Hello All-
We are running 5 node v0.94 cluster on machines w/ 32G of RAM. I'd like to utilize more of our memory for block cache. So I have a couple of questions. Currently, we are running RSs w/ 16G heap and the following memory settings: hfile.block.cache.size=0.25 hbase.regionserver.global.memstore.upperLimit=0.2 hbase.regionserver.global.memstore.lowerLimit=0.15 Typically overall stats for RS are: numberOfOnlineRegions=228, usedHeapMB=7870, maxHeapMB=16358 Used heap typically goes to ~10G then we get GC back down to about 6-8G. So by my reckoning we have about 25% of heap unutilized. More detailed, typical RS info: numberOfOnlineRegions=245, numberOfStores=245, numberOfStorefiles=249, storefileIndexSizeMB=17, rootIndexSizeKB=17688, totalStaticIndexSizeKB=11910, totalStaticBloomSizeKB=36107, memstoreSizeMB=361, readRequestsCount=-1829306476, writeRequestsCount=16153099, compactionQueueSize=0, flushQueueSize=0, usedHeapMB=11407, maxHeapMB=16358, blockCacheSizeMB=3201.95, blockCacheFreeMB=887.66, blockCacheCount=3243, blockCacheHitCount=2932634895, blockCacheMissCount=5223350, blockCacheEvictedCount=678037, blockCacheHitRatio=99%, blockCacheHitCachingRatio=99%, hdfsBlocksLocalityIndex=95 >From these stats I see my memstores are always <1G or <6% of heap. So here are my questions: 1) Are memstore settings above relevant in v0.94 w/ off heap memstores the default? 2) We have good locality and cache hits. I'd like to take block cache size to 50% and increase overall heap to 24G. Any concerns about this given stats above? 3) Any other memory recommendations? Thanks, Ted
