Re: Bundling postgreSQL with my Java application - Mailing list pgsql-performance

Saurabh Dave wrote: <blockquote cite="mid:fdd0c0f90907060017y334df602k4fff0acf07776d91@mail.gmail.com"
type="cite"><divclass="im">>No offense intended - but have you looked at the documentation for postgresql.conf?<br
/><br/> >If you are going to include PostgreSQL in your application, I'd highly recommend you >understand what
youare including. :-)<br /><br /></div> I had a look into the documentation of postgres.conf, and tried a lot with
changingparamters I thought would improve the performance, but in vain.<br /> Autovaccum is enabled by default in 8.3.7
,but i reduced the nap time so that it happens more frequently.<br /></blockquote> As others have pointed tuning is not
acaned answer  hence all the config options to start with.  But to change the configuration to something a bench mark
mustbe made.  The only way to do that is identify the common SQL commands sent to the server then run explain analyze 
soyou know what the server is doing.  Then post the the results along with Config file and we can make suggestions <br
/><br/> There is  <a class="moz-txt-link-freetext"
href="http://wiki.postgresql.org/wiki/Performance_Optimization">http://wiki.postgresql.org/wiki/Performance_Optimization</a><br
/><br/> Greg Smith is working on a tuner  <a class="moz-txt-link-freetext"
href="http://notemagnet.blogspot.com/2008/11/automating-initial-postgresqlconf.html">http://notemagnet.blogspot.com/2008/11/automating-initial-postgresqlconf.html</a><br
/><br/> But thats a monumental undertaking as one configuration setting for one type of work load can be ruinousness to
anotherwork load.<br /><br /> The one common theme is know the workload so the configuration matches.  <br
/><blockquotecite="mid:fdd0c0f90907060017y334df602k4fff0acf07776d91@mail.gmail.com" type="cite"><br /> My personal
opinionis that certain parameters in postgres.conf are simply too technical in nature for a application developer like
me,it becomes more of a trial and error kind of frustrating process.<br /></blockquote> This boils down to know the 
workload. <br /> different kinds of work loads:  <br />     A: more writing with very few  reads.<br />     B: more
readsthat are simple queries and few complex quiers with very few writes.  There is a ratio to look at in my case 10000
readsoccur before next write So we have lots of indexes aimed at those common queries.  <br />     C: Complex queries
takingminutes to hours to run on data warehouse covering  millions of records.<br />     D: equal work load between
writesand reads.  <br /><br /> There are many kinds of workloads requiring different configurations.  <br /><blockquote
cite="mid:fdd0c0f90907060017y334df602k4fff0acf07776d91@mail.gmail.com"type="cite"><br /> If there a utility that
understandsthe system specification on which postgres is going to run and change the paramters accordingly, that would
help.<br/><br /> Thanks,<br /><font color="#888888">Saurabh</font></blockquote> <snip><br /><br />