Hyperconvergence is fast becoming a de-facto standard because it delivers a flexible easily manageable IT Infrastructure based on low-cost commodity x86 hardware.
However, the reduced management and hardware costs associated with a Hyperconverged Infrastructure (HCI) deployment doesn’t preclude the need for proper planning especially when it comes to capacity management.
In fact, capacity planning when done right can yield quite a bit of value with a minimal amount of time and effort.
What is HCI and why should you care?
A hyperconverged infrastructure collapses the compute, network, and storage resources into a single manageable HCI node. The nodes are then connected (usually with little effort) to clusters that quickly scale out, “rack and stack” would be an accurate characterization.
Because the physical layers are abstracted, an HCI node is essentially an appliance. The resulting HCI based IT infrastructure is infinitely easier to manage and deploy as compared to a traditional (some would say legacy) siloed IT infrastructure - because the abstraction also collapses the required core IT management competencies (and costs) to a single platform.
Another catalyst driving HCI is hardware costs - which continue on a downward trajectory. Technologies like Flash continue to drop in price while at the same time CPU capacity is increasing; making supporting an ever-expanding workload with HCI a relatively easy proposition.
Need additional storage to accommodate more VMs or containers? The scaling-on-demand nature of hyper-convergence means it is as simple as adding an HCI node.
Another benefit of bundling everything in a chassis means there is a single throat to choke. When there is a problem, a single accountable vendor avoids finger pointing.
What is Proper Capacity Planning for HCI?
IT organizations are under increasing pressure to provision workloads quickly and efficiently. As on-premises hardware becomes increasingly commoditized there is downward budget pressure. Conversely software (i.e. public cloud deployments) are taking a significant larger bite out of IT budgets.
If we take a step back and look at siloed IT infrastructures. IT organizations had to plan and build their IT infrastructure to handle 3 to 5 years of growth. Making adjustments to these silos, especially later on, required core competencies that were highly specialized and expensive. Furthermore altering silos was disruptive.
In fact, in researching Longitude capacity planner users operating siloed environments we found that it wasn’t uncommon to see them running at a significant resource surplus.
Step forward to the incremental nature of HCI nodes, IT organizations need to be ahead of the curve. Ultimately the goal is the implementation of a pay as you go strategy, where newly provisioned resources align closely with workloads. It is about minimizing IT infrastructure and software costs. Keep in mind this isn’t a panacea as you can’t simply add a single resource, resources must be added as single block of CPU, memory, and storage.
How does Longitude help?
Automatically collects key physical and virtual performance metrics from hosts and VMs.
Longitude’s built-in knowledge base understands what key performance indicators (KPIs) to collect, how often to collect them, and most importantly how to evaluate them. This means automating the consistent collection and analysis of metrics for both hardware and software components
Alert on potential performance or availability problems for both physical and virtual components.
A fundamental principle of effective capacity planning is communicating resource depletion issues to the right people and at the right time.
Keeping the problem information relevant and timely goes a long way towards a capacity planning strategy that is proactive.
Generate comprehensive reports to show utilization and capacity issues for both physical and virtual components
Capacity planning with “What if” analyses predicts how potential changes will affect IT operations
“What if" analyses predicts where and how potential changes in resources (i.e. CPU and memory allocation) will affect operations and licensing costs.
HCI capacity planning need not be complicated or expensive. If you’re looking for comprehensive monitoring with dashboards, alerting, historical reporting, and capacity planning that’s where technology like Longitude can help.
Standing up Longitude takes just a few minutes and you’ll immediately start seeing how your hyperconverged infrastructure is performing, as well as where and how future resources will be depleted.