IT workloads grow like sponges taking on water: soaking up resources as they become available.
Data center resource scalability gained importance as digital technology started to enable every kind of business. Data center capacity planners define scalability as the back-end IT deployment's ability to grow proportionally with demand, without adding complexity or difficulty to its management.
E-commerce and online transaction processing, as well as content distribution over the Internet, prove the value of scalability. If a T-shirt maker cannot scale up its internal customer relationship management software and its Web server farm to meet increased demand, for example, the business will suffer lost orders and a damaged reputation.
Some IT organizations turn to cloud bursting to handle unexpected or unpredictable short-term rapid scaling up. Data center managers plan capacity to suit the maximum typical demand of IT workloads; beyond that, they rent resources from a cloud provider as needed. The T-shirt maker from our previous example may organically scale up data center servers and storage as its customer base grows, then also burst onto Amazon Web Services' EC2 cloud instances during the busy holiday shopping season. Both methods fit the definition of scalability.
To prove a depth of knowledge in capacity planning, define scalability in contrast with elasticity. Elasticity matches resources to a dynamic IT workload -- when it needs more resources, they scale up; when it needs less, they scale down. This dynamic reallocation is more difficult to achieve with physical servers and systems in a data center than with pay-per-use cloud resources, but with virtualization as well as cloud bursting, it is possible.