these methodologies address. Performance Traditional over-provisioning methods of ﬁ xed resources - physical servers, storage drives, network switches-no longer apply in the virtualized environment. Cloud computing offers tangible beneﬁ ts, but security issues have the potential of negating those beneﬁ ts. A virtualized data center must be supported by a virtualized security system, which must be validated by a virtualized test systems and test methodologies. Figure 1: Traditional end-to-end client server network diagram About Author Ankur Chadda is a product m a r ke t i n g m a n a g e r a t Spirent Communications and works very closely with cloud security and the applications market segment to enable them to deploy new or upgraded networks. rabilities that are xploit d. Ethernet has been a key player in the Local Area Network (LAN) and Metro Area Network (MAN) for some time. It is fast becoming the technology of choice for mobile backhaul expansion because it’s a workhorse that offers reliable packet- switched, high bandwidth characteristics that will carry the IP bandwidth load required by expanding services. Moreover, it delivers signiﬁ cant operational cost savings when compared to traditional connection-centric technologies like TDM. Additionally, new synchronization protocols such as ITU SyncE and IEEE 1588v2 have overcome the earlier downsides, like latency and jitter, to deploying Ethernet in the mobile backhaul, making the opex savings and superior bandwidth beneﬁ ts that Ethernet delivers a compelling play. While Ethernet is very attractive for all the right reasons, it brings with it new challenges that service providers and MSOs have to overcome. Performance and QoS (Quality of Service) requirements must be monitored, processes need to be established to guarantee service level agreement compliance and operators must have monitoring in place to ensure user QoE for applications like video, Web browsing, email and voice calls. Being able to efﬁ ciently and economically ﬁ nd, diagnose and ﬁ x problems in the mobile backhaul will determine proﬁ tability and ultimately, survival for the service provider or MSO. 09 Opinion 08 Answering these questions is the goal, whether testing a legacy data center or a virtualized data center. Like the virtualization of a security application, the innovation of testing virtualization lies in extending the test endpoints. As the world of computing has employed the VM to provide the many beneﬁ ts of cloud computing, test systems have extended to the virtual level to validate the functionality of applications running in the VMs, and through the iterative development process, to facilitate improvements in performance, availability, security, and scalability, the critical metrics of data center efﬁ ciency. A virtual tester is a software-based test system implemented in a virtual machine. To the network devices under test, and to the test engineer, it looks and behaves exactly as if it were a hardware tester. A virtual tester makes it possible to test cloud security at all the levels it has impact: intra-server, inter-server and infrastructure. When assessing a cyber security system that employs virtual and physical appliances, testers reside at the endpoints to generate trafﬁ c and accumulate results. •Intra-server: Virtual testers for each VM in the physical server serve as endpoints. •Inter-server trafﬁ c: A virtual tester for each VM in the separate physical servers can serve as endpoints, or a virtual tester on one end and a physical tester on the other. •Infrastructure: Virtual testers for each VM in the test serve as endpoints and a physical tester at the gateway serves as the other. The result is end-to-end testing of any IDS/IPS scenario, whether the endpoints span the whole of the data center or reside in a single physical server. A recent test conducted by Broadband Testing demonstrated the use of cloud-aware PASS methodologies to validate a cloud-aware cyber security solution. Conclusion Cloud computing offers tangible beneﬁ ts for increasing efﬁ ciency and reducing capital and operating costs for enterprises and other organizations, but security issues have the potential of negating those beneﬁ ts. A virtualized data center must be supported by a virtualized security system, which must be validated by a virtualized test systems and test methodologies. Cover Story Mobile Backhaul Testing At the service level, the cloud designer must take this into account by ensuring an adequate number of VM instances are provisioned to make dynamic access possible for all users. Cloud security must deliver the maximum number of new connections per second and ﬁ rewall bandwidth throughput while blocking threats and malicious trafﬁ c. Availability The traditional methods of providing local redundancy must also be reconsidered in a virtualized environment. Servers that can support 1,000 or more VMs can become a single point of failure if appropriate approaches to VM load balancing, automated resource scheduling and live migration to other hardware are not built into the design. Cyber security in the cloud requires maintaining optimum application response time at maximum throughput. Security Traditionally, cyber security is placed in strategic physical locations, such as at the WAN edge where requests and trafﬁ c from the Internet can be ﬁ ltered and decrypted. However, geographic locations of physical servers have less meaning in a virtualized cloud, as users might be tapping resources from VMs located on one of any number of servers or even data centers. Virtual security must be cloud-aware. In the case of live migration, where a VM moves to another server with VMotion, the security solution must migrate the proﬁ le to allow legitimate trafﬁ c access to the new physical machine to avoid downtime for the end user. Scalability The promise of inﬁ nite scale is appealing, but the elasticity of the physical infrastructure has ﬁ nite limits. Addressing this risk requires a well-thought-out network infrastructure where aggregation and core interconnects do not become the bottlenecks of the elastic demand and scale that the cloud promises, maintaining the maximum number of secure concurrent connections at maximum throughput. Virtual Test Systems for Virtual Security For both traditional and virtual data centers, testing answers questions related to PASS. In particular, testing provides the answer to the question: How secure is any given cloud? Testing a cyber security solution addresses two vital questions at a high level: 1. Does the solution block all threats while allowing legitimate trafﬁ c to pass? 2. How does the solution affect throughput, performance and scalability? Ethernet Mobile Backhaul and Ethernet Service Veriﬁ cation tual t er or each VM in the ysical er for each VM in the tual t er or each VM in the Leverage rapid transition to Carrier Ethernet for wire line trafﬁ c; enabling a single integrated wire line and mobile backhaul network Reduce management and maintenance costs Optimize the network for packet data trafﬁ c (Most mobile trafﬁ c is broadband/IP centric) Overcome TDM (T1/E1) services scalability limitations 10 11 MBH is complex & multi-layered Must ensure resilience, OAM & timing capabilities similar to TDM networks Timing / Synch Firstly, the asynchronous nature of Ethernet provides certain transmission challenges.For example, Time Division Multiplexing (TDM) services such as T1/E1 and SONET/ SDH require synchronized clocks at both the source and destination nodes. Similarly, wireless base stations require synchronization to a common clock to ensure a smooth call hand-off between adjacent cells. While there are several ways to achieve synchronization over Ethernet, The Precision Time Protocol (PTP) is a high precision time synchronization protocol for networked measurement and control systems. Accuracy in the sub-microsecond range may be achieved with low-cost implementations. It is deﬁ ned in the IEEE 1588-2002 and 1588-2008 standards, ofﬁ cially entitled "Standard for a Precision Clock Synchronization Protocol for Networked Measurement and Control Systems". Another gaining momentum is Synchronous Ethernet (SyncE). SyncE uses the physical layer interface to pass timing from node to node in the same way timing is passed in SONET/ SDH or T1/E1. This gives telecom and wireless providers conﬁ dence that networks based on SyncE will be not only cost-effective, but also as highly reliable as SONET/SDH and T1/E1 based networks. As interest from carriers and service providers grows, many Ethernet equipment vendors are developing SyncE enabled equipment targeting this lucrative new market. Trafﬁ c Setup & Management Bandwidth needs from each Base Station connection will increase 5x to 50Mbps** in two years as 3G/4G rollout accelerates. Real-time voice, video and back-ofﬁ ce applications are sensitive to poor network performance. MPLS-TP is a simpliﬁ ed version of MPLS for transport and marries the advantage of packet transport with the reliability of TDM networks, and based on the PWE3 and LSP forwarding architectures. MPLS-TP pseudowires provide circuit emulation services for ATM, TDM, Frame Relay, and Ethernet trafﬁ c originating from the base stations. Multi-segment PWs consist of multiple single-segment pseudowires stitched together at Switching-PEs Carrier Ethernet OAM for MBH Assuring consistent level of service across multi-tier networks requires monitoring of both the end-to-end and segment QoS with notiﬁ cation of any degradation in circuit quality (delay, jitter) to the cellular operator. On the TDM network any discontinuity is immediately identiﬁ ed and delay is hardly an issue. But since Ethernet is a framed technology, it is not known when to expect a frame, we need to monitor the service from one side to the other. Ethernet OAM helps monitor each service in terms of availability, frame loss and delay. In addition for each service thresholds can be deﬁ ned alarming when the service deviates the SLA. So Ethernet OAM enables fault detection, but TDM OAM tools are still required for fault isolation. Existing standards including: IEEE 802.3ah : Link OAM; IEEE 802.1ag : Connectivity Fault Management (Service OAM); ITU-T Y.1731 : Fault Management & Performance Monitoring Challenges of Ethernet Services for Mobile Carriers With the explosion in demand for Ethernet Business Services, it is imperative that mobile carriers adopt strategies that simultaneously support maintenance of their current networks, enable growth of next generation Ethernet networks and simplify service assurance for their technicians. Ethernet is by far the most robust, scalable and price prudent backhaul network alternative for the mobile carriers wanting to deploy 3G & 4G/LTE networks, which is why Ethernet is playing an ever increasing role in supporting the mobile backhaul infrastructure. Ethernet is compatible with almost every network technology, conﬁ guration or medium. It has become the choice for providers looking to eliminate single purpose service networks and transition to converged network topologies where multiple services can be efﬁ ciently and cost effectively provisioned and maintained. Ethernet provides these capabilities quickly and easily with high reliability while delivering the bandwidth needed for next generation services. Ethernet networks provide enormous beneﬁ ts allowing the Mobile Carrier to: Economically address growing bandwidth requirements currently constrained by the prohibitive costs of legacy networks Ethernet as Preferred Backhaul Architecture Challenges in deploying MBH As mobile carriers move to 3G and 4G/LTE networks, traditional DS1/E1 connections for the backhaul are not sufﬁ cient to handle this dynamic expansion in bandwidth. In addition, the high cost of bandwidth over traditional TDM SONET/ SDH technologies is also driving Ethernet as the preferred backhaul architecture. Ethernet & IP expected to dominate future MBH connections, as Ethernet solutions are more scalable and more cost- effective. But there are quite lot of challenges in deploying MBH: Cover Story Mobile Backhaul Testing Cover Story Mobile Backhaul Testing ering the bandwidth ne 13 element vendors equipped the network devices with alarms and event triggering capabilities, but the day has passed where these tools were enough because they are limited on several counts: Alarms and events are network centric and do not directly relate to the effect on the service. Alarms may be unreliable—setting up each network element to trigger on threshold crossings as the network expands and threshold values change requires a signiﬁ cant ongoing administrative effort and may not be done effectively. Alarms are triggered under conditions of hard failure while most Ethernet service issues are intermittent. Alarms are primarily targeted for core network elements and facilities. Most Ethernet service issues are in the access network. Providers are having problems despite their best alarming and monitoring efforts, which is why call centers exist. Testing and Diagnostic tools are different in that they are on-demand and can give the Ethernet service centers the detailed service-speciﬁ c troubleshooting information and analysis needed to ensure quick problem resolution. Visibility is The Answer Visibility into the performance of the backhaul network is critical now more than ever. Consider that Ethernet has always been viewed as a “plug and play” technology, delivering cost effective bandwidth with an architecture that is easy to maintain; is scalable and ﬂ exible. With the proliferation of smartphones like the iPhone, Droid and BlackBerry which are heavily data-centric, and the demands they make on the mobile backhaul and networks, Ethernet’s advantages have become even more compelling to providers. That said, the reality of being able to take advantage of Ethernet’s beneﬁ ts and simultaneously ensure carrier class quality across 3G, 4G and LTE networks requires an increased, more sophisticated, level of visibility not available from typical network equipment. Providers must manage their Ethernet backhaul. As the rule goes, you can’t manage what you can’t measure, and you can’t manage or measure something you have no visibility into. If you don’t have the right tools to gain visibility into or measure your network, you have no means of managing it. Spirent provides tools that provide the ability to provision, verify the provisioning of services up to the tower, verify performance and importantly, sectionalize or isolate problems. This is especially signiﬁ cant because many carriers don’t necessarily own the ﬁ ber or network between the tower and the central ofﬁ ce. They may contract that portion out to a third party who may possibly be a competitor, or cable company. Carriers need to be able to conﬁ rm that the connection is up, that it’s provisioned correctly, and operationally that they can monitor each section’s performance. They need to be able to rapidly see when the performance is degrading, determine why and whose network is at fault. Speciﬁ cally, determine where the trouble resides. Spirent’s tools allow you to gain granular visibility, measure and manage your Ethernet backhaul network. They give you the information you need to determine desired network performance and establish service level standards. Spirent’s tools also provide the all-important visibility into whether a trouble ticket can be remotely resolved or if a technician needs to be physically dispatched to resolve the problem, thereby giving the carrier the ability to signiﬁ cantly decrease operational expenses. Considering the amount of revenue associated with wireless customers, competition for subscriber dollars is ﬁ erce. Those providers who equip themselves with the ability to turn up the network correctly the ﬁ rst time will do so because they have the visibility into the network afforded by real-time performance monitoring, and the troubleshooting tools that ﬁ x problems quickly and cost-effectively. Consequently they will have a competitive advantage over other providers who rely solely on ﬁ eld technicians and network element statistics. The explosive growth in mobile bandwidth demand that is driving the expansion of the mobile backhaul will only serve to compound and complicate matters. In order to be successful, visibility into your backhaul and the network is the answer. Why Spirent? Spirent Communications is a global provider of integrated performance analysis and service assurance systems that enable the development and deployment of next-generation networking technology such as Internet telephony, IP Video, broadband services, 3G and 4G wireless, global navigation satellite systems, and network security equipment. Spirent has a long history of providing TDM, MPLS & Ethernet testing solutions that offer the realistic testing scenarios required to ensure Quality of Service (QoS). Many leading network equipment manufacturers and service providers utilize our solutions to meet their requirements for next- generation services whether using the Spirent TestCenter™ platform to complete the industry’s largest-ever public 10 Gigabit Ethernet test or selecting Spirent TestCenter Live solutions to better manage their expenses while meeting the service parameters essential to win in today’s competitive marketplace. Spirent is active with standards bodies and research organizations during the new technology evaluation process and provides test and diagnostic equipment to ensure the technology under evaluation meets appropriate standards. Our Service Assurance solutions utilize the expertise gained from the use of our lab test equipment by network equipment manufacturers to increase our ability to validate and resolve customer issues quickly and efﬁ ciently, with the latest technology. This total life cycle involvement with Ethernet technology uniquely positions Spirent to help mobile carriers drive their mobile backhaul migration to Ethernet or develop Ethernet mobile backhauls from the ground up, as well as deliver new Ethernet services to mass market faster with fewer resources and improved proﬁ tability, while enhancing customer satisfaction. Spirent‘s solution has been chosen by top-tier global network operators to assure their Ethernet services worldwide, resulting in the sales of more Ethernet test ports than any other vendor. In addition, Spirent was recently recognized by Frost & Sullivan as the global leader in Ethernet testing according to the research ﬁ rm’s World Gigabit Ethernet Test Equipment Market report. This award underscores Spirent’s unmatched Ethernet expertise. Spirent’s comprehensive, end-to-end solution portfolio extends from the lab with product development and veriﬁ cation, to the ﬁ eld for service provisioning, troubleshooting, and performance management, and to the in-home network to ensure QoE of multi-play applications. Spirent is proud of our award-winning family of Ethernet products and continues to invest in Ethernet Business Solutions. We are committed to maintaining our leadership in developing the Ethernet performance, test and diagnostic tools the industry requires to meet the ever increasing demand for bandwidth. Competition between the various providers has intensiﬁ ed, making it imperative that service assurance organizations are equipped with the appropriate tools. The amount and granularity of information available directly from the Ethernet switch does not address the majority of service level questions. The information provided by the switches is typically limited to aggregated counts of incoming as well as outgoing packets which doesn’t allow providers to sufﬁ ciently evaluate critical diagnostic information from a speciﬁ c port such as: Network performance: Loss, latency, and jitter Trafﬁ c: Frame sizes, protocol distribution, utilization, top talkers, top applications, etc. Speciﬁ c service: VoIP, data and video analysis Alarm Monitoring Versus Testing and Diagnostics Typically, as new services are deployed, Tier 1 technicians have little expertise and a high percent of the trouble issues are routed to a limited number of highly paid Tier 2 or Tier 3 technicians in the service centers or Network Operations Centers (NOC). The most common Ethernet service assurance tools in use today are fault or performance tools. Neither is service-centric. Instead they attempt to solve problems by looking at the network and aggregations of services. These tools, which represent a traditional approach, are absolutely necessary for service assurance and are a good start but they are not sufﬁ cient by themselves. Up until recently, this standard approach worked because network 12 Cover Story Mobile Backhaul Testing Cover Story Mobile Backhaul Testing precision and scalability. The highly accurate timing inherent in Spirent TestCenter architecture ensures the nano-second accuracy required for time sensitive applications such as Ethernet mobile backhaul without the need for additional test equipment. By combing Carrier Ethernet, MPLS and timing packages, the Spirent TestCenter system provides the industry’s most complete solution for testing converged mobile backhaul networks and devices. Applications Precision Timing Protocol compliance, performance, and interoperability testing for mobile backhaul applications Offering can be used by Network Equipment Manufacturers & Service Providers to ensure that mobile users won’t suffer from dropped calls or corrupt data Combine with Spirent TestCenter Carrier Ethernet, Routing and MPLS base packages for complete mobile backhaul system testing Combine with Spirent’s Network Impairment Emulator to simulate exhaustive set of real-world network conditions PTP remote testing using GPS synchronization Features & Beneﬁ ts Spirent TestCenter has a highly accurate internal clock with 2.5ns resolution. This accuracy means that no additional devices (GPS antennas, etc.) are required to ensure PTP accuracy, thereby simplifying the test setup Optional support for external BITS and GPS time sources. Support for test scenarios where Spirent TestCenter is not the master clock and remote test scenarios where Testing with Spirent TestCenter ensures IP & Ethernet based Mobile Backhaul solutions offer the highest quality of experience while meeting ever increasing bandwidth demands. Validate performance of IEEE 1588v2, MPLS-TP, and MPLS and Ethernet OAM with the accuracy and scale demanded by mobile networks Test backhaul capacity with immediate cause and effect analysis for each queue and real-time monitoring of over 40 per-stream metrics including packet jitter, latency, re- ordering and loss Timing & Synchronization testing In the absence of TDM synchronization, timing must be distributed to base stations via alternative methods. Synch is required to ensure QoE during mobility events (handoffs, fallback). IEEE 1588v2 deﬁ nes means to distribute timing over packet between master and slave clocks.Synchronous Ethernet distributes timing signal over the physical layer. Using ESMC protocol & Synchronization Status Messaging, networks elements can ensure supplied frequency can be traced to high quality source. IEEE 1588 and Synch Ethernet allow operators to offer converged services on one cost-effective network Spirent TestCenter IEEE 1588v2 Timing and Synchronization Base Package provides support for the Precision Timing Protocol (PTP). The package allows Spirent TestCenter ports to act as master or slave clocks, run the best master clock algorithm, and exchange PTP messages with attached devices. This enables testing of clock synchronization, 14 15 The story of 2010 is Ethernet backhaul, up 36% in 2009, and another 8% in 2010, which represents amove from 80% of total MBH spending in 2009 to 88% of all MBH spending 2010. The total worldwide mobile backhaul equipment market grew 13% from CY08, to $6.2B in CY09, yet weexpect revenue to fall 1.6% from CY09 to CY10, to $6.1B, almost entirely due to the decline of TDMmicrowave spending. Revenue isset to grow beyond 2010, reaching $8.2B in 2014, a 2009- 2014 CAGR of 5.7%. This is very healthycontinuous growth, especially for a market that already has revenue in the billions of dollars. The main drivers of the mobile backhaul market are: More phone and mobile broadband subscribers, with growth being driven in India and Chinaespecially, and other developing world markets such as Africa and Mobile Backhaul Ethernet Equipment Market is Booming Latin America; mobilebroadband subscriber numbers have already passed ﬁ xed broadband and 1.3B will be addedbetween 2010 and 2014 Increased user bandwidth for mobile broadband, mostly driven by HSPA/HSPA+; accordingto the GSMA, over 99% of WCDMA operators have deployed and commercially launched HSPA,100 HSPA+ systems will be commercially launched by end 2010, with 73 HSPA+ live at thiswriting, operating at speeds of 21M, 28M, or 42MM. WiMAX rollouts continue, while over 100operators worldwide are now committed to LTE. More new cell sites added for coverage despite spectrum zoning issues in the Americas andmany cell site sharing agreements in Europe, Africa, and India, which dampens growth—andbackhaul capacity upgrades for others to support new bandwidth LTE small cell sites; many worry about the coming explosion of small cell sites needed todeliver the increased capacity of LTE; the jury is out at this point, as world innovation leader,NTT, is driving a new approach of not deploying micro/pico base stations, but rather remote radiohead, which uses more ﬁ ber-attached radios and antennae attached to existing macro eNodeBs;we have increased our backhaul connections forecast somewhat, but will wait to forecast anexplosion of small cells until we see what others in Europe and North America will do Improved range of Ethernet microwave equipment available from all vendors with better pricing per capacity and increasing capacities, now in excess of 1GMicrowave is moving quickly from TDM-only to Ethernet and dual TDM/Ethernet. (Source are from Infonetics Research) Effectively Testing Mobile Backhaul with STC Cover Story Mobile Backhaul Testing Cover Story Mobile Backhaul Testing jury is ou Node e ha ctions ther 16 17 PTP accuracy needs to be valid across real network infrastructure Spirent TestCenter ports can operate as master or slave clocks over Ethernet, IPv4, or IPv6 and emulate complex routing and MPLS topologies. Enables users to test complex, real-world mobile backhaul scenarios with fewer DUTs Optional support for external Spirent Network Impairment Emulator.Enables users to run additional negative tests (from page16) Easy to use Wizard and Topology Editor for creating EOAM trafﬁ c VPLS implementations – supports VPLS-LDP, VPLS-BGP Conformance Test Suites for key MEN technologies Repeatable and pre-deﬁ ned test cases Provides reference system for ﬁ nding bugs and improving interoperability STC is tailor-made for handling the complexities of MBH testing Realistic MBH test scenarios involve emulation of hundreds of base stations behind CEs & PEs originating stateful application trafﬁ c. Using STC’s multi-protocol scale and device-behind-device capabilities, it is trivial to emulate above scenario—on 1 STC port!Based on deﬁ ned topology, outgoing control plane trafﬁ c will have appropriate Ethernet/ IP headers and MPLS label stack—this includes TCP Trafﬁ c Setup & Management Bandwidth needs from each Base Station connection will increase 5x to 50Mbps(Apri,2010, Infonetics forecast) in two years as 3G/4G rollout accelerates.Real-time voice, video and back-ofﬁ ce applications are sensitive to poor network performance, which requires: Each switch/router must support thousands of service level instances with QoS/CoS Verifying SLA/QoS involves advanced performance measurements such as loss, latency and jitter Service providers will have to be MEF 9 and MEF 14 compliant and handle high bandwidth / multi-protocol scale trafﬁ c Spirent TestCenter provides comprehensive Carrier Ethernet Trafﬁ c Management solution. Using STC we can test Ethernet QoS testing at scale, including: Realistic EVCtrafﬁ c generation; Generate/Analyze in real time 32k/64k EVCs – measure real time packet loss, re-sequence, delay and jitter; Analyze and verify trafﬁ c shaping and policing mechanisms; Trend with Stateful TCP Good put to see the net effect on Goodput; Support for Y. 1731 performance measurements; Flexible generator/analyzer for cutting-edge MEN technologies Besides, STC provides ﬂ exible generator/analyzer for cutting- edge MEN technologies, which can generate and analyze 802.1ad and 802.1ah (PBB/PBT) trafﬁ c,VPLS-LDP, VPLS-BGP and PWE3 with easy to use wizards etc. Ethernet and MPLS-TP OAM Ethernet OAM helps SPs provision and troubleshoot Ethernet services, monitor each service in terms of availability, frame loss and delay.In addition for each service thresholds can be deﬁ ned alarming when the service deviates the SLA. Features of Spirent TestCenter Ethernet OAM Solution as below: Powerful and ﬂ exible EOAM solution – 802.1ag, Y.1731 and 802.3ah emulation Full participant in Topology Emulation, which means the user can test EOAM over other network structures like MPLS application trafﬁ c Using Spirent TestCenter, OAM & data trafﬁ c from thousands of Carrier Ethernet MEPs can be sent to DUT over EVC: Can generate E-OAM (Segment & Link) trafﬁ c over EVC Link Trace, Loopback, AIS and RDI signals generated in combination with stateful application trafﬁ c The permutations are endless Meaningful QoS metrics measured for received packets in real-time Trillions of ﬂ ows can be overlaid on up to 32K Tx and 64K Rx streams per port 41 QoS metrics can be measured on each packet including real-time loss, short-term average latency & true MEF jitter Received trafﬁ c can be analyzed intelligently in real-time (based on Queue rules) to identify and isolate problems STC Live solution is a distributed probes test solution for Ethernet based network, which offers Multi-System, Edge to Edge - End to End Performance Monitoring and Test and Diagnostics over 1G or 10G Ethernet connections. The STC Live Probe generates trafﬁ c for active analysis of service quality and passively monitors trafﬁ c to diagnose service troubles. The Probe veriﬁ es the operational status of the underlying network and the IP services that transverse it and quickly identiﬁ es and sectionalizes any network service issues that are present then provides information to support accurate dispatches. The Probe supports both 1GigE and 10GigE interfaces and can be inserted into VLANs for on- demand active testing or passive monitoring of speciﬁ c data streams. STC Live T&D handles the STC Live Probe STC Live Mobile Backhaul Solution insertion via the mirror or span port of the network element seamlessly. Probes offer following functions: Active Test support enabling service turn-up & troubleshooting Passive visibility for link, stream, & application troubleshooting 24 x 7 PM for proactive detection of network degradation Packet Capture/Decode 10/100 Ethernet control link, -48V DC powered NEBS Level 3 compliant and CE Class A Flexible port operations Cover Story Mobile Backhaul Testing Cover Story Mobile Backhaul Testing 18 19 What can STC Live do with Mobile Backhaul Network? The Spirent STC Live Probe, deployed at the aggregation points of Mobile Backhaul networks, provides a diagnostic Probe which uses the NE for access to "join the service” in real time. The Probe provides the most extensive suite of performance and throughput tests and protocol decodes in the industry as an Ethernet aggregation point solution. The Probe increases service quality visibility and plays a vital role in measuring service. When strategically placed at the aggregation point (Ethernet Switch\Router, NTE , NID , PWE) the Probe provides maximum visibility into the network. This visibility extends to multiple locations throughout the network, including other aggregation locations or end-points and provides a complete end-to-end transport network view and service validation system with unique sectionalization features. SIAD SIAD NTE NTE NTE NTE MSN #1 MSN #2 STC Live Probe Ethernet Backhaul Application 2: Mobile Backhaul Service Maintenance With the Probe located in key locations and the power of Spirent’s Ethernet test and diagnostic solution’s active and passive test functionality providers can answer those tough service questions discussed earlier. The Probe will access the Ethernet switch and join the VLAN to be tested, by verifying following active tests on Mobile Backhaul network: 802.1ag/Y.1731 Loopbacks - allow sectionalization without being intrusive to customer trafﬁ c MAC Swap Available Features - take Ethernet frames and send them back without being intrusive to customer trafﬁ c IP address/MAC Swap loopback (IP SRC/DEST the same) – not intrusive to customer trafﬁ c UDP Echo Port Loopbacks – not intrusive to customer trafﬁ c Physical Port Loopbacks – turns all trafﬁ c around Available in NTEs, Ethernet Over devices, etc. STC Live Probe placed at the Ethernet Switch is mirrored to the VLAN or port trafﬁ c of the desired service. Once accessed, the Probe can be used to perform non-intrusive passive testing for observation and analysis of subscriber generated trafﬁ c passing through that point in the network. The Probe can monitor customer service entering and leaving the Mobile Backhaul network to help sectionalize trouble. Trafﬁ c Visibility – VLAN statistics, CoS, frame sizes, protocol distribution, utilization Data Analysis – top talkers, top applications, connection matrix VoIP Analysis – signaling & voice analysis/quality Video Analysis – channel operation & video analysis/quality Capture/Decode – deep dive analysis Application 3: Mobile Backhaul Performance Monitoring Spirent TestCenter Live Performance Monitoring (STC Live PM) solution provides real-time performance visibility across the entire footprint of the production network without impacting revenue generating trafﬁ c. By utilizing Ethernet OAM capabilities deﬁ ned by ITU Y.1731 and IEEE 802.1ag, STC Live Probes are able to provide full mesh testing to any standards-compliant device or between Probes and feed that data into a powerful centralized analysis and reporting engine. The combination of the STC Live Probe and the analysis engine is STC Live PM. STC Live PM’s exceptional visibility into the customer’s Ethernet VLAN service using both current and historical information empowers providers to offer essential Ethernet Service Level Agreements (SLA s) with absolute conﬁ dence in their ability to meet them. PM is the software that manages the performance data, enables back ofﬁ ce integration, facilitates, simpliﬁ es and improves automation, and drives the STC Live Probe. PM sectionalizes trouble into or out of a provider’s network, or between providers in shared Ethernet networks utilizing Spirent’s best in class STC Live Probe. The Probe provides real time monitoring of Ethernet service performance with scaling capability to meet carrier class requirements. PM uses the Probe to emulate the customer’s service by actively generating data patterns across the network to measure, validate, and ensure QoS. The Probe quickly assesses the subscriber’s service identifying any service issues that are present and provides information to support accurate dispatches back to STC Live PM. The Probes are able to monitor the customer’s service at three important layers: Layer 2 PM (EOAM) Layer 3 PM (TWAMP) Application Layer (Ex. VoIP and Video) Performance Monitoring Reports Reports come in several ﬂ avors, providing detailed near real time and historical data on key metrics. The reports provide views of the data in an array of intervals – 5 minute intervals for 24 hours, hourly intervals for a week, and daily intervals for a week, month, and quarter for example. STC Live PM has the capability to present information showing Ethernet QoS down to each Ethernet data stream for each Class of Service (CoS). In addition, the provider can obtain the information directly from STC Live PM or PM can feed the information into a higher level Performance OSS in their network. STC LIve PM will identify critical threshold crossings and send the alerts or SNMP alarm traps to designated fault systems. The providers can designate the SLAs for which threshold service measurements will be invoked and set these thresholds per Key Performance Metric (KPM) i.e., Frame Loss and Frame Delay for both Warning (Yellow) and Critical (Red) Threshold Violations. Spirent also provides a summary view of all of the VLANs that have crossed targeted Critical and Warning service thresholds within various periods or increments of time such as 5-minutes, hourly, or daily. Analyzing data gathered every day and throughout each day is especially valuable to determine service trends during low, average and peak activity time periods. In addition, PM has a constant source of near “real time” data that supports analysis of performance for reactive troubleshooting. These reports provide valuable information that will assist personnel in determining the “root cause” of real network service issues. STC Live PM allows the provider to slice this information numerous ways such as by service category, by switch or all switches, or by service metric. The possible retrievable data combinations are almost limitless. Spirent TestCenter Live delivers very scalable, highly accurate Ethernet based Mobile BackHaul live network testing and diagnostic solutions that address all aspects of the Ethernet life cycle including service activation, service assurance and performance management for mobile backhaul. Application 1: LTE Mobile Backhaul Service BULK Turn-up Perform Active Testing to emulate customer trafﬁ c for accurate “ﬁ rst time turn-ups” to speed the time to new service revenue. Connectivity Check – Ethernet and IP Stream Generation RFC 2544 - Committed Information Rate (bandwidth) can be reached ITU-T Y.156sam Support (Network Quality Test) Generate Ethernet Stream and validate QoS – frame loss, delay, jitter, out of order frames, etc are within speciﬁ cations Generate Multiple Streams simulating - HSIA, Video, VoIP trafﬁ c and determine integrated QoS behavior Loopback Control - for trouble isolation process Cover Story Mobile Backhaul Testing Cover Story Mobile Backhaul Testing 20 21 Independent testing lab EANTC (European Advanced Networking Test Center) recently put Cisco’s next- generation mobile network to the test using Spirent solutions. The results of these tests have been published by Light Reading in two reports. The ﬁ rst report, “Testing Cisco's Next-Gen Mobile Network” covers Cisco’s IP-RAN and Mobile Backhaul. The second report, "Testing Cisco's Mobile Core, Data Center & Business Services," focuses on the mobile core, mobile applications, the mobile operator's data center. As these tests clearly demonstrate, next-generation solutions require careful validation. This is especially true when designing, developing or deploying solutions for the transition from 2G & 3G to LTE/4G technologies and beyond. Companies cannot afford to deploy a solution with costly “hidden surprises” impacting their ability to deliver a quality product or service. These in-depth reports delve into the details of the testing done and explain the rationale for each test. The extensive scope and the magnitude of testing required to validate an end-to-end solution is evident in the network diagram. Testing Next-generation Mobile Networks ——Validating IP Ran, Mobile Backhaul and mobile core Testing Cisco's Next-Gen Mobile IP Network EANTC tested multiple areas using Spirent products and expertise including: Phase & Frequency Synchronization Hardware Upgrade (CRS-1 to CRS-3) 100GbE Throughput QoS in the IP-RAN Link & Node Resiliency in the IP RAN Backhaul NAT64 Session Setup Rate & Capacity Route Processor Redundancy All-IP RAN Scale NAT64 Module Redundancy Testing Cisco's Mobile Core, Data Center & Business Services EANTC tested multiple areas using Spirent products and expertise including: SGSN Attachment Rate GGSN Session Setup Rate & Capacity GGSN Performance With DPI GGSN Redundancy Evolved Packet Core Performance Firewall & Network Address Translation Functionality & Scalability Session Border Control Functionality & Scalability Data Center Interconnect Spirent Promise to Validate Mobile Core, Mobile Backhaul and IP RAN Spirent was the only test vendor able offer a complete solution to validate Cisco's mobile solution. Key differentiation Spirent delivered: Test Coverage - Any-G mobile technologies including 2G, 3G and LTE packet core, mobile backhaul over 1G, 10G and 100G Ethernet and stateful features like DPI and Policy Enforcement were tested Realism - Accurately modeled mobile subscriber behaviour, mobile subscriber trafﬁ c and mobile backhaul scenarios while measuring complete protocol and trafﬁ c performance Scalability - Highly dimensional, blended mix of performance at scale including ports, multiple layers of network protocols and application trafﬁ c Ease of use - Spirent's built-in, NoCode test authoring tools made it possible to recreate and measure realistic network behavior with coding custom protocol sequences and device integration Expertise - Spirent engineers brought the knowledge necessary to accelerate the concept and innovations of Cisco's next generation mobile infrastructure from ideas to being ready for service provider LTE deployment ( Please visit www.lightreading.com/mobiletest for the full article of the test report.) Spirent TestCenter Mobile Backhaul and Spirent TestCenter Live received TMCnet’s 4GWE Wireless Backhaul Distinction Award. Winners of the Wireless Backhaul Distinction Award were chosen based upon companies who successfully demonstrated a commitment to innovation toward the 4G Wireless Evolution. Spirent solution are enabling the industry migrate from legacy TDM solution to an IP Ethernet network in the backbone to meet the growing data bandwidth pressure expected for new/advanced mobile application usage. Spirent Receives 2010 4GWE Wireless Backhaul Distinction Award Cover Story Mobile Backhaul Testing Cover Story Mobile Backhaul Testing 22 23 This article describes the deﬁ nitive approach to classifying, testing and analyzing mobile backhaul while under extreme scale in a realistic test environment. Test cases majorly cover IEEE 1588V2. Measure accuracy of IEEE 1588v2 boundary clock timing This test case determines whether the DUT boundary clock replicates accurate timing to slave clocks over a long duration under a load of trafﬁ c. The precision of the boundary clock and timer variation is measured. Incorrect replication of timing as speciﬁ ed in IEEE 1588v2 by the boundary clock can lead to incorrect network behavior, dropped handset communications, random errors, and degraded bandwidth. Test Procedure 1. Reserve one tester port for the grand master clock, connect the port to the DUT, and bring up the link. 2. Conﬁ gure the connected port on the DUT to accept clocking from the emulated grand master port and boundary the timing trafﬁ c to desired slave clock ports. 3. Reserve a desired number of slave clock domains, one per test port. Connect the ports to the DUT. Conﬁ gure the connected DUT ports as slave ports. 4. Establish the link on all ports. Testing Methodology for Mobile Backhaul 5. Conﬁ gure the emulated grand master port to start timing services. Conﬁ gure the DiffServcodepoint on timing trafﬁ c to AF31. Conﬁ gure QoS on the DUT, and prioritize AF31 over all trafﬁ c. Verify that the DUT forwards timing trafﬁ c. 6. On each emulated slave port, conﬁ gure the desired number of emulated slave clocks. 7. Verify that each slave clock receives clocking form the DUT. 8. Create multiple full-mesh trafﬁ c across all test ports in the system. 9. Wait the desired pre-measure time. 10. Calculate the clock error for each slave clock from the grand master clock. 11. Calculate the clock difference between each slave clock. 12. Measure PRBS errors. Determine IEEE 1588v2 slave clock capacity This test determines the ability of the DUT to support multiple IEEE 1588v2 slave clocks while maintaining timing precision. This test case increases the number of slave clocks while measuring the precision of timing compared to the grand master clock. Understanding how many slave clocks a DUT can support while keeping precision is an important metric to refer to when provisioning mechanism network. Desired Result The clock drift from the grand master clock to each slave clock should be no more than 0.001%. The difference between slave clocks should be 0. There should be no PRBS errors. Test Procedure 1. Reserve one tester port for the grand master clock, connect the port to the DUT, bring up the link. 2. Conﬁ gure the connected port on the DUT to accept clocking from the emulated grand master port and boundary the timing trafﬁ c to desired slave clock ports. 3. Reserve a second port and connect it to the DUT slave port. Conﬁ gure the slave clocking service on the DUT for this port. 4. Establish the link on all ports. 5. Conﬁ gure the emulated grand master port to start timing services. Conﬁ gure the DiffServcodepoint on timing trafﬁ c to AF31. Conﬁ gure QoS on the DUT, and prioritize AF31 over all trafﬁ c. Verify that the DUT forwards timing trafﬁ c. 6. Create multiple full mesh trafﬁ c across all test ports in the system. 7. Loop until any clock precision drifts more than 0.001% of the grand master clock or when there are PRBS errors. a. Add an emulated slave clock. b. Wait the pre-measure time. c. Verify that each slave clock receives clocking from the DUT. d. Measure the drift on all emulated clocks. 8. Record the current number of emulated slave clocks. Desired Result The number of emulated slave clocks should match the designed peak number of slave clocks supported on the DUT. There should be no PRBS errors. Scalability of IEEE 1588v2 clocks over VPLS This test determines the maximum size of the internal VPLS table while the DUT maintains timing accuracy. In this test, the DUT VPLS and unicast BGP routing tables are incrementally ﬁ lled until timing precision decreases to below 99.999% accuracy as IEEE 1588v2 frames are transmitted across VPLS. Backhauling IEEE 1588v2 trafﬁ c across VPLS tables is a key deployment strategy. Incorrect timing can result in dropped calls and loss of bandwidth. Variable Number of Slave Clocks per Slave port Emulated Hosts Per Port Pre Measure Time Relevance Number of emulated slave clocks per port, default 100 Number of hosts per port to emulate, default 4096 Time to wait after starting clocking services before measuring drift, default 4 hours Variable Number of Emulated Slave Clock ports Number of Slave Clocks per Slave port Emulated Hosts Per Port Pre Measure Time Relevance Number of physical test ports used as slave timing ports, default 5 ports Number of emulated slave clocks per port, default 100 Number of hosts per port to emulate, default 4096 Time to wait after starting clocking services before measuring drift, default 4 hours Cover Story Mobile Backhaul Testing Cover Story Mobile Backhaul Testing clocks Test Procedure 1. Reserve tester port A and conﬁ gure it as a grand master clock. Connect the port to the DUT boundary clock and bring up the link. 2. Set the ID of the grand master clock to match the DUT. 3. Attempt to bring up the IEEE 1588v2 session and record the results. 4. On each slave port, conﬁ gure the desired number of emulated slave clocks. Set the ID of the slave clocks to match the DUT. 5. Attempt to bring up the IEEE 1588v2 session and record the results. Test Procedure 1. Reserve four tester ports. Connect the ports to the DUT and bring up the links. 2. Conﬁ gure BGP on the DUT and set the AS to 100. Conﬁ gure VPLS on the DUT and set the DUT role as a P-Router. 3. Conﬁ gure the DUT port connected to test port A to accept clocking from the grand master clock emulated on tester port A. Peer BGP with the DUT. 4. Setup test port B as a slave clock. Peer BGP with the DUT. 5. On both test port A and B, emulate a P Router->PE Router- >CE Router and clocking endpoint. 6. Begin clocking services. 7. Loop until known upper limits for AS count, route count and VPLS tunnel counter are reached. 8. Record the number of BGP peers, advertised AS path count, routes, and VPLS tunnels. Desired Result The BGP peers, advertised AS path count, routes, and VPLS tunnels counts should be the maximum in each category that the DUT is designed to carry. Desired Result Neither the grand master clock nor any slave clock should successfully peer with the DUT Determine whether the 1588v2 grand master clock is resilient to DDoS attacks This test determines whether the boundary clock can provide accurate timing to the network while subject to distributed denial of service (DDoS) attacks. In this test, the boundary clock is attacked while the timing protocol is bridged across the DUT. Protocol resilience is a critical part of deployment in a production network the timing trafﬁ c to desired slave clock ports. 5. Reserve a desired number of slave clock domains, one per test port. Connect the ports to the DUT. Conﬁ gure the connected DUT ports as slave ports. 6. Establish links on all ports. 7. Conﬁ gure the emulated grand master port to start timing services. Conﬁ gure the DiffServcodepoint on timing trafﬁ c to AF31. Conﬁ gure QoS on the DUT, and prioritize AF31 over all trafﬁ c. Verify that the DUT forwards timing trafﬁ c. 8. On each emulated slave port, conﬁ gure the desired number of emulated slave clocks. 9. Verify that each slave clock receives clocking from the DUT. 10. Create multiple full mesh trafﬁ c across all test ports in the system. 11. Wait the desired pre-measure time. 12. Calculate the clock error for each slave clock from the grand master clock. 13. Calculate the clock difference between each slave clock. 14. Measure any PRBS errors. Desired Result The clock drift from the grand master clock to each slave clock should be no more than 0.001%. The difference between slave clocks should be 0. There should be no PRBS errors. Test Procedure 1. Reserve one tester port for the grand master clock, connect the port to the DUT, and bring up the link. 2. Reserve one port as the Distributed Denial of Service (DDoS) attacking port. Conﬁ gure line rate trafﬁ c with a PDU of Ethernet>IP>UDP. Increment the UDP source and destination ports. Connect this port to the boundary clock. 3. Start attack trafﬁ c toward the boundary clock. 4. Conﬁ gure the connected port on the DUT to accept clocking from the emulated grand master port and boundary Variable Maximum BGP AS Maximum Routes per AS Maximum VPLS Tunnels per AS pair Variable Set of 1588v2 Slave Clocks Emulated Slave Clocks per Port DUT Clock ID Relevance Maximum number of AS per DUT, default 10,000 Maximum number of routes per emulated AS, default 100000 Maximum number of tunnels per AS pair, default 4000 Relevance Ports emulating slave clocks, default 4 Number of slave clocks emulated per port, default 100 ID of the DUT boundary clocks Variable Number of Emulated Slave Clock ports Number of Slave Clocks per Slave port Emulated Hosts Per Port Pre Measure Time Relevance Number of physical test ports used as slave timing ports, default 5 ports Number of emulated slave clocks per port, default 100 Number of hosts per port to emulate, default 4096 Time to wait after starting clocking services before measuring drift, default 4 hours Duplicate clock identity detection in IEEE 1588v2 This test determines whether the DUT can detect and avoid duplicate clock identities in the network. In this test, the DUT clock identify string is replicated on emulated clocks on the test interface to determine whether the DUT rejects the connection. Users are allowed to determine the clock identity and thus allows for duplication of identities, which can lead to incorrect timing behavior in the network. 24 25 Cover Story Mobile Backhaul Testing Cover Story Mobile Backhaul Testing New Product New Product 26 27 Spirent TestCenter 3.60 Spirent TestCenter 3.60 Enhancement Overview Mobile Backhaul Enhancements to Topology Emulation Spirent TestCenter 3.60 are now available. Spirent TestCenter 3.60 adds key elements of Mobile Backhaul, Data Center, Intelligent Results, and also adds several productivity enhancements. Data Center and Virtualization Enhancements Spirent TestCenter’s mature Data Center and virtualization platform become enhanced with the addition of the VIC protocol. Spirent TestCenter Virtual enhances storage protocol testing in hypervisors and across the network with SCSI workload generation for FibreChannel and FCoE. Advanced Analysis & Reporting Spirent TestCenter 3.60 continues to enhance Intelligent Results capabilities. With Dynamic Views, the user can create sophisticated and powerful queries on both conﬁ gure and measured parameters using a simple and intuitive wizard. Spirent TestCenter 3.60 adds Wireshark™ support for analysis of captured data. Post test data mining also becomes more ﬂ exible and intuitive with brand new Spirent TestCenter 3.60 Results Reporter. Productivity & Large Test Scale Enhancement Spirent TestCenter 3.60 simultaneously makes testing easier and more scalable for the user. A new ‘Getting Started’ screen has been added to enable new users to become productive in a few minutes. Detailed List of Spirent TestCenter 3.60 Enhancements Mobile Backhaul IEEE 1588v2 Support Stateful emulation of Grandmaster, Master, Best, and Slave Clocks under scale Make live changes to running clocks; use Topology Emulation to trend clock performance with other protocols and application services Synchronous Ethernet Support on all Ethernet Modules Spirent TestCenter enhances Mobile Backhaul testing by adding 1588v2, SyncE, seamless MPLS, and TWAMP support to our unique Topology Emulation architecture. As the network becomes more hieratical and virtualized, the user can now test clocking performance under high load throughout the network. With seamless MPLS, the user can easily test key technologies such as LDP LSP PING, entropy labels, and TWAMP. Because these protocols are conﬁ gured with Topology Emulation, the user can test complex, multi- protocol topologies with simplicity. Emulate the master Clock and changes over time with Command Sequencer Test G8262 and G8264 MPLS-OAM LDP-LSP PING/ Trace Route support with easy to use wizards TWAMP Client & Server (RFC 5357) Easily conﬁ gure TWAMP client and server with other routing and MPLS protocols Test live changes while the Device Under Test (DUT) is under load Core Routing VPLS Psudowire Redundancy Based on ‘draft-ietf-pwe3-redundancy-02’ implementation Measure device under test (DUT) switch over redundancy MPLS QoS Mapping Full display of hardware based dynamic MPLS mapping Measureable with Dynamic Views and Results Reporter Dynamic Results Views Point to Multi Point RSVP (P2MP) Added “Make before Break” functionality Make live changes while sessions are active Multi-Topology ISIS Test IPv4 and IPv6 routers and routes in a single area (L1/L2) Flow Label for PW Flow aware transport of PW over an MPLS PSN using Entropy labels Based on ‘draft-ietf-pwe3-fat-pw-03’ draft Access PPPv6 over L2TPv2 IPv4 IPCPv6 over L2TP Conﬁ gure both LAC and LNS for L2TPv2 PPPoEv6 Server Serve both PPPoE and PPPoEv6 on the same port DHCPv6 Client / Server Emulation Topology Emulation “Home Gateway” link class coordinated DHCPv6 client with network events automatically Tunnel DHCPv6 DHCPv6-PD Client Support both ‘PPPoEv6oL2TPv2oIPv4 - Client’ and ‘PPPoEv6oL2TPv2oIPv4- Server’ Topology Emulation “Home Gateway” link class coordinated DHCPv6-PD client with network events automatically Tunnel DHCPv6-PD Data Center / Virtualization VN- Technology (802.1Qbh – Edge port bridging) Full stateful emulation of VIC Protocol to test switches SCSI workload generation for FC / FCoE on Spirent TestCenter Virtual Initiator emulation on virtual machines FCF to FCF conformance - extended topologies FC Services Conformance Test Suite Test FC conformance VF Port Emulation Emulate a FC switch port Layer 2-7 Trafﬁ c 802.1x Supplicant Support Login to the network dynamically Media Encoded RTP for SIP Embed a “WAV” ﬁ le in the RTP stream for SIP Trafﬁ c Measure QoE (MOS-LQ) Analysis - Results Reporter General Enhancements Default to full-screen on launch and remember the screen size for subsequent launches Dynamic template loading Easily locate the database folder Option to apply a template change to ALL iterations Ability to print Results Reporter content with an option to re-align page size Export Enhancements Option to split up an CSV ﬁ le for tables with more than 65k entries Ability to select speciﬁ c tables/graphs for HTML export Ability to export the same table across all iterations Template Enhancements Tables will have alternate colors between rows Ability to create 3D graphs in the template page Table Enhancements Ability to have column left/center/right justiﬁ cation Ability to copy/paste table rows without headers Ability to deﬁ ne number of rows per page Graph Enhancements Option to change graph line colors and style Analysis - Real-Time Dynamic Result Usability Enhancement Friendly naming of ﬁ elds in regular expressions Result Count display added Selectable drill down or table view of results Result Browser Table and Chart Export to clipboard or ﬁ le within Spirent TestCenter GUI Analysis – Capture Wireshark™ Support Stream live capture to Wireshark and decode on the ﬂ y Decode Spirent Test Signature ﬁ eld No cost option Enhanced Hardware Capture Filters User regular expression (AND/OR /NOT) to capture based on ranges Deﬁ ne your own live counter through based on multi- ﬁ eld range conditions Works with Spirent Wireshark™ Live Next Generation Measurement Manual Schedule Mode Deﬁ ne TX side bursting Allows users to precisely control packet transmission Latency Calibration Allows users to zero out optics, cables, and other ﬁ xed offsets to measure ultra-low latency devices Productivity, User Support, & Large Scale Testing Microsoft Windows 7 32- and 64-bit OS Support Multi-threaded Architecture GUI and Test Kernel reside in separate processes Spirent TestCenter Launch Wizard Allows new user to get started with complex topologies created in minutes Web Installer Users can download Spirent TestCenter GUI client from Spirent TestCenter chassis Faster port and chassis recycle times Hypermetrics CM / CV recycle as fast as 7 seconds New 2U Chassis with Whisper Mode New 2U Chassis that dynamically sets the FAN RPM based on environment conditions. Conformance Test Suites (CTS) Ethernet Link Management Interface (E-LMI) MEF16 CTS Veriﬁ es MEF16 implementation with over 130 test cases G8032 (Ethernet Ring Protection Switching) CTS (BETA Feature) Verify G8032 protocol conformance using G8032 CTS Verify R-APS logic, state machine, PDU validation,single- ring topology in Revertive and Non-Revertive modes Over 700 test cases Spirent Avalanche 3.60 is released. Spirent Avalanche 3.60focuses on enhanced usability, industry leading performance & scale with true realism, enhanced security testing, storage testing and enhanced OS support. Avalanche 3.60 is the world class leader in cloud based services and user emulation testing under extreme load. Spirent Avalanche 3.60 Enhancement Overview Spirent Avalanche 3100 Phase 2 (3.56) With a 50% performance increase over the Spirent Avalanche 3100, Spirent Avalanche again sets the industry record for high performance testing without sacriﬁ cing realism. Protocol Expansion with SAPEE QuickFlow Spirent Avalanche SAPEE QuickFlow usability enhancement brings simple point and click addition of client and server protocols to your test case. With the ability to add protocol packs, Spirent Avalanche allows for dynamic addition of new protocols between releases. Storage Spirent Avalanche 3.60 enhances the user’s ability to use storage realism under extreme scale. With enhancements to CIFS block size and server timestamp, there user can test with high accuracy and scale. NFS Kerberos support allows for more realistic NFS testing. Protocol Enhancement With SMTP authentication support, DNSTCP and DNSSEC, SSL Refactor enhancements, and HTTP Adaptive Streaming, users can test more broadly and with higher conﬁ guration options Security Tunneling – IPSEC (Alpha) Enhanced user interface allows for repaid building of thousands of S2S tunnels with ease. Spirent Avalanche 3.60 Detailed List of Avalanche 3.60 Enhancements SAPEE QuickFlow Simple point and click addition of protocols to Action Lists Maintains SimUser realism while dramatically simplifying protocol additions Protocol Packs Allows plug in of additional protocols to Avalanche SMTP Authentication Support The AUTH command is an ESMTP command (SMTP service extension) that is used to authenticate the client to the server The AUTH command sends the clients username and password to the e-mail server AUTH can be combined with some other keywords as PLAIN, LOGIN, CRAM-MD5 and DIGEST-MD5 (e.g. AUTH LOGIN) to choose an authentication mechanism The authentication mechanism chooses how to login and which level of security that should be used SSL improvements including NTLM v1/v2 Support Conﬁ gurable SSL record size for both client and server Ability to support one digital certiﬁ cate per user using forms database New Product New Product 28 29 Support for NTLMv1 and NTLMv2 authentication for HTTP protocol under Client/Proﬁ le and Action List CIFS-Settable Block Size With this CIFS enhancement, user has the ability to set the CIFS read and write block size request in the range of 1024 to 65535 bytes CIFS-Settable Server Timestamp On the server side, user can select how Avalanche assigns a new timestamp each time a ﬁ le is accessed There are two options for selecting the timestamp: Generate on the ﬂ y – assign a new timestamp each time a ﬁ le is accessed Constant – assign a timestamp when a ﬁ le is ﬁ rst accessed, and then retain that timestamp on subsequent requests NFS-Kerberos Support NFS support with Kerberos authentication support DNSTCP Support With the support of DNS over TCP in Avalanche, users are Spirent’s Virtual Drive Test (VDT) – Conversion Tool is an invaluable aid in bringing “virtual drive testing” to your wireless device lab. Virtual drive testing reduces the cost and time-to-market liabilities inherent in live drive testing. Drive testing has always been a “necessary evil”: it is expensive and imposes a drag on time-to-market goals. It requires you to expose precious intellectual property to the outside world. And when results are gathered, there’s no way to tell whether deltas in the results are due to design tweaks or changes in the ﬁ eld environment. On top of all that, complex multiple-antenna (diversity and MIMO) scenarios are becoming the norm, exacerbating all these issues. For years RF engineers have dreamed of a way to take real- world RF data from drive testing, store it, and use it to re- create the same, repeatable RF scenario on the lab bench. Spirent’s Virtual Drive Test (VDT) for the SR5500 Wireless Channel Emulator brought that capability to the market for those who had there sources to convert captured drive-test data. Now it’s faster and easier than ever to take advantage of Spirent’s Virtual Drive Test capabilities. The powerful VDT- Conversion Tool takes RF data collected during drive testing and automatically converts it within minutes for storage and playback in the lab. Beneﬁ ts Reduced development costs – Minimize the need for expensive drive testing. Reduced time-to-market – Fewer physical drive tests eases the strain of tight deployment schedules. Start “drive testing” whenever you see the need… within minutes. able to send messages over UDP or TCP and DNS servers bind to UDP/TCP port 53 The beneﬁ t of sending messages to DNS over TCP is that it takes advantage of the performance of UDP but also has a backup failover solution for longer queries DNSSEC Support HTTP Adaptive Streaming Added support for HTTP Adaptive Streaming for Microsoft SmoothHD clients With SmoothHD client support, Avalanche provides a complete solution of content (video) delivery network (CDN), IPTV, and VoD testing Windows 7 OS Support Support for Windows 7 32-bit and 64-bit OS IPSec (Alpha Enhancement) Combining the ability to setup large number of IPSec site to site and wizard/policy generator provide users the ease of use and the most secure and scalable way to test VPNs. This feature improves the IPSec usability within the Avalanche software and also provides security integrated into network infrastructure, industry-leading VPN technology, high performance for IPSec advanced security services by providing strong encryption, authentication, and integrity to network services testing solution. Content File System Enhancement Performance improvements for distributing content over multiple ports 50% or greater improvement in system response time Spirent Avalanche 3100 Phase 2 (3.56) 4x10G port and 12x1G port Scale testing of multi 10G layer 4-7 trafﬁ c State engines support faster setup and sustained higher connectivity 50% Increases in performance Greatly improves encryption and security testing performance Enhanced product quality – An absolutely repeatable drive test on a bench top. Quickly home in on RF issues that might otherwise go unnoticed until after deployment Key Features Supports all bands for GSM, WCDMA & CDMA/EV-DO Converts 4+ hours’ worth of drive test data within a fewminutes All data can be graphically displayed to allow easy analysis. Spirent Virtual Drive Test (VDT) – Conversion Tool New Product New Product 30 31 Spirent customers have the opportunity for optimized testing as a result of our varied and broad services offerings. The graphic below illustrates how Global Services adds value, maximizing and enhancing the customer’s evolving experience with their Spirent investment. From the point of purchase, customers who need to assure that their Spirent gear maintains its reliability and productivity in critical test cycles adopt Support Services. Customers who need their test gear up and running in the shortest time possible choose Implementation Services. Customers needing to expand their team’s expertise employ Education Services. As deadlines and costs tighten, solutions from Professional Services minimize testing risks or maximize productivity with lab and test automation. All of Spirent’s services are designed to help our customers bring their products and services to market faster and to keep critical costs down. This article provides insight into two vital areas of Spirent Wireless services: Support Services and Professional Services. 33 Services Services Spotlight on Wireless Services By Patrick Barry Global Services: Adding Value to the Customer Experience 32 Spirent Wireless Support Services With dedicated Spirent Wireless Support Service centers in Eatontown, NJ; Paignton, England; Beijing, PR China, and with the new Bangalore, India service center (as well as distributors in Japan and Korea), Spirent Wireless Support is global in scale and coverage. Spirent’s main customer base in this growing line of business consists of wireless handset manufacturers and wireless service providers. As the global need for wireless handsets and service expands at an accelerating pace with no end in sight, Spirent is on the front lines. All major handset manufacturers have Spirent systems and almost all handsets in US have been tested with Spirent equipment. With the timing of product or service releases being critical factors in marketplace success, having a support plan in place for Spirent’s range of test products – including the AirAccess, SR5500, as well as the C2K and 8100 platforms – is critical for customers needing to maintain their competitive edge. “Keeping applications at peak performance so they perform accurate and repeatable tests that deliver results all the time is what our support is all about,“ says Rick Oliver, Manager of Support Programs from the Eatontown, New Jersey ofﬁ ce. The System ASA (Annual Support Agreement) contract is the primary offering for Wireless Support Services, with coverage including: unlimited phone support, software updates, hardware maintenance and yearly calibrations. The team delivering this support includes dedicated phone support professionals, in-house repair and calibration specialists, as well as Field Support Engineers (FSEs) who are located strategically around the globe, centered near our strategic customers in each region. The Spirent Support hardware support team distinguishes itself in important ways. Brad Ball, Customer Service Manager out of Eatontown, takes pride in observing that the hardware repair team consistently turns around all their repairs and calibrations within ﬁ ve days. Cliff Bailey, Application Support Manager in the Eatontown ofﬁ ce - to whom the FSEs and backline teams report - observes that centralization of support services processes and produces, for all of Spirent’s Support Services, is nearing completion. Bailey now has on his team the ﬁ rst frontline phone representative for Wireless who is also qualiﬁ ed to cover all the other Spirent support product lines (Networks and Applications, Positioning and Service Assurance). In the AsiaPAC region, the Wireless Support team not only directly supports Spirent’s customers, but also works with distributors and trains their engineers to assure delivery of the highest quality Spirent support for customers. New Wireless Professional Services from Spirent Currently, the Spirent Wireless Professional Services team is developing a customized set of test suites called the Development Test Library for two of the top ﬁ ve Wireless service providers in North America. The Spirent Professional Services team worked with the Spirent Wireless product group and the service providers to create rigorous sets of standardized test plans. All mobile device manufacturers who propose to have their devices deployed in the service providers’ network must pass this battery of acceptance tests. The test plan comprises approximately 75 test cases, focused on throughput performance and include pass/fail criteria for easy analysis of post-test results. The output of the Professional Services includes the test cases and documentation supporting the test methodology. In addition, warranty support is included. Ultimately, Spirent delivers an automated acceptance and validation process developed and endorsed by the service providers. This solution is standardized against the wireless providers’ requirements and is reusable for all devices that ﬁ t the requirements criteria. The Professional Services group was chosen for these engagements based on the Spirent Wireless product group’s close relationship with the service providers, and with a track record of developing testing solutions for service providers that are an exact ﬁ t with their needs. Major beneﬁ ts Spirent delivers to these leading wireless service providers include an understanding of the 8100 platform, test case development, acceleration of device testing and faster time to market for mobile devices. The testing is a tremendous time saver for the service providers since the cases are over 12 hours in duration and would be signiﬁ cantly longer if performed manually. Spirent Wireless Professional Services has also been working with another top-ﬁ ve service provider in North America on a UTS veriﬁ cation pretest plan of their development library to facilitate smooth integration of all handsets being introduced into the service provider’s network. This streamlines the integration for the service provider, since all mobile devices have been pretested before they are ofﬁ cially submitted to the service provider for deployment. This requirement means the handsets have a high chance of ﬁ nal passage, which speeds both the product and service to market. In addition, this UTS offering is seen to meet a growing demand in the Asia-Paciﬁ c region. These customized partnerships delivering customized development test libraries with a growing number of major service providers reﬂ ect a dynamic synergy that is in sync with the accelerated technology advancements that are part of today’s wireless world. Wireless Services Solutions from Spirent It’s clear as the progressive challenges of global wireless technology continue to evolve, the Global Services Wireless team grows with them, delivering solutions that are closely synchronized with the targeted needs of Spirent customers. 3 4 1 2 Why Test Your New Cloud? 10 Things You Need to Know Q: We are planning to migrate to a cloud computing solution. My highly-qualiﬁ ed IT team has ample experience with designing and implementing new networks. If we have their stamp of approval, is it really necessary to test our newly-proposed cloud computing solution any further? A: The cloud infrastructure — which is a blend of physical and virtual devices, existing and emerging technologies (some in early stages of adoption) — makes or breaks the cloud services quality of experience. What looks good on paper may not be in sync with the ever-evolving complex realities of cloud computing. Findings from Nemertes Research state that launching an IT initiative without testing can completely wipe out the anticipated operational and cost beneﬁ ts. Testing for performance, availability, security, and scale (the cornerstone components of the PASS test methodology) assure the success of a cloud initiative by identifying vulnerabilities that may have been missed in the design and implementation phases. Proper and timely testing before launch assures the delivery on the promises of cloud computing. With the promise of robust and broad range of next-generation networking offerings, the awareness of Cloud Computing’s advantages is widely-acknowledged. The rich potential beneﬁ ts of Software as a Service (SaaS), Infrastructure as a Service (IaaS) and Platform as a Service (PaaS), available in combination or individually in scalable data centers or other venues, are well appreciated. The pathway to delivering those advantages in public, private or hybrid clouds, however, require the right level of due diligence to achieve. Q: What exactly is the PASS test methodology? A: In response to recognizing the myriad vulnerabilities stemming from the complex and possibly confounding convergence of processes which constitute the implementation of any cloud computing solution, Spirent Communications developed the industry’s ﬁ rst holistic test methodology to validate the performance, availability, security and scalability (PASS) of cloud computing. These four categories are the cornerstone of success or failure of any IT initiative and have been developed based on top concerns in the industry by IT managers. The Spirent PASS methodology addresses both physical and virtual appliances and is speciﬁ cally designed to test services and infrastructure between any point in the cloud environment. Q: Recognizing that effective cloud testing needs to address virtual, physical and combined environments, what are some of the speciﬁ c tests required for effective evaluation of the performance of the cloud? A: Performance tests from industry standards that address these requirements include: IETF RFC 2544 Network, RFC 2889 LAN Switch, RFC 3918 Multicast, RFC 5180 IPv6 Network and Draft Data Center Bridging Benchmarking. Q: What are some of the critical components of testing availability in cloud computing? A: Cloud Infrastructure as a Service (IaaS) Live Migration is of crucial importance: the movement of IaaS VM instances without the requirement of powering down. Automated resource scheduling, which automatically moves IaaS VM instances at deﬁ ned load thresholds, is another important component. High availability is also critical in cloud computing, which includes the ability to instantly restart IaaS instances on HW failures. These components require virtualized test environments to produce the depth of metrics required to accurately assess the availability of your solution. 5 8 9 10 6 7 Q&A on testing Q: Are adjacent cloud services a serious security threat? A: Adjacent cloud services can be security threats and these concerns need to be addressed for any cloud solution. Cloud security solutions require traditional security components to work in concert with virtual security elements. You need to test them individually and as a whole, which requires test cases, hardware and virtualized components to assure effective and comprehensive test coverage. Areas where cloud security should be assessed include: user authentication IPSec, SSL VPN, 802.1x, Network Access Control (NAC), RADIUS and network attack/negative trafﬁ c testing. Testing of ﬁ rewalls, IDS/IPS and complex deep packet inspection should also be part of comprehensive security testing. Q: How critical is it to ensure SLAs in cloud? How will cloud providers be able to provide assurance to the end-user? A: Assuring SLAs in the cloud is critical, especially in a public cloud, as tenants need to be assured of the performance they receive. Continuous testing and monitoring is important, in addition to the pre-deployment strategies. For this providers will have test equipment integrated in the production network so that it can be used on an as-needed basis. Q: I’m considering a cloud solution in phases. Does PASS testing take into account my needs for future growth? A: Scalability and performance testing identiﬁ es the breaking points of the system and quantiﬁ es the delta between current utilization levels and maximum, indicating the headroom in the current implementation and deployment. With that information about your real-world headroom, you know to what degree your cloud is future- proofed, and where your subsequent development needs to occur in future phases of your cloud solution growth strategy. Q: Once Quality of Experience (QoE) has been assured to Cloud customers for SaaS, IaaS and/or the PaaS components, what are the main advantages of the cloud solution from a provider or end user’s perspective? A: Cloud provides inﬁ nite scalability with opportunity to grow in sync with business requirements and deliver a pay for what you use model. Some other key advantages are reduced cost, increased storage, highly automated environment, ﬂ exibility and mobility. Also it allows IT to shift focus, no longer having to worry about constant server updates and other computing issues. Q: I recognize the need to test my cloud solution with the PASS methodology. I’ll just have my in-house team take over with these guidelines in mind, right? A: Best practices clearly indicate testing, before an IT initiative goes live, is an effective practice to mitigate costly risks. However, relying on in-house engineering resources, which may lack the required up-to-date expertise and experience, can produce unreliable test results. The choice of a qualiﬁ ed test partner is critical. Employing a partner with a team of test experts conversant in the challenges of cloud computing, with expertise in the ﬁ eld of data center testing, is essential. Also, since the cost of the required test equipment can be prohibitive, selecting a partner that also manufactures a broad array of networking test equipment, and provides rental options, offers added advantages. Q: What speciﬁ c qualiﬁ cations do I look for when choosing a test partner? A: A neutral third party; Testing as a core competency; Virtual and physical test systems for true end-to-end testing of cloud services; An established global name in the test and measurement industry; Experience and expertise; Quality test engineers; Provides test rental equipment (preferably manufactured by same company); An established delivery process; Extensive automation expertise; Articulation of testing beneﬁ ts / ROI vs Risk; Holistic understanding of networks; Successful engagements with references. 35 Q&A on testing 34 of su d on d t A: Clou as a Ser n is s at deﬁ por ailu s. These y o 37 Recently Spirent announced that a major network equipment manufacturer has installed the Spirent SR5500 Wireless Channel Emulator for testing beamforming capabilities of Time Domain-Long Term Evolution (TD- LTE) base stations at its laboratories in Shanghai, China. Performance testing of network equipment is critical for the successful deployment of TD-LTE, a technology which is expected to ﬁ nd widespread adoption in China, India and several other countries in Asia. Unlike the Frequency-Domain implementation of LTE currently being deployed in North America and Europe, TD-LTE uses a single shared frequency band for both uplink and downlink. Beamforming is an adaptive antenna technique used to maximize the performance of wireless systems. It typically makes use of eight antennas on the base As a part of the ongoing effort to advance test automation techniques, Spirent recently successfully demonstrated interoperability of its solutions with the Network Test Automation Forum (NTAF) framework. This interoperability test underscores Spirent’s commitment to work with industry leaders to deﬁ ne and implement standardized automation interfaces. NTAF, which is tasked with establishing a standard for the way in which network test solutions communicate with each other, is expected to deliver the ﬁ rst set of ﬁ nalized speciﬁ cations by the end of Q1 2011. The plug fest served to validate the draft NTAF speciﬁ cation, identifying and resolving differences in how each participant implemented the NTAF protocol, and demonstrating acceptance from multiple NTAF members. An integrated and collaborative test automation approach is critical for today’s engineers to effectively validate performance of the complex, multi-technology, multi- protocol and multi-vendor networks. Leading network equipment manufacturer selects Spirent SR5500 Wireless Channel Emulator for testing advanced TD-LTE base stations Spirent helps NTAF achieve major milestone towards open automation speciﬁ cation New Spirent MIMO Test Solution Conﬁ guration Helps the Industry Deliver on the Promise of TD-LTE Spirent Demonstrates Interoperability with NTAF Test Automation Framework Spirent Update Spirent Solution Identiﬁ es Smartphone Performance Issues From Voice, Data and Location Service Interaction In an effort to validate the Metro Ethernet Forum’s External Network-to-Network Interface (ENNI) inter-working across network operators worldwide, the Carrier Ethernet Global Interconnect demonstration conducted by EANTC and co-organized by ancotel used Spirent solutions to test interconnectivity, interoperability and service provisioning. The demonstration validates Carrier Ethernet services across different continents from a variety of service providers including Beeline, BICS, Equinix, Expereo International, Kazakh Telecom, TeraGate, Tinet, Ucomline and XO Communications. Earlier this year the MEF ratiﬁ ed interconnect standards that enable service providers to understand how to expand their service coverage by seamlessly interconnecting with other providers to deliver an end-to-end service while maintaining global service level agreements (SLAs). The results of the Global Interconnect test event show that, while signiﬁ cant progress has been made in ENNI implementations, interoperability and provisioning challenges continue to exist. Building on the previous phases of the Global Carrier Ethernet Interconnect demonstrations that were kicked off in January, this event focuses on validating Ethernet Service OAM and on verifying service performance across exchange points in New York and Frankfurt, Germany and networks spanning service providers in seven countries. This effort builds on the validation of MEF’s ENNI speciﬁ cation and multi-provider Class of Service (CoS) implementation options. Spirent’s market leading performance monitoring and test and diagnostics solution, Spirent TestCenter Live™, was used to generate trafﬁ c, analyze how the trafﬁ c was mapped at the interconnect switch, and verify compliance with Ethernet OAM standards. The solution provided the critical visibility December 13, 2010 — Spirent announced the availability of the industry’s ﬁ rst automated R&D solution for testing smartphone performance in the presence of simultaneous voice, data and location-based services (LBS). Built on Spirent’s 8100 Development Library solution, this new service interaction test capability is the latest innovative Spirent tool designed speciﬁ cally to test 3G UMTS device performance. Although today’s smartphones are running an ever- increasing number of applications, these applications must share the device’s limited resources, and this can lead to performance challenges. For example, the use of A-GPS location on smartphones for navigation or other location- based applications can disrupt the reception of business Results of Carrier Ethernet Interconnect Demonstration Shows Signiﬁ cant Progress in ENNI Implementations and Highlights Challenges of Service Provisioning New Test Capability Helps Ensure a Positive End User Experience of Smartphones in the Presence of Multiple Simultaneous Services Spirent Veriﬁ es Global Carrier Ethernet Network Interconnectivity and Service Performance Across Multiple Service Spirent Update needed to validate EOAM trafﬁ c as it traversed the network. In addition, Spirent TestCenter Live™ was used to validate provisioning as the Ethernet Virtual Circuits (EVCs) were provisioned, allowing EANTC to rapidly address any network conﬁ guration issues. EANTC also used Spirent TestCenter ™ for additional bandwidth proﬁ le testing. Spirent enables service providers to quickly validate the performance of Carrier Ethernet services delivery by offering test suites compliant with MEF speciﬁ cations. In addition to supporting MEF standards on Spirent TestCenter, Spirent offers Spirent TestCenter Live, the only integrated test and diagnostic and performance monitoring solution that is fully compliant with standards such as ITU Y.1731 and IEEE 802.1ag. e-mail, reduce data download speeds and can also cause important calls to drop, even threatening the reliability of emergency calls. Spirent’s new service interaction testing capability was used to carry out tests on a number of commercial smartphones, which ran multiple simultaneous services in the lab under real-world conditions. The results, documented in a recent Spirent white paper, highlight some of the signiﬁ cant performance threats to key elements of smartphone functionality. The ﬁ ndings, along with the innovative test methodology used to obtain them, are also the subject of a recent Spirent webinar, available now for on-demand viewing. station and two on the mobile device (8x2). Spirent’s SR5500 is used to simplify and accelerate the testing associated with complex beamforming implementations. The SR5500 is a modular system which creates realistic synchronized radio paths by affecting a transmitted radio signal in the same way as a live environment (i.e. noise, radio-wave reﬂ ections, etc.). For this application, Spirent has developed a fully-integrated 8x2 bi-directional (downlink and uplink) SR5500 conﬁ guration controlled by a single software user interface. For more information on Spirent’s channel emulator and TD- LTE testing visit http://www.spirent.com/Solutions-Directory/ SR5500.aspx 36 echniq or dize ticipant im Spirent Helps Nokia Siemens Networks Verify its Evolved Packet Core Solution is Best-in-Class Recently Nokia Siemens Networks has selected Spirent Communications to validate the performance of its Evolved Packet Core (EPC) gateway Flexi NG, a solution that will play a key role in next-generation wireless network deployments. With the market leading Spirent Landslide, Nokia Siemens Networks is able to verify the performance of its EPC solutions, while reducing the cost and time associated with bringing them to the market. “Spirent’s test equipment has helped us verify our EPC solutions to deliver on our promise of proﬁ table, high-performance and reliable networks to our customers,” said Petri Pöyhönen, head of Converged Internet Connectivity business line at Nokia Siemens Networks. “We evaluate IP-based end-to-end networking solutions in our ResIP Center, and trust Spirent’s testing solutions for efﬁ cient veriﬁ cation of multiple mobile technologies simultaneously. Spirent’s integrated test methodologies and intelligent results allow our engineers to validate the performance of Nokia Siemens Networks’ EPC solution in high-scale mobility scenarios.” Nokia Siemens Networks is using Spirent Landslide to assess the scale and data performance of its EPC gateway against complex mobility and real-world trafﬁ c scenarios to ensure successful live network operations. Spirent Landslide is a comprehensive end-to-end platform that emulates millions of 2G, 3G and LTE mobile data subscribers, simultaneously accessing the wireless network, using various access models. It emulates all key wireless network elements and combines highly scalable control plane and application data to provide a real-world emulation of millions of mobile nodes in various stages of activation, deactivation, and hand-off between cells. Nokia Siemens Networks validates performance, scalability and reliability of its Flexi NG evolved packet core gateway with Spirent Landslide Spirent eAirAccess recognized for its role in advancing the development of LTE technologies Spirent Participates in Ethernet Alliance Multi-Vendor Interoperability Demo, Speaks on High Performance Cloud Computing at SC10 Spirent Receives 2010 4GWELTE Visionary Award Spirent Helps Advance Ethernet Technologies in High Performance Computing Environments December 14, 2010 – Spirent announced that Spirent’s eAirAccess wireless network emulator was presented the 2010 LTE Visionary Award. The award was presented by Technology Marketing Corporation (TMC) in conjunction with Crossﬁ re Media. eAirAccess is a true multi-radio access technology (RAT) network emulator used for testing mobile devices. Its real-time state machine emulates an entire post-deployment environment including LTE, WCDMA, GSM, HSPA, CDMA, EV-DO and Evolved High-Rate Packet Data (eHRPD) services, with multiple cells available per technology. “Spirent’s eAirAccess is uniquely designed to address the complexity around handovers between radio access technologies, especially since LTE deployment will need to coexist with legacy networks such as CDMA/EV-DO or UMTS/HSPA for many years,” said HeshamElHamahmy, vice-president at Spirent’s wireless division. “The eAirAccess solution has already found commercial success and it’s an honor to now be recognized by the industry.” Spirent Update 38 New Orleans – November 16, 2010 — Spirent Communications, a leader in data center and cloud computing testing, has joined the Ethernet Alliance in a live multi-vendor interoperability event that demonstrates the performance of next generation data center technologies during SC10 in New Orleans, November 15 - 18. Spirent’s involvement in this event underscores the company’s commitment to advancing Ethernet technologies within the high power computing (HPC) industry, through comprehensive testing. As data centers get larger, faster and more complex, new technologies such as virtualization, FCoE and 40/100G Ethernet aim to help organizations move terabits of diverse trafﬁ c types onto a single, converged network. Validation and performance assessment are vital ﬁ rst steps in implementing these new technologies. In a data center context, that means testing each technology, not just by itself but also in concert with many other data center components, old and new.

Access Edge

Network Equipment

Cloud/Data Center

Core

Mobility

High Speed Ethernet

Automation

Monitoring

Security

Performance and Acceptance

Location Technologies

Research and Development

Applications

Test Environments

Technologies

Products

Services

Testing World Magazine

Company

Media Center

Partnerships & Alliances

Contacts

Connect