This syndicated post originally appeared at Zeus Kerravala's blog.
It’s the end of August, which means “Tis the season.” What season is that? It’s now the end of the summer, which means back to school for our kids. It’s also NFL preseason, so all of us going through football withdrawal are close to getting some real football soon – time for Tebow mania, the football equivalent of SDNs. But, for those of us in tech, the end of August means VMworld time. For me, it means my email inbox is full of messages from PR vendors wanting me to “stop by the booth” and check out the latest and greatest.
This week, Arista and ExtraHop got out in front of the impending noise and announced a strategic partnership between to deliver an integrated solution called the ExtraHop-Arista Persistent Monitoring Architecture. Despite the totally unimaginative name, the product should be compelling to highly virtualized organizations or those considering a move to a software defined data center – the obvious sweet spot for VMworld.
The premise of the joint solution is that the software defined data center is significantly more complex than legacy ones. Delivering applications used to be fairly straightforward, although highly inefficient as each application had its own dedicated infrastructure. Today, application delivery is anything but that. Everything is virtualized, shared, converged and mobile.
This means traditional management tools that look at physical topology simply won’t work in these highly virtualized environments. However, physical infrastructure isn’t going away, and management tools that monitor the virtual environment have no real knowledge of the physical world. What’s required is visibility and correlation of data across the entire application delivery stack, and that’s what the ExtraHop/Arista offering is designed to provide.
The Persistent Monitoring Architecture integrates the Data ANalyZer (DANZ) features in Arista’s EOS operating system with ExtraHop’s Context and Correlation Engine (CCE) for persistent visibility of mobile workloads. The combined solution automatically discovers, classifies and maps the dependencies of infrastructure and related performance. As the data center changes based on virtual machine migrations, or other changes the solutions dynamically adapts, using the ExtraHop APIs to ensure IT has a persistent view of performance. No more blind spots caused by VM migrations.
One of the keys to the solution is that the DANZ data is continually being sent to the ExtraHop device, so there is no agent, polling or re-discovery required. These management methods may have been fine when the data center supported “best effort” traffic, but today’s software-defined, highly virtualized environments are anything but best-effort. The solution uses the continuous auto-discovery of the virtual and physical infrastructure, and the associated MAC addresses to ensure persistent SPAN capabilities on any port or VLAN.
One of the primary value propositions of a SDDC is that it simplifies data center operations. However, without the proper management tools, virtualization and “software defined” will actually make things more complicated and could increase operational expense. The integrated solution from ExtraHop and Arista gives IT visibility up and down the IT stack and cross-team visibility make the SDDC manageable. This is a great example of the flexibility and programmability of Arista’s operating system as well as strong evidence for the analytic capabilities of ExtraHop. As the vision of the SDDC moves closer to fulfillment, we should expect to see more partnerships like this.
Latest posts by Zeus Kerravala (see all)
- Thinking About the Cloud 2020 - March 20, 2017
- SD-WAN: Right for the Cloud Era? - March 14, 2017
- Microsoft and NVIDIA partner to bring GPUs to the public cloud - March 8, 2017