The cloud, 5G and the Internet of Things (IoT) have combined to form one of the biggest growth forces in the age of networking. Each of these trends brings new applications and services. Combined, they’re transforming the way networks are designed, built, secured and operated. Each has led to new architectures that are necessary to support the enormous scale of users, devices, applications, infrastructure and data.
To ensure service level and capacity as network traffic continues to increase, enterprises need high-speed networks. Telecom networks and data centers that serve hundreds of thousands of users are deploying 100 Gbps network links to keep up with the growing demand. Consequently, there’s a need for testing and troubleshooting of the networks at this link speed.
Replay for More-Realistic Tests
As the number of customers and services that rely on stable data networks increases, quality and maturity are critical when deploying new solutions. Testing new network solutions or new versions of networking products and applications is a challenge because synthetic test cases fail to disclose all potential issues. Replaying real production-network traffic provides more-realistic test scenarios. Moreover, the test coverage can be extended by replaying traffic from many different networks.
When network problems manifest themselves, having a system of efficient troubleshooting reduces the disruptions and service downtime. Network-traffic replay is a powerful tool for reproducing and analyzing issues in a lab environment without affecting the production network. Identifying network issues requires precise replay in both throughput and timing. For example, recreating traffic microbursts in a lab requires that the solution support full throughput and precise timing of replayed packets.
Your Choices in Traffic Replay
You can buy complete solutions for capture and replay of 100 Gbps network traffic today. In some cases these products match the actual need, but in others they are far too expensive, have too many features or lack customizability to special use cases.
You can also create a custom replay solution using a COTS server, a standard Linux OS and a 100 Gbps smart NIC. The COTS server has the horsepower and I/O performance to run 100 Gbps replay. Linux enables use of open-source software as well as easy application development and feature customization. The smart NIC ensures full throughput transmission for any packet size, with precise transmit timing; some smart NICs having one-nanosecond time resolution.
If you choose to build your own 100 Gbps replay solution, understand that you need certain qualities in the elements to ensure high performance.
Make maximum memory bandwidth available in the server so you can achieve maximum replay performance. The server should be populated with the maximum number of DDR4 memory blocks to allow full use of all the memory channels. Another recommendation is to use dual-rank DDR4 or memory blocks running at least 2,133 MHz or higher. To replay large files, the packet-store output performance limits the maximum average replay speed. Consider fast solid-state disks for high-speed replay.
Not just any parts will do. For example, the smart NIC must support a 100 Gbps network interface with line-rate capture and transmit for any packet size. It must also support high-precision time stamping of received traffic, high-precision transmit timing, an efficient buffer system and an optimized DMA scheme over the PCIe interface.
Some smart-NIC vendors include tools for capture and replay in their software suite; open-source capture and replay applications are available as well. The application should employ server memory for buffering traffic from the packet store before transmission. This feature compensates for the difference between the 100 Gbps peak network packet rate and the maximum read rate from the packet store.
By 2020, the world will contain four billion connected people, 25 million applications, 25 billion embedded systems, 50 billion devices and sensors, and 50 trillion Gbps of data. In other words, high-speed networks are becoming a necessity. Along with them comes the need to test and troubleshoot 100 Gbps links to ensure quality service. Replaying actual network traffic enables you to test outside the production network so you avoid affecting service. Complete capture and replay solutions are available on the market today, or you can create your own using quality components. Building a solution yourself ensures that you and your customers get just what you need without overspending.
About the Author
Sven Olav Lund is a senior product manager at Napatech and has over 30 years of experience in the IT and telecom industry. Before joining Napatech in 2006, Sven Olav was a software architect for home media-gateway products at Triple Play Technologies. From 2002 to 2004 he worked as a software architect for mobile-phone platforms at Microcell/Flextronics ODM and later at Danish Wireless Design/Infineon AG. As a software engineer, Sven Olav started his career architecting and developing software for various gateway and router products at Intel and Case Technologies. He has an MSc degree in electrical engineering from the Danish Technical University.