Datacenters are fraught with hidden operational hazards, and because virtual infrastructure is extremely complex and dynamic, many vulnerabilities go undetected and can be onerous to find. On the other hand, known hazards are often ignored because administrators underestimate or do not understand their scope, severity and risk.
CloudPhysics data scientists have compiled a Datacenter Insights Report (registration required) that exposes where operational hazards lurk in an organization’s IT infrastructure, jeopardizing the health and safety of the datacenter. The report is available to any organization running VMware vSphere 4.1 and above.
Using its global data set, which has more than 50 trillion points of machine metadata from datacenters of all shapes and sizes around the world, CloudPhysics has found:
- 43% of organizations risk application downtime every week due to disk full condition in the guest
- 41% of HA clusters do not have admission control enabled
- 26% of disk space is dead
- 16% of virtual machines are unused (powered off or suspended)
- 25% of Windows virtual machines will have reached end of support life next year
- 22% of VMware vSphere 5.5 servers remain unprotected from SSL Heartbleed.
On average, each “bully” virtual machine in the datacenter causes 2.2 datastore contentions daily and negatively impacts 5 other virtual machines, degrading application performance
The company also released a “Halloween Cookbook” that compiles tips and tricks for using its analytics to rid the datacenter of hazards that may be haunting it.