[ 
https://issues.apache.org/jira/browse/HBASE-8894?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13803910#comment-13803910
 ] 

Liang Xie commented on HBASE-8894:
----------------------------------

0) hardware or os:

System | Supermicro; X9SCD; v0123456789 (Desktop)
Service Tag | 0123456789
Platform | Linux
Release | CentOS release 6.2 (Final)
Kernel | 2.6.32-220.el6.x86_64
Architecture | CPU = 64-bit, OS = 64-bit
Threading | NPTL 2.12
SELinux | Disabled
Virtualized | No virtualization detected
# Processor ##################################################
Processors | physical = 1, cores = 4, virtual = 8, hyperthreading = yes
Speeds | 8x3292.141
Models | 8xIntel(R) Xeon(R) CPU E3-1230 V2 @ 3.30GHz
Caches | 8x8192 KB
# Memory #####################################################
Total | 31.3G
Free | 219.9M
Used | physical = 31.1G, swap allocated = 33.5G, swap used = 213.6M, virtual = 
31.3G
Buffers | 438.0M
Caches | 20.4G
Dirty | 1340 kB
UsedRSS | 9.2G
Swappiness | 60
DirtyPolicy | 20, 10
DirtyStatus | 0, 0
Locator   Size     Speed             Form Factor   Type          Type Detail
========= ======== ================= ============= ============= ===========
1333 MHz
1333 MHz
1333 MHz
1333 MHz
ChannelA-DIMM0 8192 MB  1333 MHz          DIMM          DDR3          
Synchronous
ChannelA-DIMM1 8192 MB  1333 MHz          DIMM          DDR3          
Synchronous
ChannelB-DIMM0 8192 MB  1333 MHz          DIMM          DDR3          
Synchronous
ChannelB-DIMM1 8192 MB  1333 MHz          DIMM          DDR3          
Synchronous

1) important vm options:
-Xmx2560m -XX:MaxDirectMemorySize=2560m -Xms2560m -Xmn512m -Xss256k 
-XX:MaxPermSize=128m -XX:PermSize=128m   -XX:+UseConcMarkSweepGC -verbose:gc 
-XX:+PrintGCDetails -XX:+PrintGCDateStamps -XX:SurvivorRatio=1 
-XX:+UseCMSCompactAtFullCollection -XX:CMSInitiatingOccupancyFraction=75 
-XX:+UseCMSInitiatingOccupancyOnly -XX:+CMSParallelRemarkEnabled 
-XX:TargetSurvivorRatio=80  -XX:+CMSScavengeBeforeRemark -XX:ConcGCThreads=8 
-XX:ParallelGCThreads=8 -XX:PretenureSizeThreshold=4m 
-XX:+CMSConcurrentMTEnabled -XX:+ExplicitGCInvokesConcurrent 
-XX:-UseBiasedLocking ...
I used the absolutely the same vm options during both bucket cache and l2 cache 
testing.

2) important config:
hbase.bucketcache.ioengine=offheap
hfile.l2.bucketcache.ioengine=offheap
hbase.bucketcache.size=0.7
hfile.l2.bucketcache.size=0.7
hfile.block.cache.size=0.7
hbase.regionserver.global.memstore.lowerLimit=0.1
hbase.regionserver.global.memstore.upperLimit=0.1

3) jdk version:
java -version
java version "1.6.0_37"
Java(TM) SE Runtime Environment (build 1.6.0_37-b06)
Java HotSpot(TM) 64-Bit Server VM (build 20.12-b01, mixed mode)

4) i just tested 100% read,  no write at all.

additional,  my test data size : 
dfs -du -s -h /hbase/hytst-replication/ycsb_xieliang
3.6g /hbase/hytst-replication/ycsb_xieliang
it's a YCSB table with 500w records. 
I ensure there's a read ops warmup before test running,  until almost all data 
be loaded into os cache. so in this scenario the data could not fit into hbase 
cache all, but no disk io to dominate the test result:)

> Forward port compressed l2 cache from 0.89fb
> --------------------------------------------
>
>                 Key: HBASE-8894
>                 URL: https://issues.apache.org/jira/browse/HBASE-8894
>             Project: HBase
>          Issue Type: New Feature
>            Reporter: stack
>            Assignee: Liang Xie
>            Priority: Critical
>         Attachments: HBASE-8894-0.94-v1.txt, HBASE-8894-0.94-v2.txt
>
>
> Forward port Alex's improvement on hbase-7407 from 0.89-fb branch:
> {code}
>   1 r1492797 | liyin | 2013-06-13 11:18:20 -0700 (Thu, 13 Jun 2013) | 43 lines
>   2
>   3 [master] Implements a secondary compressed cache (L2 cache)
>   4
>   5 Author: avf
>   6
>   7 Summary:
>   8 This revision implements compressed and encoded second-level cache with 
> off-heap
>   9 (and optionally on-heap) storage and a bucket-allocator based on 
> HBASE-7404.
>  10
>  11 BucketCache from HBASE-7404 is extensively modified to:
>  12
>  13 * Only handle byte arrays (i.e., no more serialization/deserialization 
> within)
>  14 * Remove persistence support for the time being
>  15 * Keep an  index of hfilename to blocks for efficient eviction on close
>  16
>  17 A new interface (L2Cache) is introduced in order to separate it from the 
> current
>  18 implementation. The L2 cache is then integrated into the classes that 
> handle
>  19 reading from and writing to HFiles to allow cache-on-write as well as
>  20 cache-on-read. Metrics for the L2 cache are integrated into 
> RegionServerMetrics
>  21 much in the same fashion as metrics for the existing (L2) BlockCache.
>  22
>  23 Additionally, CacheConfig class is re-refactored to configure the L2 
> cache,
>  24 replace multile constructors with a Builder, as well as replace static 
> methods
>  25 for instantiating the caches with abstract factories (with singleton
>  26 implementations for both the existing LruBlockCache and the newly 
> introduced
>  27 BucketCache based L2 cache)
>  28
>  29 Test Plan:
>  30 1) Additional unit tests
>  31 2) Stress test on a single devserver
>  32 3) Test on a single-node in shadow cluster
>  33 4) Test on a whole shadow cluster
>  34
>  35 Revert Plan:
>  36
>  37 Reviewers: liyintang, aaiyer, rshroff, manukranthk, adela
>  38
>  39 Reviewed By: liyintang
>  40
>  41 CC: gqchen, hbase-eng@
>  42
>  43 Differential Revision: https://phabricator.fb.com/D837264
>  44
>  45 Task ID: 2325295
>  7 ------------------------------------------------------------------------
>   6 r1492340 | liyin | 2013-06-12 11:36:03 -0700 (Wed, 12 Jun 2013) | 21 lines
> {code}



--
This message was sent by Atlassian JIRA
(v6.1#6144)

Reply via email to