Hi Ali, The best practice is to use the noop scheduler on array of SSDs behind your block device (Hardware RAID controller). If you are using only one SSD disk, the deadline scheduler is the best choice to reduce IO latency. It is not recommended to set cfq on SSDs disks.
Regards, Roni Balthazar On 15 March 2015 at 09:03, Ali Akhtar <ali.rac...@gmail.com> wrote: > I was watching a talk recently on Elasticsearch performance in EC2, and they > recommended setting the IO scheduler to noop for SSDs. Is that the case for > Cassandra as well, or is it recommended to keep the default 'deadline' > scheduler for Cassandra? > > Thanks.
