5 Considerations for a hyper-converged solution

By Richard Hogg - Presales Consultant, Proact IT UK.

1 – What is your target application stack?

We need to consider how the applications will grow. VDI platforms look great on HC platforms, you just keep stacking boxes up as the user-count grows. But what if you have an ERP system or OLTP database that requires a large central database server which uses massive CPU and RAM allocations? You may find that you can’t deliver this in an HC node. How is your application licenced? Can you use the HC stack, or does it need to be segregated? How does this affect your strategic plan for the datacentre architecture?

2 – Where is the architectural pinch point?

Are you running a fairly balanced IT environment, with compute, RAM, network and storage all scaling fairly linearly? If so, great. But if not, you need to look under the covers of your proposed HC infrastructure. Are you going to deliver fixed blocks as you scale? Or can you deploy a skewed architecture with compute-only nodes? Or, perhaps a more storage-heavy environment, with either ‘fat’ nodes, or the ability to still use an external storage platform?

3 – How do I do DR or Backup?

Are you able to bring your existing backup platform with you, or are you being shoehorned into another backup platform? Do you need to layer on a third-party replication technology, or is it part of the HC stack? How easy is it to perform DR and what are the RTO/RPO levels you can set?

4 – But what if my infrastructure doesn’t need a refresh all at once?

Most IT budgets are set up around large infrastructure refreshes that are funded from a central pot. HC can be a benefit and a hindrance in these scenarios. You can pay for new projects as they arrive through the ability to scale in incremental chunks, but this needs a change in mind-set, and may cause you to have under-utilised infrastructure elsewhere for a lengthy period while you transition.

5 – How resilient is it?

Probably the most critical element of any business application infrastructure is the ability to withstand failures. Can the system handle the loss of power to a rack without corruption or data loss? Does it have the ability to ensure that at least one copy of data is outside a single rack? What happens in a node failure?

Is it right for you?  Sign up for our Hyper-converged Assessment Workshop

Sign up to Proact’s Hyper-converged Assessment Workshop and receive a free one-to-one consultation. In this 2-3 hour session we will discuss concepts, the current hyper-converged marketplace and use-cases to help you find out if hyper-converged is right for your organisation. For more information please contact us by clicking here.

Hyper-converged Infrastructure Webinar - Friday 27th March

This webinar offers an overview of the key vendors you should be taking notice of when considering hyper-converged infrastructure. It examines how HCI differs from conventional infrastructure, the potential risks, size of investment needed and when it may or may not be the right approach. Find out more here.