This is the forth and last article (here is the previous one) of a series regarding memory management in VMware ESX/ESXi strategy advised by Eco4Cloud and anabled by Eco4Cloud’s workload consolidation, smart ballooning and troubleshooter solutions. The topic of this article is VMs rightsizing.
Assigning the right amount of resources to the virtual machines helps to ensure maximum performance of your workloads and the efficient use of your underlying hardware. Today adding resources to a virtual machine when required is just two clicks away, so we’d better get rid of our (bad) habit of over provisioning.
Often times the hardest part of right sizing VMs is convincing the application owners that they don’t really need as many resources as they think they do. In the physical world, adding more CPUs and more memory could unlikely degrade performance. In the virtual world, on the other hand, this is certainly not the case.
In fact, many often wonder if there is overhead when creating a large virtual machine with excessive assignments of vCPUs and RAM that may or may not be used, letting the ESX/ESXi sort it out. Actually, adding more virtual memory could actually result in making matters worse if the host is contended, increasing memory ballooning and swapping.
Eco4Cloud Troubleshooter is a software which performs continuous monitoring of well-known virtualization options and sets immediate warning/alerts when wrong configurations are detected.
Among other misconfigurations, Troubleshooter reports oversized and undersized VMs, and allows resources reconfiguration. Troubleshooter achieves that by analyzing past workloads and suggesting the optimal resources assignment.
In this overview, we saw how to take advantage of Transparent Page Sharing, Memory Limits, Memory Ballooning and Memory Rightsizing in the best way with Eco4Cloud solutions.
Field results show that, even in a small server farm, tens of TeraBytes of RAM can be saved without compromising performances, but actually increasing them.
This allows to reduce and/or delay CapEx to buy new RAM or even entire physical servers altogether, as memory is typically the bottleneck resource of a virtualized environment.
Call to Action – Request a PoC immediately to take advantage of Eco4Cloud memory optimization features.