Cache performance issues

Question

I was able to add indexes to a cache. This cache holds objects which contain all the rows in one of our database tables.
I run a sql query using hibernate on the database table, and then run the same query using filters on the cache. I am not noticing a significant performance gain. Is there something I'm doing wrong?
Here's what I'm trying to do:
QueryMap cache = (QueryMap)this.getPrimaryDAO().getCache();
Filter filterStateEq = new EqualsFilter("getStateCode", state);
Filter filterCompanyEq = new EqualsFilter("getCompanyCode", company);
Filter filterCoverageEq = new EqualsFilter("getCovCode", coverage);
Filter filterEffDateLE = new DateLessEqualsFilter("getEffectiveDate",
effectiveDate);
Filter filterExpDateGE = new DateGreaterEqualsFilter("getExpirationDate",
effectiveDate);
Filter filterAnd = new AllFilter(new Filter[]
{filterStateEq, filterCompanyEq,
filterLobEq, filterEffDateLE,
filterExpDateGE});
Set filteredSet = cache.keySet(filter);
Basically I'm trying to simulate a sql query:
select * from where state =
and company =
and covcode =
and ......

Answer 1

Hi Asim,
The code looks good, and it is quite natural to expect performance numbers similar to the DB considering you are executing the query on a single thread. When you will see the difference is under load on multiple machines. Let’s say you have 10 nodes configured as cache servers (storage enabled). Each of theses nodes will own approximately 10% of the total data in the cache. When a client thread calls a distributed query, it will be executed on each cache server node in parallel against the partial dataset they own. This means that with Coherence the performance of your queries will scale near linearly. Providing similar scalability with a typical commercial DB is going to be incomparably more expensive.
Best regards,
Gary Hawks
Tangosol