Which workloads are best run on-premises?

Which workloads are best run on-premises?

We’ve talked about the benefits of Hybrid IT, which allows organizations to run some applications in the cloud and others on-premises.

This ability to customize your workload placement sounds great in theory, and for most organizations a hybrid infrastructure truly is the best choice. But even if you understand the benefits and know that 100% cloud isn’t right for you, you still have plenty of decisions to make.

You’re certainly not alone if you’ve ever second-guessed your strategy, or wondered “which workloads do I run where?”

It’s not easy. Determining the right workload placement strategy is tough, and every organization is different. The workloads you currently (or plan to) run, as well as the lifecyle of those applications, play a key role. But considerations such as company size, security, cost and your IT staff all factor into the decision too. Fortunately, there are guidelines to help you navigate this path.

Workload lifecycle can impact your workload placement strategy

To create the most effective strategy, you want to start by defining your objectives, top priorities (security? response time? performance? dependability?) and most used workloads. Once you’ve identified these, you’ll have a clearer vision to help guide your strategy. And while it’s true that many workloads can run in the cloud or on-premises, there are certain workloads that make more sense to run on-premises.

So, which ones are they?

Workloads That are Best Left On-Premises

Organizations that have a choice where to run their workloads typically choose to run mission critical workloads on-premises. When performance is critical, you’ll want to keep those applications onsite and maintain tighter control. Although there are no absolute rules, a good rule of thumb is to run the following workloads on-premises:

Unstructured Data Analytics

Many businesses choose to run unstructured data applications on-prem for security reasons. In fact, 31% of hybrid IT organizations report that they choose to run unstructured data analytics on-premises because their IT infrastructure is more secure (than running in the cloud).[i] Response time is also critical with unstructured data analytics, so running on-prem helps organizations respond faster to business needs.

Structured Data Management & Analytics

Structured data management and analytics often use sensitive information, so security is a top priority.[i] In fact, 37% of hybrid IT organizations cite “our on-premises IT infrastructure is more secure” as a top reason they choose to run structured data management and analytics on-premises.[i] Organizations often need to consider regulatory and compliance issues, so running on-prem gives firms a greater piece of mind. Not to mention it can also be more cost efficient.

Business Applications (including CRM & ERM)

Business applications are often revenue-generating and considered “mission critical” to the business. You need them to keep daily operations going, as disruption would cause significant problems. Because reliability is an absolute must, most organizations want to maintain tight control and choose to run business applications on-prem.

Engineering/Technical

Engineering and technical applications often have custom requirements that need training or specific expertise. They may be for innovation or R&D testing (intellectual property) so security is critical. They are also commonly subject to regulatory and compliance laws. For these reasons, engineering and technical workloads are best run on-premise.

No two businesses are alike. Firms grow, technologies advance and priorities change. For these reasons, it’s critical to develop a workload placement strategy based on your specific needs, resources and business objectives.

[i] Source: Workload Placement Separates the Winners from the Losers in IT, an IDC White Paper Sponsored by Dell EMC

 

 By Emily O'Shaughnessy

Published with permission from https://blog.dellemc.com/en-us/