HBase Read Path openinx@apache.org Abstract ❏ Client Side ❏ Server Side ❏ Tuning Part-1 Client Side HBase Client ClientScanner ClientScanner cache(queue) scanner.next() RegionServer-0 RegionServer-1 (old generation) ● Less mixed GC(s) and shorter STW time. End-to-end offheap on the read-path (HBASE-11425) BucketCache StoreFileScanner Copy the Block from BucketCache(offheap) to onheap. Rpc Handler accumulate multiple results until reach max result size even if reach batch limit ○ Related issue: HBASE-21206 ● BlockSize ? Part-3 Tuning Tuning ● Read Distribution ● Locality ● Short Circuit Read

HBase Practice At Xiaomi huzheng@xiaomi.com About This Talk ● Async HBase Client ○ Why Async HBase Client ○ Implementation ○ Performance ● How do we tuning G1GC for HBase ○ CMS vs G1 ○ Tuning Tuning G1GC ○ G1GC in XiaoMi HBase Cluster Part-1 Async HBase Client Why Async HBase Client ? Request-1 Response-1 Request-2 Response-2 Request-3 Response-3 Request-4 Response-4 Request-1 66% Availability: 0% Why Async HBase Client ? ● Region Server / Master STW GC ● Slow RPC to HDFS ● Region Server Crash ● High Load ● Network Failure BTW: HBase may also suffer from fault amplification

HBase Practice At XiaoMi tianjy1990@gmail.com openinx@apache.org Part-1 Problems In Practice Problems in XiaoMi ❏ Problem 1. How to satisfy the regular demand of scanning table without affecting analysis need to scan a large number of data from hbase ❏ They are executed by mapreduce or spark, that put a heavy burden on HBase Scan snapshot directly ❏ HBase already provides this feature: TableSnapshotInputFormat TableSnapshotInputFormat (ClientSideRegionScanner) ❏ Construct regions by snapshot files ❏ Read data without any HBase RPC requests ❏ Required READ access to reference files and HFiles Snapshot ACL ❏ HDFS ACL could

HBASE-21879 Read HFile ’s Block into ByteBuffer directly. 1. Background For reducing the Java GC impact to p99/p999 RPC latency, HBase 2.x has made an offheap read and write path. The KV are allocated Case In above pictures, the p999 latency is almost the same as G1GC STW cost (~100ms). After HBASE-11425​ , almost all memory allocations should be in the offheap, there should be rarely heap allocation As the basic idea part said, the first thing is to design a global ByteBuffAllocator. In ​HBASE-11425​ , we have introduced an offheap memory management policy as following: 1. Set a max memory

wikipedia.org/wiki/B-tree How to Store Petabytes++ ? Likely need “No SQL” databases HBase, Cassandra, MongoDB, many more HBase covered in Hadoop/Spark modules later this semester 17

● Middleware & Proxy ● NewSQL 1970s 2010 2015 Present MySQL PostgreSQL Oracle DB2 ... Redis HBase Cassandra MongoDB ... Google Spanner Google F1 TiDB RDBMS NoSQL NewSQL Architecture TiKV TiKV

following configuration and tune it to fit your environment. [desktop] app_blacklist=zookeeper,hbase,impala,search,sqoop,security use_new_editor=true [[interpreters]] [[[sparksql]]] name=Spark SQL other third-party libraries, such as Hudi, Iceberg, Delta Lake, Kudu, Apache Paimon (Incubating), HBase，Cassandra, etc. We also provide sample data sources like TDC-DS, TPC-H for testing and benchmarking value> fs.azure.block.blob.with.compaction.dir /hbase/WALs,/tmp/myblobfiles fs.azure org.apache

following configuration and tune it to fit your environment. [desktop] app_blacklist=zookeeper,hbase,impala,search,sqoop,security use_new_editor=true [[interpreters]] [[[sparksql]]] name=Spark SQL integrate with other third-party libraries, such as Hudi, Iceberg, Delta Lake, Kudu, Flink Table Store, HBase，Cassandra, etc. We also provide sample data sources like TDC-DS, TPC-H for testing and benchmarking value> fs.azure.block.blob.with.compaction.dir /hbase/WALs,/tmp/myblobfiles fs.azure org.apache

following configuration and tune it to fit your environment. [desktop] app_blacklist=zookeeper,hbase,impala,search,sqoop,security use_new_editor=true [[interpreters]] [[[sparksql]]] name=Spark SQL integrate with other third-party libraries, such as Hudi, Iceberg, Delta Lake, Kudu, Flink Table Store, HBase，Cassandra, etc. We also provide sample data sources like TDC-DS, TPC-H for testing and benchmarking value> fs.azure.block.blob.with.compaction.dir /hbase/WALs,/tmp/myblobfiles fs.azure org.apache

following configuration and tune it to fit your environment. [desktop] app_blacklist=zookeeper,hbase,impala,search,sqoop,security use_new_editor=true [[interpreters]] [[[sparksql]]] name=Spark SQL integrate with other third-party libraries, such as Hudi, Iceberg, Delta Lake, Kudu, Flink Table Store, HBase，Cassandra, etc. We also provide sample data sources like TDC-DS, TPC-H for testing and benchmarking value> fs.azure.block.blob.with.compaction.dir /hbase/WALs,/tmp/myblobfiles fs.azure org.apache