Sajal, thanks for your input! I have almost finished working on the fix, it should get to one of the next releases of Ignite. You can track this ticket https://issues.apache.org/jira/browse/IGNITE-1159 to see when it's merged to master.
2015-07-29 6:34 GMT-07:00 Yakov Zhdanov <[email protected]>: > Sajal, thanks for pointing out and investigation. This is a known issue > that on very large batches there are slowdowns in MVCC. I believe this has > been fixed by Alex Goncharuk and will be released soon. Alex is this the > case? > > --Yakov > > 2015-07-29 14:07 GMT+03:00 Sajal Gupta <[email protected]>: > >> Trying the below sample where I am putting 20000 entries in cache within >> transaction on a single node(JRE 7, heap 4GB),it takes more than 20sec to >> commit. The thread dumps below shows thread executing >> GridCacheMvccManager.addNext most of the time.The same sample with >> pessimistic mode takes less than 5sec >> Let me know if there is any other setting that should be specified or is >> it the best we can get considering everything is happening in memory on >> single node? >> URL url = >> TestCache.class.getResource("/unittest/gridgain/example-ignite.xml"); >> Ignite ignite = Ignition.start(url); >> System.out.println(); >> System.out.println(">>> Cache API example started."); >> >> CacheConfiguration cfg = new CacheConfiguration(); >> cfg.setName("myCache"); >> cfg.setAtomicityMode(CacheAtomicityMode.TRANSACTIONAL); >> cfg.setCacheMode(CacheMode.PARTITIONED); >> cfg.setMemoryMode(CacheMemoryMode.OFFHEAP_TIERED); >> cfg.setRebalanceMode(CacheRebalanceMode.ASYNC); >> >> cfg.setWriteSynchronizationMode(CacheWriteSynchronizationMode.FULL_ASYNC); >> cfg.setOffHeapMaxMemory(2 * 1024L * 1024L * 1024L); >> IgniteCache<String,byte[]> cache = ignite.createCache(cfg); >> byte[] valArr = new byte[16000]; >> for(int i=0;i<valArr.length;i++){ >> valArr[i] = 0x0; >> } >> long t1 = System.currentTimeMillis(); >> try (Transaction tx = >> (Transaction)Ignition.ignite().transactions().txStart(TransactionConcurrency.OPTIMISTIC,TransactionIsolation.REPEATABLE_READ)){ >> for(int i=0;i<20000;i++){ >> cache.put("K"+(long)i,valArr); >> } >> tx.commit(); >> } >> long t2 = System.currentTimeMillis(); >> System.out.println("Time taken:" + (t2-t1)); >> >> >> "main" prio=6 tid=0x0000000002c33800 nid=0x454 runnable >> [0x0000000002b4e000] >> java.lang.Thread.State: RUNNABLE >> at java.util.LinkedList.add(LinkedList.java:336) >> at >> org.apache.ignite.internal.processors.cache.GridCacheMvccManager.addNext(GridCacheMvccManager.java:733) >> at >> org.apache.ignite.internal.processors.cache.distributed.dht.GridDhtCacheEntry.addDhtLocal(GridDhtCacheEntry.java:230) >> at >> org.apache.ignite.internal.processors.cache.distributed.dht.GridDhtCacheEntry.tmLock(GridDhtCacheEntry.java:244) >> at >> org.apache.ignite.internal.processors.cache.transactions.IgniteTxManager.lockMultiple(IgniteTxManager.java:1586) >> at >> org.apache.ignite.internal.processors.cache.transactions.IgniteTxManager.prepareTx(IgniteTxManager.java:882) >> at >> org.apache.ignite.internal.processors.cache.transactions.IgniteTxLocalAdapter.userPrepare(IgniteTxLocalAdapter.java:410) >> at >> org.apache.ignite.internal.processors.cache.distributed.near.GridNearTxLocal.prepareAsyncLocal(GridNearTxLocal.java:852) >> > >
