This is the first in our TECHRead series of posts to highlight interesting and valuable technical resources developed by the joint Citrix and HP engineering teams. Understanding the performance of virtual machines is important as this feeds directly into informing decisions about how many virtual machines can be run on a given host as well as inform scalability decision making for server and storage resources. Capturing and analyzing virtual machine performance in your specific environment with your specific workloads is the only way to know what is really going on in your systems.

XenServer records various performance statistics persistently for long term access and analysis of historical trends. Where storage is available to a VM, the statistics are written to disk when a VM is shut down. Statistics are stored in RRDs (Round Robin Databases), which are maintained for individual VMs (including the control domain) and the server. The fact that the data is captured and stored is great – what you do with the data is the important part.

HP has created this paper to describe a process for analyzing XenServer persistent performance logs for your host system. The methods used and suggestions provide in this paper were developed by analyzing the performance of Citrix XenDesktop running on XenServer on HP BladeSystem servers and HP StorageWorks P4000 SAN storage. The suggestions for generating graphs from the RRA data sets is an example of the very practical, useful information provided in this paper. There is nothing like seeing data charted and displayed over time to aid in analyzing factors impacting performance.

So check out this paper, and if you find it helpful, or have suggestions send HP your feedback at this link.