Why workloads define cloud strategies
Workloads vary widely, ranging from enterprise applications found in the typical datacenter to mobile apps born into the cloud. Most businesses begin using the cloud for distributed cloud native apps but rapidly realize the need to also incorporate the vast base of existing enterprise application workloads into a standardized cloud architecture. That can present challenges as the different categories of workloads each bears its own set of specific hardware, storage, networking, availability and redundancy requirements as well as different overarching architectural characteristics.
Traditional scale-up application workloads
The majority of existing enterprise applications that live in the datacenter fall into this workload category. Traditional applications achieve scale by scaling up to increase the size of the application and database infrastructure. As a result, the architecture of these workloads is somewhat limited in scalability.
Enterprise workloads are also traditionally designed to run on reliable hardware and utilize complex technologies to ensure availability. Traditional workloads require fault tolerant architectures and are built using enterprise-grade infrastructure components.
Cloud native scale-out application workloads
The new generation of applications typically associated with cloud computing fall into this workload category. These applications achieve scale by scaling out across many loosely coupled, commodity-grade computing, networking and storage nodes. As a result, these workloads are able to achieve dynamic scaling and elasticity.
Cloud native workloads are built for infrastructure that isn’t expected to be resilient. Instead, they are designed to simply and efficiently handle the failure of any given node. Because they don’t require enterprise reliability, these workloads can use commodity and open-source architecture components.
Consider both workloads for long-term success
Due to these inherent differences, workloads are one of the key factors organizations need to consider when developing a cloud strategy and selecting a cloud orchestration platform. By planning ahead to address both types of workloads, it’s possible to establish a solution that meets evolving use cases and long-term business requirements.