hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From lars hofhansl <la...@apache.org>
Subject Re: how to explain read/write performance change after modifying the hfile.block.cache.size?
Date Sun, 23 Nov 2014 00:16:17 GMT
Yep. Came here to say that. The block cache is 40% of the heap, so if that's 400mb with the
old default setting of 1g.
      From: "Liu, Ming (HPIT-GADSC)" <ming.liu2@hp.com>
 To: "user@hbase.apache.org" <user@hbase.apache.org> 
Cc: lars hofhansl <larsh@apache.org> 
 Sent: Saturday, November 22, 2014 3:35 PM
 Subject: RE: how to explain read/write performance change after modifying the hfile.block.cache.size?
   
Thank you Nick,

I will increase the heap size.
The workstation is a development workstation. People use it to code and build software, and
do unit testing. And I use 'who' and 'ps' to make sure it is exclusive to me when I did the
test, but it is not accurate. You are right, the benchmark is not accepted in that env. I
just use this to 'develop' a benchmark tool, since I wrote a new YCSB driver for your system
on HBase, but the following test is using native HBase driver and test on Hbase. For the real
benchmark, we will run it on a separated cluster later. But I hope the data at least make
sensor even on a shared env.
The heap configuration is something I really need to check , thank you.

Best Regards,
Ming

-----Original Message-----
From: Nick Dimiduk [mailto:ndimiduk@gmail.com] 
Sent: Saturday, November 22, 2014 5:57 AM
To: user@hbase.apache.org
Cc: lars hofhansl
Subject: Re: how to explain read/write performance change after modifying the hfile.block.cache.size?

400mb blockcache? Ouch. What's your hbase-env.sh? Have you configured a heap size? My guess
is you're using the un configured default of 1G. Should be at least 8G, and maybe more like
30G with this kind of host.

How many users are sharing it and with what kinds of tasks? If there's no IO isolation between
processes, I suspect your benchmarks will be worthless on this shared environment.

-n

On Friday, November 21, 2014, Liu, Ming (HPIT-GADSC) <ming.liu2@hp.com>
wrote:

> Thank you Lars,
>
> There must be something wrong with my testing yesterday. I cannot 
> reproduce the issue anymore. Now, changing the cache.size from 0 to 
> 0.4 will not slow down the write perf dramatically, but still will 
> slow down write (Yesterday I saw a 7x slowdown, today it is about 1.3x 
> slowdown which is acceptable for me). As Ted pointed out, it is 
> possible that memstore cannot get enough memory when more RAM give to 
> block cache so it flush more frequently, but I really need more 
> reading to understand when memstore will flush.
> And one thing I noticed is when I restart hbase for the very first 
> test, the performance is best, then the second time, it is slower, 
> both read and write, and slower and slower in the following tests and 
> get to a stable point after about 3 or 4 times, in each run I will 
> read 5,000,000 rows and update 5,000,000 rows. There are too many 
> factors affect the read/write OPS in hbase...
>
> My purpose is to find a proper way to evaluate performance, since we 
> are going to change something in hbase and it is good to have a base 
> benchmark so we can compare the performance after change. So I must 
> make sure the perf test itself make sense and should be trusted.
>
> I saw an entry in the log may help to see the cache settings in my system:
> hfile.LruBlockCache: Total=373.54 MB, free=13.13 MB, max=386.68 MB, 
> blocks=5655, accesses=17939947, hits=14065015, hitRatio=78.40%, , 
> cachingAccesses=17934857, cachingHits=14064420, 
> cachingHitsRatio=78.42%, evictions=15646, evicted=3861343, 
> evictedPerRun=246.7942657470703
>
> My testing environment is a workstation with 12 core CPU, 96G memory 
> and
> 1.7 T disk. But it is a shared workstation, many users share it and I 
> started hbase in standalone mode with hbase-site.xml as below :
>
> <configuration>
>  <property>
>    <name>hbase.rootdir</name>
>    <value>hdfs://localhost:24400/hbase</value>
>  </property>
>  <property>
>    <name>hbase.zookeeper.property.dataDir</name>
>    <value>hdfs://localhost:24400/zookeeper</value>
>  </property>
>  <property>
>    <name>hbase.master.port</name>
>    <value>24560</value>
>  </property>
>  <property>
>    <name>hbase.master.info.port</name>
>    <value>24561</value>
>  </property>
>  <property>
>    <name>hbase.regionserver.port</name>
>    <value>24562</value>
>  </property>
>  <property>
>    <name>hbase.regionserver.info.port</name>
>    <value>24563</value>
>  </property>
>  <property>
>    <name>hbase.zookeeper.peerport</name>
>    <value>24567</value>
>  </property>
>  <property>
>    <name>hbase.zookeeper.leaderport</name>
>    <value>24568</value>
>  </property>
>  <property>
>    <name>hbase.zookeeper.property.clientPort</name>
>    <value>24570</value>
>  </property>
>  <property>
>    <name>hbase.rest.port</name>
>    <value>24571</value>
>  </property>
>  <property>
>    <name>hbase.client.scanner.caching</name>
>        <value>100</value>
>  </property>
>  <property>
>    <name>hbase.client.scanner.timeout.period</name>
>        <value>60000</value>
>  </property>
>  <property>
>      <name>hbase.bulkload.staging.dir</name>
>      <value>hdfs://localhost:24400/hbase-staging</value>
>  </property>
>  <property>
>    <name>hbase.snapshot.enabled</name>
>    <value>true</value>
>  </property>
>  <property>
>    <name>hbase.master.distributed.log.splitting</name>
>    <value>false</value>
>    </property>
>    <property>
>      <name>zookeeper.session.timeout</name>
>      <value>90000000</value>    :-) I just want to make sure never timeout
> here, since I get timeout so many times...
>    </property>
>    <property>
>      <name>hfile.block.cache.size</name>
>      <value>0.4</value>
>    </property>
> </configuration>
>
>
> [liuliumi@ YCSB]$ free -m
>              total      used      free    shared    buffers    cached
> Mem:        96731      46828      49903          0        984   
  32525
> -/+ buffers/cache:      13317      83413
> Swap:        48432        12      48420
>
>
> [liuliumi@ YCSB]$ lscpu
> Architecture:          x86_64
> CPU op-mode(s):        32-bit, 64-bit
> Byte Order:            Little Endian
> CPU(s):                12
> On-line CPU(s) list:  0-11
> Thread(s) per core:    1
> Core(s) per socket:    6
> Socket(s):            2
> NUMA node(s):          2
> Vendor ID:            GenuineIntel
> CPU family:            6
> Model:                44
> Stepping:              2
> CPU MHz:              3065.894
> BogoMIPS:              6130.96
> Virtualization:        VT-x
> L1d cache:            32K
> L1i cache:            32K
> L2 cache:              256K
> L3 cache:              12288K
> NUMA node0 CPU(s):    0,2,4,6,8,10
> NUMA node1 CPU(s):    1,3,5,7,9,11
>
>
> Thanks,
> Ming
>
> -----Original Message-----
> From: lars hofhansl [mailto:larsh@apache.org <javascript:;>]
> Sent: Friday, November 21, 2014 4:31 AM
> To: user@hbase.apache.org <javascript:;>
> Subject: Re: how to explain read/write performance change after 
> modifying the hfile.block.cache.size?
>
> That would explain it if memstores are flushed due to global memory 
> pressure.
>
> But cache and memstore size are (unfortunately) configured independently.
> The memstore heap portion would be 40% (by default) in either case.So 
> this is a bit curious still.
> Ming, can you tell us more details?- RAM on the boxes- heap setup for 
> the region servers- any other relevant settings on hbase-site.xml- 
> configs on the table/column family you're writing to (like bloom filters, etc).
>
> That would help us diagnose this.
>
> -- Lars
>
>      From: Ted Yu <yuzhihong@gmail.com <javascript:;>>
>  To: "user@hbase.apache.org <javascript:;>" <user@hbase.apache.org 
> <javascript:;>>
>  Sent: Thursday, November 20, 2014 9:32 AM
>  Subject: Re: how to explain read/write performance change after 
> modifying the hfile.block.cache.size?
>
> When block cache size increases from 0 to 0.4, the amount of heap 
> given to memstore decreases. This would slow down the writes.
> Please see:
> http://hbase.apache.org/book.html#store.memstore
>
> For your second question, see this thread:
>
> http://search-hadoop.com/m/DHED4TEvBy1/lars+hbase+hflush&subj=Re+Clari
> fications+on+HBase+Durability
>
> Cheers
>
> On Thu, Nov 20, 2014 at 8:05 AM, Liu, Ming (HPIT-GADSC) 
> <ming.liu2@hp.com <javascript:;>>


> wrote:
>
> > Hello, all,
> >
> > I am playing with YCSB to test HBase performance. I am using HBase
> 0.98.5.
> > I tried to adjust the hfile.block.cache.size to see the difference, 
> > when I set hfile.block.cache.size to 0, read performance is very 
> > bad, but write performance is very very very good....; when I set 
> > hfile.block.cache.size to 0.4, read is better, but write performance 
> > drop dramatically. I disable the client side writebuffer already.
> > This is hard to understand for me:
> > The HBase guide just said hfile.block.cache.size setting is about 
> > how much memory used as block cache used by StoreFile. I have no 
> > idea of how HBase works internally. Typically, it is easy to 
> > understand that increase the size of cache should help the read, but 
> > why it will harm the write operation? The write performance down 
> > from 30,000 to 4,000 for your reference, just by changing the 
> > hfile.block.cache.size from 0
> to 0.4.
> > Could anyone give me a brief explanation about this observation or 
> > give me some advices about what to study to understand what is block
> cache used for?
> >
> > Another question: HBase write will first write to WAL then to memstore.
> > Will the write to WAL go to disk directly before hbase write 
> > memstore, a sync operation or it is possible that write to WAL is 
> > still buffered somewhere when hbase put the data into the memstore?
> >
> > Reading src code may cost me months, so a kindly reply will help me 
> > a lot... ...
> > Thanks very much!
> >
> > Best Regards,
> > Ming
> >
>
>
>
>


  
Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message