Re: random-access I/O

[Aftab Khan]

>So given that the READs are the slow part of the whole process, the only thing that I can do to speed this up is increase the buffer_pool correct?

Please read James and my replies on your differernt topic 'Mass Import Innodb' which I believe relevent to your query:

http://forums.mysql.com/read.php?22,512689,514019#msg-514019
http://forums.mysql.com/read.php?22,512689,514164#msg-514164
http://forums.mysql.com/read.php?22,512689,514171#msg-514171

>Finally in one of your previous replies you mentioned that partitioning could help me. In my case would it be better to use RANGE or KEY partitioning? With the first option most of the queries will draw data from 1st partition (since the records are sorted by popularity). On the second option the "popular" records will be stored across all partitions. Does MySQL reads multiple partitions at the same time for the same query or is this done in a serial manner?

read this article to understand how mysql decide partition(s) to fecth required records:
http://dev.mysql.com/doc/refman/5.5/en/partitioning-pruning.html

Range partitioning - you would have to add a new partition probably each day to keep more "popular" records and from other end you can archive (if required) and delete the oldest partition (the least popular one). However, with KEY partitioning you cannot drop partitions from tables that are partitioned by HASH or KEY in the same way that you can from tables that are partitioned by RANGE or LIST.

http://dev.mysql.com/doc/refman/5.5/en/partitioning-management-range-list.html
http://dev.mysql.com/doc/refman/5.5/en/partitioning-maintenance.html

If your client(s) would hit the most popular partition then this would improve READs.

Let us know if you require more assistance.