IDS Forum

Re: Performance Impact after storage migration

Posted By: Art Kagel
Date: Wednesday, 9 October 2013, at 6:45 a.m.

In Response To: Performance Impact after storage migration (VIKAS HIVARKAR)

Vikas:

Onstat -F will tell you whether LRU writes or Chunk writes are dominating.
With your lru_min/max_dirty settings at 1&3 I'd also guess that LRU writes
are dominating. Normally on v11.50+ with non-blocking checkpoints, most
sessions are not affected by long checkpoints, but it does happen. Before
you turned on the auto checkpointing, did the onstat -g ckp show long block
times?

You say that you "upgraded" your storage. That usually means a bigger
badder SAN. If you now have all of your chunks, including logical and
physical logs, rootdb, temp spaces, and your data, all on a single
monstrous structure on the SAN you may be experiencing IO bottlenecks
during the checkpoints and, also during periods of more intense LRU
flushing, that is affecting the engine's ability to move data into the
cache when it is needed. Similarly, how was(were) the SAN structure(s) for
the instance configured? What RAID level? How big is the stripe blocking?
How many spindles are in each structure? Is it possible that the same
sets of spindles are being shared with other applications that have
different access patterns than the database (Windows filesystems and
especially mail servers are especially bad to share with for example). How
are your read and write IO service times as the engine sees them (onstat -g
iof)?

Art

Art S. Kagel, Principal Consultant

Advanced DataTools (www.advancedatatools.com)
Blog: http://informix-myview.blogspot.com/

Disclaimer: Please keep in mind that my own opinions are my own opinions
and do not reflect on my employer, Advanced DataTools, the IIUG, nor any
other organization with which I am associated either explicitly,
implicitly, or by inference. Neither do those opinions reflect those of
other individuals affiliated with any entity with which I am affiliated nor
those of the entities themselves.

On Wed, Oct 9, 2013 at 3:47 AM, VIKAS HIVARKAR <vikas.hivarkar@gmail.com>wrote:

> Hello All,
>
> IDS 11.50.FC8W2 on HP-UX B.11.31 U ia64
>
> Recently we have upgraded our storage and the hp boxes and since then we
> see
> intermittent performance issues.
>
> onstat -g ckp
> AUTO_CKPTS=On RTO_SERVER_RESTART=180 seconds Estimated recovery time 1304
> seconds
>
> We observed that the checkpoint was taking long time and blocking the
> transactions hence we set the RTO_SERVER_RESTART to 180
> post this the AUTO_CKPTS was ON and now I see there is very very less
> number
> of checkpoint except for the checkpoints triggered
> by backup so I believe most of the writing is done by LRU flushing
>
> Is it normal not to have regular interval checkpoints ? could the excessing
> LRU flusing is causing the intermittent performance problems?
>
> onstat -c |grep ^BUFFER
> BUFFERPOOL
>
> size=2K,buffers=9000000,lrus=512,lru_min_dirty=1.000000,lru_max_dirty=3.000000
> BUFFERPOOL
>
> size=4K,buffers=3000000,lrus=512,lru_min_dirty=1.000000,lru_max_dirty=3.000000
> BUFFERPOOL
>
> size=8K,buffers=1200000,lrus=512,lru_min_dirty=1.000000,lru_max_dirty=3.000000
> BUFFERPOOL
>
> size=16K,buffers=3500000,lrus=512,lru_min_dirty=1.000000,lru_max_dirty=3.000000
>
> onstat -g seg
> Segment Summary:
> id key addr size ovhd class blkused blkfree
> 32789 52564803 c000000015380000 348160 5296 M 84 1
> 65551 52564801 c000000100000000 77832777728 912540720 R* 19002141 2
> 393236 52564802 c000001400000000 7372800000 86401888 V* 1127853 672147
> Total: - - 85205925888 - - 20130078 672150
>
> onstat -p
> Profile
> dskreads pagreads bufreads %cached dskwrits pagwrits bufwrits %cached
> 3533376905 40019324337 129263240373 97.27 119250498 619708349 2931928150
> 95.94
>
> isamtot open start read write rewrite delete commit rollbk
> 168657611197 15194194925 5119943273 78536749289 1844472464 29194093
> 56456814
> 32472838 8777
>
> gp_read gp_write gp_rewrt gp_del gp_alloc gp_free gp_curs
> 0 0 0 0 0 0 0
>
> ovlock ovuserthread ovbuff usercpu syscpu numckpts flushes
> 0 0 0 987815.69 148729.15 1685 3891180
>
> bufwaits lokwaits lockreqs deadlks dltouts ckpwaits compress seqscans
> 420005993 316802 28144601035 0 0 61682 104707403 2348872108
>
> ixda-RA idx-RA da-RA RA-pgsused lchwaits
> 1053637354 85987054 623849831 1727787827 84014387
>
>
>
> *******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>

--089e013d14cadc215f04e84c96de

Messages In This Thread

• Performance Impact after storage migration
  VIKAS HIVARKAR -- Wednesday, 9 October 2013, at 3:47 a.m.
  • Re: Performance Impact after storage migration
    Art Kagel -- Wednesday, 9 October 2013, at 6:45 a.m.
    • Re: Performance Impact after storage migration
      VIKAS HIVARKAR -- Friday, 11 October 2013, at 7:38 a.m.
      • Re: Performance Impact after storage migration
        Art Kagel -- Friday, 11 October 2013, at 10:26 a.m.
        • Re: Performance Impact after storage migration
          VIKAS HIVARKAR -- Wednesday, 16 October 2013, at 3:26 a.m.
          • Re: Performance Impact after storage migration
            Art Kagel -- Wednesday, 16 October 2013, at 7:05 a.m.
            • Re: Performance Impact after storage migration
              VIKAS HIVARKAR -- Friday, 18 October 2013, at 8:09 a.m.
              • Re: Performance Impact after storage migration
                Art Kagel -- Friday, 18 October 2013, at 8:12 a.m.

IDS Forum is maintained by Administrator with WebBBS 5.12.