At Interop New York this week Arista Networks announced two new switches to the 7000 product family, targeted at being leaf switches for high-density and storage racks, providing ultra-low latency and support for VXLAN. The Arista 7050QX-32 and 7050SX-128 leaf switches offer 32 40GbE or 96 10GbE ports and 8 40GbE wire speed ports. The high efficiency 1 or 2u switches feature ultra low-latency, VXLAN and NVGRE capabilities. When used in conjunction with the Arista 7000 series of fixed and modular switches, it allows networks to scale to over 100,000 10 GbE servers in a low latency leaf and spine network.
This latest introduction expands the portfolio of switches available with consistent Arista EOS in the form of a single binary image across all products, greatly simplifying network administration.The flexibility of L2 and L3 multipath design options, combined with support for open standards, provides maximum flexibility, scalability, and network wide virtualization.
“The new Arista 7050QX-32 gives us an unprecedented switching capacity in a 1RU low-power package. Arista’s stable and capable EOS allows us to expand the complexity of our data processing network while keeping pace with our ever-expanding data center requirements,” said John Belshaw, IT Architecture Advisor, CGG.
Arista also announced new network tracers to complement existing health and network virtualization offerings for a full suite of Arista EOS Network Telemetry applications. Path Tracer provides end-to-end visibility into Layer 3 ECMP (equal cost multi-path routing) networks, proactively detecting, logging and reacting to gathered information.
"Cloudera Enterprise, our Platform for Big Data, gives our customers access to all their data whether it's structured or not. It's the foundation next generation of data management infrastructure, and provides insights that haven't been possible until now." said Alan Saldich, vice president of marketing for Cloudera. "Arista's Map Reduce Tracer provides similar insights to network administrators in understanding network performance and big data workloads".