SONiC is an open source network operating system based on Linux that runs on switches from multiple vendors and application-specific integrated circuits (ASICs), contributed to the Open Compute Project led by Microsoft. SONiC offers a full-suite of network functionality such as Border Gateway Protocol (BGP) and Remote Directory Memory Access (RDMA) that has been production-hardened in the data centers of some of the largest cloud service providers. It offers users the flexibility to create network solutions while leveraging the collective strength of a large ecosystem and community. SONiC is based on a disaggregated architecture (separate switch silicon, white box switch and network operating system software) within an open source community.
“SONiC’s disaggregated architecture enables the fast growing open source data center and the community is working to ensure that all of the components work together,” said Dean Lee, senior director, Cloud Solution Unit at Keysight’s Network Applications & Security group (formerly Ixia Solutions Group). “Keysight’s participation in the SONiC Test Working Group will help companies deploying SONiC ensure optimal performance and resiliency of their data centers.”
“The SONiC community welcomes Keysight’s support,” said Gerald Degrace, Head PM for Host NIC, Switching, SONiC and WAN, Microsoft Azure Networking, Microsoft. “With Keysight’s long history and experience in testing switches and routers, it will expand the test coverage of the SONiC test bed. Keysight’s contribution will be valuable especially in the areas of conformance, interoperability and traffic congestion control.”
Keysight will offer the SONiC community a data center fabric test environment to stress SONiC performance and scale. This test environment can scale in multiple dimensions including the number of hosts/servers, top of rack (TOR) switches, leaf switches and spine switches to simulate a hyperscale data center environment. The test environment will enhance the existing test bed, including the testing for:
- Forwarding performance and resiliency – to measure resiliency of the data center in case of data path failure
- Control plane conformance and performance – to identify potential interoperability issues and characterize performance bottlenecks and limits
- Congestion avoidance and buffer management – to ensure that traffic flows are prioritized properly and buffer/queue is utilized efficiently