Hi,
I tried scaling ignite with 10, 20, 50, 100, 200, 500 threads for 1M gets.
The results were not as expected. For example, my application took 19 ms per
record for 20 threads which also took 49 ms per record for 50 threads.
Analyzing thread dump, I could see the below trace in numbers almost
equivalent to number of threads in every single dump I took.
*java.lang.Thread.State: WAITING (parking)
at sun.misc.Unsafe.park(Native Method)
at java.util.concurrent.locks.LockSupport.park(LockSupport.java:304)
at
org.apache.ignite.internal.util.future.GridFutureAdapter.get0(GridFutureAdapter.java:177)
at
org.apache.ignite.internal.util.future.GridFutureAdapter.get(GridFutureAdapter.java:140)
at
org.apache.ignite.internal.processors.cache.GridCacheAdapter.get0(GridCacheAdapter.java:4563)
at
org.apache.ignite.internal.processors.cache.GridCacheAdapter.get(GridCacheAdapter.java:4537)
at
org.apache.ignite.internal.processors.cache.GridCacheAdapter.get(GridCacheAdapter.java:1350)
at
org.apache.ignite.internal.processors.cache.IgniteCacheProxyImpl.get(IgniteCacheProxyImpl.java:907)
at
org.apache.ignite.internal.processors.cache.GatewayProtectedCacheProxy.get(GatewayProtectedCacheProxy.java:608)
at *
I have a transactional cache and another atomic cache. This waiting was
shown for both of them for get operations.
Any suggestions so as to improve this situation is most welcome. Is
increasing system thread pool an option to overcome this?
--
Sent from: http://apache-ignite-users.70518.x6.nabble.com/
