On Tuesday, June 4, 2013, Rahul Ravindran wrote: > Hi, > > We are relatively new to Hbase, and we are hitting a roadblock on our scan > performance. I searched through the email archives and applied a bunch of > the recommendations there, but they did not improve much. So, I am hoping I > am missing something which you could guide me towards. Thanks in advance. > > We are currently writing data and reading in an almost continuous mode > (stream of data written into an HBase table and then we run a time-based MR > on top of this Table). We currently were backed up and about 1.5 TB of data > was loaded into the table and we began performing time-based scan MRs in 10 > minute time intervals(startTime and endTime interval is 10 minutes). Most > of the 10 minute interval had about 100 GB of data to process. > > Our workflow was to primarily eliminate duplicates from this table. We > have maxVersions = 5 for the table. We use TableInputFormat to perform the > time-based scan to ensure data locality. In the mapper, we check if there > exists a previous version of the row in a time period earlier to the > timestamp of the input row. If not, we emit that row.
If I understand correctly, for a rowkey R, column family F, column qualifier C, if you have two values with time stamp 13:00 and 13:02, you want to remove the value associated with 13:02. The best way to do this is to write a simple RegionObserver Coprocessor, which hooks to the compaction process (preCompact for instance). In there simply, for any given R, F, C only emit the earliest timestamp value (the last, since timestamp is ordered descending), and that's it. It's a very effective way, since you are "riding" on top of an existing process which reads the values either way, so you are not paying the price of reading it again your MR job. Also, in between major compactions, you can also implement the preScan hook in the region observer, so you'll pick up only the earliest timestamp value, thus achieving the same result for your client, although you haven't removed those values yet. I've implemented this for counters delayed aggregations, and it works great in production. > We looked at https://issues.apache.org/jira/browse/HBASE-4683 and hence > turned off block cache for this table with the expectation that the block > index and bloom filter will be cached in the block cache. We expect > duplicates to be rare and hence hope for most of these checks to be > fulfilled by the bloom filter. Unfortunately, we notice very slow > performance on account of being disk bound. Looking at jstack, we notice > that most of the time, we appear to be hitting disk for the block index. We > performed a major compaction and retried and performance improved some, but > not by much. We are processing data at about 2 MB per second. > > We are using CDH 4.2.1 HBase 0.94.2 and HDFS 2.0.0 running with 8 > datanodes/regionservers(each with 32 cores, 4x1TB disks and 60 GB RAM). > HBase is running with 30 GB Heap size, memstore values being capped at 3 GB > and flush thresholds being 0.15 and 0.2. Blockcache is at 0.5 of total heap > size(15 GB). We are using SNAPPY for our tables. > > > A couple of questions: > * Is the performance of the time-based scan bad after a major > compaction? > > * What can we do to help alleviate being disk bound? The typical > answer of adding more RAM does not seem to have helped, or we are missing > some other config > > > > Below are some of the metrics from a Regionserver webUI: > > requestsPerSecond=5895, numberOfOnlineRegions=60, numberOfStores=60, > numberOfStorefiles=209, storefileIndexSizeMB=6, rootIndexSizeKB=7131, > totalStaticIndexSizeKB=415995, totalStaticBloomSizeKB=2514675, > memstoreSizeMB=0, mbInMemoryWithoutWAL=0, numberOfPutsWithoutWAL=0, > readRequestsCount=30589690, writeRequestsCount=0, compactionQueueSize=0, > flushQueueSize=0, usedHeapMB=2688, maxHeapMB=30672, > blockCacheSizeMB=1604.86, blockCacheFreeMB=13731.24, blockCacheCount=11817, > blockCacheHitCount=27592222, blockCacheMissCount=25373411, > blockCacheEvictedCount=7112, blockCacheHitRatio=52%, > blockCacheHitCachingRatio=72%, hdfsBlocksLocalityIndex=91, > slowHLogAppendCount=0, fsReadLatencyHistogramMean=15409428.56, > fsReadLatencyHistogramCount=1559927, fsReadLatencyHistogramMedian=230609.5, > fsReadLatencyHistogram75th=280094.75, fsReadLatencyHistogram95th=9574280.4, > fsReadLatencyHistogram99th=100981301.2, > fsReadLatencyHistogram999th=511591146.03, > fsPreadLatencyHistogramMean=3895616.6, > fsPreadLatencyHistogramCount=420000, fsPreadLatencyHistogramMedian=954552, > fsPreadLatencyHistogram75th=8723662.5, > fsPreadLatencyHistogram95th=11159637.65, > fsPreadLatencyHistogram99th=37763281.57, > fsPreadLatencyHistogram999th=273192813.91, > fsWriteLatencyHistogramMean=6124343.91, > fsWriteLatencyHistogramCount=1140000, fsWriteLatencyHistogramMedian=374379, > fsWriteLatencyHistogram75th=431395.75, > fsWriteLatencyHistogram95th=576853.8, > fsWriteLatencyHistogram99th=1034159.75, > fsWriteLatencyHistogram999th=5687910.29 > > > > key size: 20 bytes > > Table description: > {NAME => 'foo', FAMILIES => [{NAME => 'f', DATA_BLOCK_ENCODING => 'NONE', > BLOOMFI true > LTER => 'ROW', REPLICATION_SCOPE => '0', COMPRESSION => 'SNAPPY', > VERSIONS => '5', TTL => ' > 2592000', MIN_VERSIONS => '0', KEEP_DELETED_CELLS => 'false', BLOCKSIZE > => '65536', ENCODE_ > ON_DISK => 'true', IN_MEMORY => 'false', BLOCKCACHE => 'false'}]}