By Vikram Ramesh, Senior Director of Product and Digital Marketing –
Any enterprise IT worker can attest to the fact that IT availability doesn’t always equate to strong performance. Frustratingly, though, many Web apps run on cloud services that come with service-level agreements centered solely on availability without taking more valuable metrics into account. As long as an app is operating, there’s no issue with noncompliance of the SLA. Performance-based SLAs are unfortunately almost unheard of among cloud providers, but it’s crucial for latency-sensitive applications and industries to understand the difference between guaranteed availability and optimized performance.
Often, cloud vendors advertise their services as coming with a Five 9s guarantee, saying their platform will be available 99.999 percent of the time, with only planned, intermittent downtime as part of planned upgrades or maintenance. But that guaranteed availability doesn’t come with a baseline for guaranteed application performance. Applications on these cloud providers’ platforms will operate smoothly more often than not, but delays and latency extensions will happen from to time. When they occur, it doesn’t take long for end users to take notice.
There are obvious benefits for companies to move critical workloads to the cloud. The cost savings alone have drawn some, while others are intrigued by the perceived flexibility. Savvy enterprises must approach any SLA with a cloud vendor with some reservation, though. If a provider can’t – or won’t – give a defined performance standard the same level of consideration as availability, then it may not be the best time to migrate an app to its environment if it’s one to which employees and customers need performing at optimal levels at all times.
There isn’t a clear-cut guide to making the best decision for migrating an app to the cloud. Whether it’s a timing issue or the actual terms proposed by competing cloud services, it’s a consideration IT departments need to make based on their own needs. There are a number of statistics and metrics proposed by cloud vendors to show off their ability to manage applications workloads, but not all of them are applicable for all companies. Every organization wants to guarantee availability for their critical workloads. However, when availability guarantees don’t come with assured performance, it’s important for IT departments to understand where standard SLAs may not really be the standard on which to rely.
Availability is important, but your company needs guaranteed performance, too. Learn how IPM with Virtual Instruments can help you achieve that goal.