Surprising Cloud Constraints You Don’t Have On-Prem

Moving to the cloud from the datacenter opens your organization up to a world of new possibilities. You’re able to offload the day-to-day management, necessary maintenance, and exorbitant costs of an on-premises datacenter. Your team can focus on more high-value projects that have a direct impact on the business. And if your organization is taking steps to modernize, you’re better positioned to do just that: whether it’s through the integration of new technologies or by offering new business models like a SaaS offering of your product.

The benefits and possibilities of the cloud are well-documented. What isn’t is the constraints you might run up against as you enter the world of cloud and leave the datacenter behind. In this post, we’ll take a look at some aspects of the cloud that might leave much to be desired and how you can overcome them.

Achieving the Speed You Need

Depending on what sort of workloads you are running on the cloud, the performance you get might be enough. But what if your organization recently decided to go completely all in on the cloud and now you have to move your databases out of the datacenter? That’s when things start to slow down…

Let’s first understand how performance on the cloud works: When you purchase cloud resources from a public cloud vendor, the level of performance you get is based off of the amount of resources you have. The more you invest in the cloud, the faster your speeds will be. But what if you need even more speed then you’re automatically allocated? Then you need to overprovision with even more cloud resources – cloud resources you don’t necessarily need or will use. This can make fast performance on the cloud both pricey and wasteful.

But what if you need even FASTER performance? The fastest money can buy! Well, then – let’s talk about throttles. Since the public cloud is a shared environment amongst customers, the cloud vendors will put limits – or throttles — on how much performance any single customer can achieve in order to keep performance steady for the entire customer base. Meaning the fastest speeds money can buy might STILL not be fast enough if you have workloads with ultra-high performance needs.

Cloud Efficiency

Another constraint you’ll find as you adopt the cloud more and more is how efficient your cloud usage is. On-prem you have multiple data services – such as snapshots, deduplication, and thin provisioning – that help to minimize how much storage space is being used. In the public cloud, these data services are not available out of the box. So if your team is taking snapshots of workloads for Dev/Test purposes, you’re going to use up additional cloud resources.

Making it even trickier is if the team then forgets to turn off these additional resources when they’re done using them, or during down times such as nights and weekends. Additional copies of data that aren’t being used are just wasting more and more cloud resources – and unnecessarily driving up your cloud bill.

Overcoming Cloud Constraints with Silk

The Silk Cloud DB Virtualization Platform makes it easy to overcome these cloud constraints and get the most performance for the least amount of cloud resources. The platform sits between your cloud database and storage layers. From there, it optimizes performance with no additional database changes or tuning necessary. It does this by uniquely decoupling performance from cloud capacity, enabling continuous performance optimization that is significantly higher than what the public cloud providers can offer.

In addition, Silk leverages those same enterprise data services you’re use to from your on-prem days. These data services increase the efficient use of cloud resources by shrinking your capacity footprint – and in turn, your cloud bill.

Sentara Healthcare’s Story

One company that was able to leverage Silk’s superior performance and enterprise data services was Sentara. A not-for-profit healthcare organization in Virginia and North Carolina, Sentara had decided to move its electronic health records (EHR) system onto the Microsoft Azure cloud. The migration process was proving to be difficult since Azure natively wasn’t able to give Sentara the performance they needed.

On top of that, the team projected that their current cloud spend would grow 10% year-over-year. This was unacceptable! The team needed faster performance to enable cloud adoption, while also ensuring they stayed within budget.

With Silk, Sentara was able to achieve 3x faster performance than they were seeing on native cloud alone. And with Silk’s data compression services, they could reduce the number of cloud resources used by 2:1, helping to cut their cloud costs by up to 20%.