Re: CPU Bound Writes

2013-07-23 Thread aaron morton
That's very old documentation, try using the current docs. Although the statement is syntactically correct, it will become CPU bound before becoming memory bound. That statement says nothing about the IO use. Cheers - Aaron Morton Cassandra Consultant New Zealand @aaronmorton

Re: CPU Bound Writes

2013-07-21 Thread Mohammad Hajjat
Aaron, here is the source: http://www.datastax.com/docs/0.8/cluster_architecture/cluster_planning Thanks! On Sun, Jul 21, 2013 at 4:57 PM, aaron morton wrote: > > Wouldn't this make Writes disk-bound then? I think the documentation may > have been a bit misleading then "Insert-heavy workloads w

Re: CPU Bound Writes

2013-07-21 Thread aaron morton
> Wouldn't this make Writes disk-bound then? I think the documentation may have > been a bit misleading then "Insert-heavy workloads will actually be CPU-bound > in Cassandra before being memory-bound"? What is the source of the quote ? Cheers - Aaron Morton Cassandra Consultant

Re: CPU Bound Writes

2013-07-20 Thread Mohammad Hajjat
Thanks/Shukran, Jon! :) Wouldn't this make Writes disk-bound then? I think the documentation may have been a bit misleading then "Insert-heavy workloads will actually be CPU-bound in Cassandra before being memory-bound"? On Sat, Jul 20, 2013 at 12:12 PM, Jonathan Haddad wrote: > Everything i

Re: CPU Bound Writes

2013-07-20 Thread Jonathan Haddad
Everything is written to the commit log. In the case of a crash, cassandra recovers by replaying the log. On Sat, Jul 20, 2013 at 9:03 AM, Mohammad Hajjat wrote: > Patricia, > Thanks for the info. So are you saying that the *whole* data is being > written on disk in the commit log, not just som

Re: CPU Bound Writes

2013-07-20 Thread Mohammad Hajjat
Patricia, Thanks for the info. So are you saying that the *whole* data is being written on disk in the commit log, not just some sort of a summary/digest? I'm writing 10MB objects and I'm noticing high latency (250 milliseconds even with ANY consistency), so I guess that explains my high delays?

Re: CPU Bound Writes

2013-07-19 Thread Patricia Gorla
Kanwar, This is because writes are appends to the commit log, which is stored on disk, not memory. The commit log is then flushed to the memtable (in memory), before being written to an sstable on disk. So, most of the actions in sending out a write are writing to disk. Also see: http://www.data

CPU Bound Writes

2013-07-19 Thread Kanwar Sangha
"Insert-heavy workloads will actually be CPU-bound in Cassandra before being memory-bound" Can someone explain why the internals of why writes are CPU bound ?