honestly, I don't know the answer to that, but I'll take a guess. In request only mode, ActiveMQ is probably returning a response immediately, then queueing up the message. In could be that in that case, it batches the insert, so there could be a longer delay. In contrast, request response has a longer elapsed time from the thread perspective but shorter time from postgresql.
I would ask the activeMQ guys to get a definitive answer. James Strachan is usually good about providing a good explanation. peter On 12/12/05, Kim Loh <[EMAIL PROTECTED]> wrote: > > Hi, > > Based on Point-To-Point Sampler in JMeter, > > 1. For Communication Style drop list, May I know the difference > between > Request Only and Request Response? > > 2. From JMeter Log, the results obtained from above two options are > below: > Why there is the difference elapsed time sending message to postgresql > database between these two methods which the Request Only slower than the > request response using the same message via JMS ActiveMQ provider. > > Communication Style: Request Response > ------------------------------------- > Thread Started : 16:42:13 > Thread Done : 16:42:18 > > ==> TimeStamp in Postgresql: 16.42.15 > > > Communication Style: Request Only > ------------------------------------- > Thread Started : 16:17:24 > Thread Done : 16:17:24 > > ==> TimeStamp in Postgresql: 16.17:49 > > > Thanks. > > Regards, > Kim > >
