db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Olav Sandstaa <Olav.Sands...@Sun.COM>
Subject Re: Options for syncing of log to disk
Date Fri, 01 Sep 2006 13:00:44 GMT
Suresh Thalamati wrote:
> Olav Sandstaa wrote:
>> I have run some simple tests where I have changed mode from "rws" to
>> "rwd" for the Derby log file. When running a small numbers of
>> concurrent client threads the throughput is almost doubled and the
>> response time is almost halved. I am enclosing two graphs that show
>> this when running a given number of concurrent "tpc-b" clients. The
>> graphs show the throughput when running with "rws" and "rwd" mode 
>> when the
>> disk's write cache has been enabled and disabled.
>
> I am really surprised to notice the difference between rws / rwd 
> throughput in your graphs,  Especially when the write cache is 
> enabled.  When I ran small tests long time ago, did not notice any 
> substantial difference between rws/rwd, on windows for sure.

I was not surprised of the results with the disk's write cache disabled, 
but very surprised by the large effect then the write cache was enabled. 
I had expected to so hardly any difference in that case. This was on 
Solaris 10, on Linux 2.6 I did not see any difference between these case 
(both with and without a write cache on the disk).

> Could u please post more details on the hardware/OS/JVM ..etc.  and 

The hardware is a dual Opteron server with two reasonable fast Fujitsu 
SCSI disks (one for log and one for the database). If I remember 
correctly the disks have a 6 MB cache (shared between reading and 
writing), Solaris 10 x86 and Sun JVM 1.5 update 4.

> What is the work done in each transaction ?

The transactions are similar to TPC-B trasactions. Each consists of 
three updates, one insert and one select.

Olav


Mime
View raw message