Soon the backup power will be available for our new datacenter and the redesign to make use of VMware vCloud Suite is nearing completion. Soon, our full private cloud will be ready for our existing workloads. These workloads however now run within a XenServer based public cloud. So the question is, do we stay in a poorly performing public cloud (mentioned in our Public Cloud Reality series) or move back to our own private cloud? As the Clash put it “Should I Stay or Should I Go Now.”
We are actually leaning towards moving on or more importantly back to our own Private Cloud where we can optimize for the performance we require. For a small enterprise this is always a tough decision, on one hand you have the cost of maintaining your own systems and the other you have the fact that we would be maintaining our own systems so we could test and play with new and old products. On the one hand, I gotta say the Cloud has been relatively maintenance free, well so has our own data center and private cloud. We occasionally change things in our own private cloud, but not all that often. VMware vSphere “Just Works”.
So with Maintenance Costs not an issue, staying in the cloud ends up being an extra expense to us. So why should we stay in the cloud? Is it more secure? Is it a better performer? Do I have the visibility I did not have within our own data center? Why stay in the cloud if you do not gain more from it than being in your own data center? Is this part of an upgrade path? Are we future proofing our applications? Does the Cloud alleviate a single point of failure?
Ultimately everything boils down to cost
I brought this up on a cloud talk where I was told that a CIO will stay in the cloud if they get better compliance reporting. My response was that the CIO will stay in the cloud only if it is cost effective. Ultimately, that is the decision to make.
You may recall we moved to the cloud not to increase security (it is less secure than we had before), not to increase performance (it is slower than before), not to gain visibility (we had more access in our own data center and have the same logging mechanisms in place, but instead to migrate our data center nearly 2500 miles. For us the cloud was about staying up and available. If the cloud performed better, was more secure, there was a good chance we would stay.
Public Cloud Reality: Should We Stay?
The reality is that IaaS security is what you make of it, performance is dependent on noisy neighbors and your systems not being too noisy as well, maintenance is no better or worse for well designed cloud applications, and you have to spread the load to get better performance which eats through your budget quickly. Big systems do not work well in a public cloud, yet many small systems can. Which means you need a well designed cloud application to take advantage of the public cloud fully.
We are leaning to pulling back from the cloud, our original goal was met, now can we meet the rest of our goals using a private cloud. What would you do?
Share this Article:
Latest posts by Edward Haletky (see all)
- Continuous Integration, Deployment, and Testing - July 22, 2016
- Serverless: Business Plan or an Approach to Technology? - July 21, 2016
- Root Cause Analysis Is Not Dead - July 13, 2016