MAN testing goes next-generation
Physical layer SDH/SONET NGN, Ethernet testing in one platform
New services like VoIP and IPTV require telecom operators to move away from legacy networks primarily designed to carry voice and constant bit rate data traffic and implement next-generation transport networks that incorporate new types of traffic. Companies providing communications services to the enterprise market are going to find themselves using existing Ethernet over SDH/SONET (EoS) infrastructure. Doing so drastically changes the requirements for testing metropolitan area networks (MANs), however. Traditional physical layer tests no longer provide the in-depth testing required for valid and reliable results.
These tests, when relied on as core measurement techniques, can often leave networks at risk. As telecom operators continue to upgrade legacy networks with new packet-based capabilities, an entirely new generation of complex test methodologies and equipment is emerging that incorporates traditional tests and verifies the performance of SDH/SONET Next-Generation Networks (NGN).
Traditionally, vendors and service providers have partitioned knowledge between transport technology and Ethernet technology. The transport environment is dominated by static, synchronous connections, while data networks carry bursty, less predictable traffic. As such, transport networks are monitored typically by verifying the status of alarms and synchronous errors, while data networks are monitored by counting frames, throughput, and the speed at which fast frames are delivered. Service providers deploying or maintaining MANs depended on three critical tests to ensure network performance and data delivery: SDH/SONET physical layer verification, network element testing and out-of-service bit error rate testing (BERT).
Figure 1: VCAT testing set-up.
Physical layer verification enables service providers to verify the performance of the link sending information in an out-of-service state. Using a traditional test set, technicians generate an SDH or SONET signal with framing structure and test pattern, generally a pseudo random binary sequence (PBRS) that represents what live traffic could look like. Once the signal is generated, the technicians can verify physical layer measurements and performance including frequency, optical power, SDH/SONET framing and error performance analysis conforming to ITU or Telcordia/ANSI recommendations.
An out-of-service bit error rate test is used to determine whether or not frames are received at the destination. Using predetermined stress patterns comprising a sequence of logical ones and zeros, the test set generates signals over time to provide a detailed view of the actual performance. From this data, a bit error ratio (BER) is calculated by taking the ratio of erroneous data bits to the number of bits transmitted. Initial tests to bring up a network typically take about 15 minutes. Longer tests spanning two hours to 24 hours are then conducted to secure more reliable data and ensure that the circuit is ready to transport live traffic.
Network element tests determine whether or not each individual element on a network, such as add and drop multiplexers (ADMs) or digital cross-connects (DXCs) are functioning properly and enabling data to reach its intended destination. A test set is used to generate signals that should cause alarms or errors. From a remote location, the technician monitors the network to ensure the proper alarms or errors are transmitted downstream by the network element. At the same time, the technician checks for remote indicators, which should be transmitted upstream. This is a critical step in determining the health of a network. It enables technicians to verify that network management systems respond by activating the automatic switching protection (APS) when defects and anomalies are identified. When functioning, the APS can prevent information loss due to network failures.
Next-generation MAN tests
Today’s SDH/SONET networks are constantly evolving to carry existing traffic, support emerging data protocols, and interconnect different regions of the world. Communications equipment must accommodate as many common protocols as possible, in order to address market segments worldwide, and interoperate with other vendor equipment. For example, a typical routing switch platform must support a long list of transport protocols including Fast Ethernet, Gigabit Ethernet, PoS, MPLS, ATM, PDH (DS1, DS3, etc.), and SDH/SONET. A look at telecom markets and standards reveals a myriad of SDH/SONET-based protocols that equipment vendors need to support in different regions of the world to carry voice and data traffic. New technologies such as Gigabit Ethernet and PoS must co-exist with decade-old technologies such as PDH and ATM.
Figure 2: GFP testing
Telecom operators have invested billions of dollars in their existing networks. With NGN they are faced with the choice of rebuilding these networks from the ground up with the newest equipment and technology, or with incrementally upgrading existing infrastructure by adding new equipment and packet-based capabilities. As expected, most are choosing to upgrade. This path, however, and the need to deliver next-generation services using next-generation protocols, requires an entirely new class of testing, to ensure transport networks can carry packet-based traffic efficiently.
Three critical tests are required to ensure the performance of NGN:
- Virtual concatenation tests including path error performance and differential delay,
- Generic frame procedure (GFP) mapping and demapping,
- Link capacity adjustment verification including bandwidth management.
VCAT, an extension of SDH/SONET (ANSI T1.105/ITU-T G.707), maintains the required SDH/SONET influence within the context of service reliability, optimizing the network bandwidth based on customer requirements. On the other hand, the extension of VCAT with LCAS (ITU-T G.7041) introduces a dynamic element to SDH/SONET that is more akin to a data protocol environment. To complete the picture, GFP (ITU-T G.7042) either provides mapping of Ethernet or Fibre Channel frames into octet-based transport technologies like SDH/SONET. GFP_F provides mapping for frame-based payloads like Ethernet/Fast Ethernet and GFP-T maps block-based payloads (i.e. Gigabit Ethernet or Fibre Channel) into SDH/SONET.
Virtual concatenation provides both bandwidth efficiency and flexibility using the service providers’ current SDH/SONET equipment for any traffic rate. It has unprecedented efficiency and flexibility since all traffic grades can be broken into the smallest available granularity, and transported as independent traffic, and then re-assembled at the terminating points. Up to 93 percent bandwidth efficiency is achieved when optimizing the virtual concatenation group (described below) based on the payload. For instance, transporting 10M Ethernet payload efficiently implies the use of 5 VC12a (SDH) or 7 VT1.5s (SONET), as opposed to using a traditional VC3 or STS-1 SPE pipe where the efficiency barely reaches 20 percent; VCAT alleviates the need to expand network capacity while minimizing the changes to the existing infrastructure (see Figure 1).
Figure 3: LCAS testing.
Testing SDH/SONET includes the injection and monitoring of alarms at the line/regenerator section, section/multiplex section, and path levels. Traffic from a single source may be separated into unique “containers,” separated into a virtual concatenation group (VCG). These containers are identified and managed using a multi-frame indicator that increments sequentially as frames are transmitted and a sequence ID number that is unique to each member. To maximize bandwidth, the containers may be routed using different physical paths. Intrusive and non-intrusive tests that monitor alarms and errors on each VCG member can verify that each container reaches its destination error-free.
By its nature, the virtual concatenation process can cause containers in a VCG to arrive out of order at the path terminating equipment (PTE). VCAT equipment can buffer the information to account for delays and to enable correct reconstruction of the original data signal. With today’s advanced test sets, technicians can create delays on each member of the VCG to simulate different physical path lengths, characterize high and low order container differential delay performance, and ensure reassembly is performed quickly and accurately according to the sequence ID number, a critical specification of virtual concatenation designs. Test sets monitor and identify three disruptions in VCAT that can cause errors at the termination point:
- loss of multi-frame (LOM), based on the multi-frame indicator,
- loss of sequence (SQM), based on the sequence number,
- loss of alignment (LOA), when the maximum differential delay is exceeded.
Differential delay is the relative arrival time measurement between members of a VCG.
Technicians can test differential delay by injecting and measuring the delay up to the maximum supported delay on the equipment or up to the absolute maximum delay defined by ITU or Telcordia by sending different types of delays – static where the delay value changes every five to ten seconds, or random. These variables are intended to simulate real-life situations.
Another aspect of VCAT is the size of each member and the maximum number of members in a VCG. Since the main premise of EoS is to carry Ethernet, enough bandwidth is aggregated to correspond to 10 Mbps, 100 Mbps, or 1 Gbps. The SDH/SONET containers, which members use in VCAT, are either high-order or low-order paths. Supported high-order paths in SONET are either STS-3c-X-v or STS-1-X-v, while in SDH they are either VC-4-X-v or VC-3-X-v. These are normally used to transport 100M/1000M Ethernet frames or 1/2G Fibre Channel traffic. In low-order, they are generally VT1.5-X-v in SONET and VC-12-X-v in SDH and are typically filled with low rate Ethernet traffic (10M).
Generic framing procedure (GFP) is the intermediate step between variable rate types of traffic and a constant octet-based transport protocol such as SDH/SONET or OTN (See Figure 2). GFP wraps the Ethernet frames with a GFP header and FCS field before filling the SDH/SONET containers. In GFP mode, data is carried asynchronously, where idle frames are transmitted when no traffic data is available. In this context, the concepts of traffic bandwidth, bandwidth profiles, and traffic streams apply. Traffic monitoring becomes a matter of ensuring that frames are not lost or received out of order. Using advanced test sets, technicians can inject errors into the header to verify response of a network element at the termination point.
Used in tandem with virtual concatenation, link capacity adjustment schemes (LCAS) (see Figure 3) can provide SDH/SONET networks with unprecedented bandwidth management flexibility. LCAS can dynamically change virtual concatenated path sizes, as well as automatically recover from path failures. This technology is important when providing bandwidth-on-demand, thus increasing bandwidth efficiency and optimizing service provider revenue potential. As such, link capacity adjustment verification has become mandatory for bandwidth management.
New test sets verify the different states of the signaling protocol and respond to LCAS commands when verifying NE LCAS functionality. By increasing or decreasing traffic from either the access or transport side, test and analysis can determine whether or not network elements register a change in bandwidth automatically. Test functions to provide error injection and monitoring associated with LCAS messages are also required.
Ethernet over SONET testing Once the SDH/SONET infrastructure has been tested and verified, network performance must be verified while carrying Ethernet traffic. To provide the most accurate QoS measurements, the test procedure should follow the network architecture and services provided. Technicians must generate and measure test traffic for every service provided, which may include voice and video services along with traditional data and network management traffic. Each type of traffic carries with it different Class of Service (CoS) settings, different bandwidth usages, and different packet loss, latency, and jitter requirements. The test set not only must simultaneously generate each traffic type, but also be able to measure and monitor each one independently.
Typical QoS metrics include the achievable throughput of the circuit, the number or percentage of lost frames, the frame delay and frame delay variation. These metrics can then be compared to a CoS definition or service level agreement (SLA) to give a pass/fail grade on the circuit. Verifying throughput is usually a simple matter. Traffic is generated at or above the expected throughput rate using frames tagged with a sequence number. Technicians can easily see whether or not frames are lost by checking the tags of the received frames. Lost frames may be acceptable, depending on the type of service; so a quantitative measurement is required. Frame delay or latency is typically measured as a roundtrip delay measurement very similar to a propagation delay test for traditional transport networks. The variation in frame delay, or frame jitter, can have a significant impact on real-time applications such as voice and video. Failure to meet QoS requirements is usually due to the performance or configuration of the network elements, which can vary based on frame size and traffic rate, and so multiple QoS tests may be required to see the whole picture.
In summary, EoS helps service providers deliver end-to-end Ethernet and storage area networks services by providing more flexibility and efficiency to existing SDH/SONET infrastructure. The combination of VCAT, for better bandwidth granularity, LCAS, for dynamic bandwidth, and GFP, as a generic framing mechanism, extend the transport network capabilities. This introduces a new set of requirements to test and validate the service. Today, test equipment companies are delivering new, comprehensive systems that include physical layer SDH/SONET NGN, and Ethernet testing in one platform, enabling technicians to test every single part of the network, quickly and reliably. For service providers, these systems virtually guarantee the quality of network they can deliver and fully tested services that meet service level agreements quickly and less expensively.