Servers are used in every data center to supply computing resources. These include — processing cycles, memory space, network, and disk I/O among others. With the increase in workload and computing demands, server resources must scale to meet those demands. So, we need to scale:-scale up and scale out. In this blog post, we will tell you the best way to do so. Note that both are not the same and we will tell you why.
Scale out: You add more server resources in order to handle the workload. For instance, you have 8 servers running with an approximate usage of 90% overall. If you plan to scale the workloads for your business, the current number of servers might not be sufficient to handle it. So you scale up by adding more number of servers.
Scale up: With advances in technology and computing power there has been a vast increase in server resources for each new design. It is now possible to replace an aging server with a model utilizing far more processing, memory, and I/O capability than previous versions. But the catch here is that it occupies the same physical footprint and often consumes less energy. This approach is known as “scale-up”
These are the two provisions available for scaling server provisions. But how does one identify when to scale up? If you have the following ticks on your checklist then it’s time to scale your server provisions.
- If your networks and physical machines are shared with other companies, you might face some issues. You can’t rely on the consistency of performance in any three dimensions: network, storage and compute.
- This would mean that the response time of each node is not constant and predictable.
- There is unavailability of nodes reduces the overall response time
Both scale-up and scale-out approaches are efficient means of adding computing resources to a data center environment. What needs to be understood here is that these two are not mutually exclusive. A scale-out could be the right approach when a large number of smaller nodes are needed. On the other hand, a scale-up server approach might be right for a major virtual server consolidation initiative. Here more workloads must reside on the fewer physical servers. The reliability of scale-up and scale-out servers are equally good. It’s time to decide and upgrade your servers!
How do you start? Both AWS and Lenskart introduces a new series of LearnUps which will teach you the capabilities and power of the AWS platform and how it can be leveraged to give your software the boost it needs and Lenskart will help you understand as to how to manage your infrastructure as they do.