This chapter will focus on how to optimize Accumulo's performance. As Accumulo uses Hadoop and ZooKeeper, we need to start with optimizing their basic performance.
A common setup of Accumulo consists of multiple nodes and usually a rack-mounted server system. Setting up a network for a multi-node Accumulo is very important, and will cause problem if it's not done correctly.
In this chapter, the focus is going to be on the configuration part to optimize Accumulo (and subsystems) for performance. It might sound strange to focus on the configuration in this chapter, but the best way to optimize Accumulo for performance is to know what kind of tasks it is used for and use that knowledge to tweak Accumulo.