Since the release of Virtual SAN one of the most popular topics of discussion about Virtual SAN revolves around solution sizing and performance capabilities. For the most part, the majority of guidance around Virtual SAN designs has been focused on capacity sizing and performance characteristics of virtual machine workloads.
However, there are other aspects of sizing and design criteria for Virtual SAN, specifically those related to system-wide performance and availability during data management operations. The data management operations of Virtual SAN are focused around data resynchronization and rebalancing amongst all the all copies of data. The functions and impact of these operations should be part of all Virtual SAN design and sizing exercises for optimal results.
The design of data management operations is intrinsic to the value proposition of Virtual SAN. It is important to know the events that activate them and also understand the impact they introduce during normal operations. Inadequate size and design can have an impact on the overall performance expectation and availability capabilities of the solution.
This white paper provides detailed information about the Virtual SAN data management operations, their functions and the type of events that triggers them as well as recommendations for achieving performance and recoverability results based on cluster design and sizing. Virtual SAN data management operations can be downloaded from the VMware Virtual SAN product page as well as the directly link provided below:
For future updates on Virtual SAN (VSAN), Virtual Volumes (VVols), and other Software-defined Storage technologies as well as vSphere + OpenStack be sure to follow me on Twitter: @PunchingClouds
Since the release of Virtual SAN one of the most popular topics of discussion about Virtual SAN revolves around solution sizing and performance capabilities. For the most part, the majority of guidance around Virtual SAN designs has been focused on capacity sizing and performance characteristics of virtual machine workloads.
