Strategy

Adapting right-sizing to the hybrid IT era

By Kong Yang, Head Geek™, SolarWinds
Strategy
Published: 27 June 2018

Right-sizing is a key business process when assessing performance against cost. Businesses and the C-suite should ensure they’re periodically conducting this proactive, strategic planning process to make sure they are operating with the correct level of resources. The same applies for IT environments.

Traditionally, right-sizing on-premises IT environments is an extensive and manual process, requiring significant time from IT professionals. This is because right-sizing for on-premises IT environments first requires determining speeds and feeds for compute, memory, network, and storage. With this information, IT professionals can then begin to manually scale infrastructure, networking, and application stacks to their optimal size. However, completing this process, while having to consider potential future requirements of the data centre, is a difficult and complicated task.

Migrating services to hybrid IT environments that span both on-premises and the cloud has dramatically altered the right-sizing process for IT professionals, as within cloud-based environments, right-sizing is very different. In hybrid IT environments, all of the underlying infrastructural matters are looked after by cloud service providers (CSPs), leaving internal IT departments with more time to concentrate on application performance. Working with CSPs also means that IT professionals can benefit from auto-scaling, the ability to automatically scale the resources that are being allocated to an application at any given time. In solely on-premises environments, it takes a significant amount of time to build out the required infrastructure services, but with the cloud, IT professionals can scale up or down with ease.

While cloud migration has made the process of right-sizing significantly less challenging in many respects, it has also introduced new issues as a result of working in collaboration with CSPs. Principally, it has become problematic for internal IT professionals to gain full visibility into the workloads, applications, and infrastructure that they are tasked with managing, once they have been migrated to the cloud.

To gain an optimal sense of clarity when right-sizing in hybrid IT environments, IT professionals and the C-suite should adopt the following best practice methods within their data centres:

Introduce baseline monitoring. To gain a clear understanding of what “normal” should look like in a data centre, you first need to know what the key performance metrics are across all services. Monitoring your baselines with discipline will make any deviations from normal performance easy to identify. Therefore, upon instances when performance output does deviate from the baselines, you can react quickly before Quality-of-Service (QoS) begins to suffer.

Use metrics to find the root cause. Once a system has been put in place to monitor your data centre’s performance baselines, you are then able to not only identify performance issues, but also determine their root cause. This can be done by combining and correlating time-series metrics with historical performance metrics from all applicable hybrid IT data sources – including applications, compute, network, storage, virtualisation, web, and cloud – into a single dashboard. This will then allow you to visualise the relationship between all possible causes, making the origin of the issue easy to identify.

Trust but verify. Working with CSPs has transformed the workload of many IT professionals, but migrating services and applications to a third party doesn’t mean that you are no longer responsible for them. To help ensure that the QoS to your organisation’s internal and external users is maintained, it is crucial that best practices are upheld and standards do not slip. Verifying your CSP’s architecture will go a long way in guaranteeing that this is the case.

With these best practice principles in place, IT professionals and the C-suite can then look to optimise their services in line with price and performance. There are huge cost benefits and performance improvements to be had when moving applications to the cloud, but there must also be clarity across your hybrid environment to help ensure that the correct decisions are made when right-sizing.

Right-sizing in on-premises environments is an important skill, and one that should not be neglected. However, hybrid IT is the current reality, and it is therefore vital that all IT professionals work to adapt this skill to make it applicable in the cloud.