If you have an HC3 system or are thinking about using an HC3 system, your HC3 system most likely is or will be running some or all of the critical workloads in your organization. Unfortunately, nothing lasts forever and those systems will at some point need to be scaled out, repurposed, or retired. It is never too early to start planning for or at least thinking about when and how you may need to make changes. Here are the most common examples we see.
Scale Out
There may come a point where your HC3 systems are running out of disk space, RAM, or CPU capacity. In fact, we encourage you to not over-buy and over-provision because you get a better value by waiting to buy more capacity when it is needed rather than 3-5 years before it is needed. That’s why we made it so easy to scale out when needed.
With HC3 you can add more capacity seamlessly to your existing HC3 cluster and you can add in a new node that is configured to meet your resource needs. You can choose an appliance configuration heavier on storage, heavier on RAM, heavier on CPU, or all of these if needed. Mixing and matching nodes is one of the key benefits of HC3 over many of our competitors. There are some common sense limitations within all of the possible combinations of nodes but we can help you choose the appliance and configuration to meet your needs. Scale out is also the time to consider whether you may want to start transitioning to an all-flash system and plan accordingly, knowing that this new node may outlive your existing nodes. In some cases, adding a second production HC3 cluster makes more sense than expanding an existing one. You can manage multiple HC3 clusters in our web UI “at a glance”.
Replace and Repurpose
Another option when your resource capacity is in question is upgrading your cluster by buying newer, bigger, faster nodes and repurposing your old cluster or in some cases selected nodes. Even if you have the capacity you need now, but are looking at another need like DR, replacing your primary cluster and repurposing your old gear may be the best option for future planning.
The old cluster or select nodes can then be repurposed for another site, DR, test and dev, or any other purpose you desire. A second cluster or multiple HC3 systems are still easy to manage within the HC3 web interface, using the remote clusters tab for monitoring. Repurposing increases your ROI and gives you more resources to deploy elsewhere. Re-deploying at another site closer to users might reduce their connection latency. DR sites generally need less capacity which is perfect for an older, smaller cluster or single node and tend to be the most common use case we see for repurpose.
Retirement
Eventually, your nodes will simply need to be retired for good. If you have planned ahead, you’ve likely added new nodes over time for scale-out and as the oldest nodes reach their end of life, they can more gracefully be retired from the cluster, with newer nodes taking their place. This prevents all the nodes in the cluster from reaching end-of-life at the same time, increasing the risk of failures. It also helps with budget planning to have a regular schedule for adding new nodes to the cluster over time rather than requiring a full forklift upgrade.
With HC3 clusters you can add new nodes to a cluster and evict retiring nodes without any downtime if you are replacing retiring nodes with cluster compatible nodes. Since we are regularly refreshing our appliance lines with newer and better hardware, the new nodes you add to replace the retiring nodes will likely be superior in every way to the retiring nodes. That is our goal.
Summary Regardless of where you are in the life cycle of your HC3 system, it’s never wrong to think about the next step in the life cycle. Feel free to contact us if you have questions. We are happy to help walk you through any and all of your options for moving forward.