Reading Performance in Hadoop Cluster

Question

I have an average key-value pair size of 100 bytes. My primary access is random needs on the table. I want to speed up my reading performance on Hadoop Cluster. So, what should I do that will speed up random reading performance on my cluster?

Can someone help!

Thanks in advance!

Answer 1

For speeding up reading performance of your cluster one thing you can do is that you can decrease the block size because larger block size is preferred if files are primarily for sequential access.  Smaller blocks are good for random access, but require more memory to hold the block index, and may be slower to create.

Hope this will clear your doubt.