CloudPhysics vSphere Design Based on Big Data Analytics

March 11, 2014 1 By Eric Shanks

SAMSUNG CSCOne of the companies I was most interested in seeing at the GestaltIT Virtualization Field Day 3, was CloudPhysics.  I was already a little familiar with the company because I’d written a post on my experience in the lab.  While my original post was obviously good, you can’t really get a more passionate and knowledgeable explanation of the solution than from the Co-founder and CTO Irfan Ahmad.  The presentations can be found online.

All travel expenses and incidentals were paid for by Gestalt IT to attend Virtual Field Day 3. This was the only compensation given.

 

The Big Data Solution

From my experience in the financial industry, I can tell you that financial decisions are not usually made by a guess.  A lot of time and money goes into building solutions that will allow analysts to make very informed predictions about what will happen over time.  Every minute data point is tracked, put into data cubes, dissected and forecast so the best resolution can be made.

Think about how we make IT design decisions now.  Sure, we’ll do some analysis and load testing to see how to size an environment, but is that really the same thing?  Wouldn’t it be great to have a way to tell us what would happen if we made certain decisions about how we designed our infrastructure?  Is it cheaper to own our own hardware, or migrate it to Amazon Web Services?  Should we buy some flash for caching and if so, how much?  How should I configure my High Availability Cluster?

CloudPhysics is empowering administrators to use large data sets to determine a course of action.

CPcards1

Data Security

In order to make these kinds of decisions, CloudPhysics is using metrics from all of their customers.  The larger the data set, the more accurate the forecast will be.

This is a great idea, but some organizations may be concerned about giving up their configuration data for fear that they’ll lose confidential or intellectual property.  Rest assured that CloudPhysics isn’t stealing your confidential information.

CloudPhysics is collecting configuration and performance information for:

  • VMs
  • Hosts
  • Clusters
  • Resource Pools
  • Datastores

In addition this information is transferred via SSL encrypted sessions and under no circumstances are passwords, IP Addresses or personal data transferred.

Obviously there is an ID that has to be transferred so that you can login to the portal and view your infrastructure, but even this data is housed separately to further ensure privacy.  To learn more please check out their website.

 

Summary

I think you’ll be seeing much more from this company down the road, but their vision to treat IT design just like any other analytical process is refreshing.