Ok, so I have some problems with put.
randomWrite test shows 28K/sec and 4ms response, 99th percentile
Mine shows 30ms 99th percentile.
I'm doing just htable.put, where Put object is less than 1KB
2015-08-12 9:37 GMT+02:00 Serega Sheypak :
> I agree.
> 99% <= 112.09 milliseconds
> I could make
I agree.
99% <= 112.09 milliseconds
I could make 3 gets during 112 MS.
2015-08-12 9:24 GMT+02:00 Vladimir Rodionov :
> OK, this is actually checkAndPut -> get - check -put. Latency is dominated
> by get operation. Unless you have SSDs 10-40 ms mean read latency is
> normal.
>
> -Vlad
>
> On Tu
OK, this is actually checkAndPut -> get - check -put. Latency is dominated
by get operation. Unless you have SSDs 10-40 ms mean read latency is normal.
-Vlad
On Tue, Aug 11, 2015 at 11:24 PM, Serega Sheypak
wrote:
> Hi, here is it:
> https://gist.github.com/seregasheypak/00ef1a44e6293d13e56e
>
Hi, here is it:
https://gist.github.com/seregasheypak/00ef1a44e6293d13e56e
2015-08-12 4:25 GMT+02:00 Vladimir Rodionov :
> Can you post code snippet? Pastbin link is fine.
>
> -Vlad
>
> On Tue, Aug 11, 2015 at 4:03 PM, Serega Sheypak
> wrote:
>
> > Probably I found something. Response time decre
Can you post code snippet? Pastbin link is fine.
-Vlad
On Tue, Aug 11, 2015 at 4:03 PM, Serega Sheypak
wrote:
> Probably I found something. Response time decreases when parallelism grows.
> What I did:
>
> 1. wrap business logic controller into java main class. My controller does
> some logic a
Probably I found something. Response time decreases when parallelism grows.
What I did:
1. wrap business logic controller into java main class. My controller does
some logic and puts/gets to hbase with checkAndPut (sometimes)
2. create HConnection
3. pass HConnection to controller
4. wrap controll
>How about GC activity? ApplicationStopTime? Do you track that?
yes, jviusalm says it's ok, newrelic also doesn't show something strange.
HBase also says it's OK.
Profiler says most time thread is waiting for response from hbase side. My
assumption is:
1. I have weird bug in HBase configuration
2.
How about GC activity? ApplicationStopTime? Do you track that?
Is the issue reproducible? or you got it first time?
Start with RS logs and try to find anything suspicious in a period of a
very high latency. 1.5 sec HBase write latency does not look right.
-Vlad
On Tue, Aug 11, 2015 at 2:08 PM,
Hi Vladimir!
Here are graphs. Servlet (3 tomcats on 3 different hosts write to HBase)
http://www.bigdatapath.com/wp-content/uploads/2015/08/01_apps1.png
See how response time jump. I can't explain it. Write load is really-really
low.
all RS have even load. I see request-metrics in HBase master we
Monitor GC events (application stop time). Your RS may have nonoptimal
hotspot GC settings. Search Internet on how to tune GC large heaps.
-Vlad
On Tue, Aug 11, 2015 at 9:54 AM, Vladimir Rodionov
wrote:
> *Common questions:*
>
>
>1. How large is your single write?
>2. Do you see any Reg
*Common questions:*
1. How large is your single write?
2. Do you see any RegionTooBusyException in a client log files
3. How large is your table ( # of regions, # of column families)
4. RS memory related config: Max heap
5. memstore size (if not default - 0.4)
Memstore flush
hba
Hi, we are using version 1.0.0+cdh5.4.4+160
We have heavy write load, ~ 10K per econd
We have 10 nodes 7 disks each. I read some perf notes, they state that
HBase can handle 1K per second writes per node without any problems.
I see some spikes on "writers". Write operation timing "jumps" from 40-
12 matches
Mail list logo