Yeah, that's really microsecond latency. Note, though that this isn't the full request timing, its just the storage proxy down, so it doesn't account for any latency added by thrift or the network.
-ryan On Wed, Nov 10, 2010 at 1:43 PM, Rock, Paul <paul.r...@teamaol.com> wrote: > Afternoon all - I'm playing with 0.7beta3 on some boxes I have here at the > office and while checking out the stats from one of my tests I'm seeing Write > Latency being reported as "0.009 ms". I haven't done any timing yet in my > client, but is this really microsecond latency, or is there a mismatch > between the numeric and the label? Granted, I'm not loading the complex up at > all just writing with a single thread to play with pycassa so the cluster > doesn't have anything to do but handle my write, but I'd like to make sure > before I run off trying to talk my manager into something :-) > > Column Family: NameServer2Domain > SSTable count: 0 > Space used (live): 0 > Space used (total): 0 > Memtable Columns Count: 39718 > Memtable Data Size: 2531109 > Memtable Switch Count: 0 > Read Count: 0 > Read Latency: NaN ms. > Write Count: 39718 > Write Latency: 0.009 ms. > Pending Tasks: 0 > Key cache capacity: 200000 > Key cache size: 0 > Key cache hit rate: NaN > Row cache: disabled > Compacted row minimum size: 0 > Compacted row maximum size: 0 > Compacted row mean size: 0 > >
