>
> As a follow up,
>
I unit tested mutli-threaded inserts, reads and updates up to 100000
entries and have seen awesome performance, lets say under 8 msecs average
while on a taxed VM - I did a simple unit test with a loop that drops
runnables into the thread executor. But, when I mix inserts and reads... it
falls apart quickly.
I've read the performance tuning page - not much about multi-threading....
can someone comment on some experiences with doing high concurrency with
H2? My application reads from multiple JMS queues/topics - each on its own
thread, then dumps it to the H2 DB. Other threads grab the messages and
"process" them - we are seeing 1+ second read times when under "light"
loads. I am using connection pooling with the apache BasicDataSource -
using the default - Im concerned because I see threads in the BLOCKED state
from inside the H2 DB Drivers.
Example code:
@GET
@Path("/flood")
@Produces("text/plain")
public String flood(@Context UriInfo info)
{
String s = info.getQueryParameters().getFirst("count");
int count = Integer.parseInt(s);
ExecutorService executor = Executors.newFixedThreadPool(25);
final Map<String,String> p = new HashMap<String,String>();
p.put("A", "value1");
p.put("B", "value2");
for (int i=0; i<count; i++)
{
final int idx = i;
Runnable r1 = new Runnable() {
@Override
public void run() {
long start = System.currentTimeMillis();
WebServices.this.incoming.storeMessage("MESSAGE-" + idx, "SOURCE-" + (idx %
5), 84600000, p);
long diff = System.currentTimeMillis() - start;
calc(idx, diff);
}
};
executor.execute(r1);
Runnable r2 = new Runnable() {
@Override
public void run() {
//long start = System.currentTimeMillis();
Object o = WebServices.this.incoming.getMessage("SOURCE-" + (idx % 5), -1);
//System.out.println(o);
//long diff = System.currentTimeMillis() - start;
//calc(idx, diff);
}
};
executor.execute(r2);
}
return "D0NE";
}
--
You received this message because you are subscribed to the Google Groups "H2
Database" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/h2-database.
For more options, visit https://groups.google.com/d/optout.