As covered in the previous chapters, latency is the delay that occurs before an operation can complete, sometimes also known as lag. You may not be able to control the latency of the infrastructure that your software application runs on, but you can write your application in such a way that it can cope with this latency in a graceful manner.
The two main types of latency that we will discuss here are network latency and disk latency. As the names suggest, these are, respectively, the delay in performing an operation over the network and the delay in reading from or writing to a persistent storage medium. You will often deal with both at the same time, for example, a database (DB) query to a server on a remote virtual machine will require the following operations:
- A network operation from the web server to the DB server
- A network operation from the DB server to a remote disk on a Storage Area Network (SAN)
- A disk operation to look up data on the physical drive