Exadata v2 Smart Scan Performance Troubleshooting article

I finally finished my first Exadata performance troubleshooting article.

This explains one bug I did hit when stress testing an Exadata v2 box, which caused smart scan to go very slow – and how I troubleshooted it:

Thanks to my secret startup company I’ve been way too busy to write anything serious lately, but apparently staying up until 6am helped this time! :-) Anyway, maybe next weekend I can repeat this and write Part 2 in the Exadata troubleshooting series ;-)

Enjoy! Comments are welcome to this blog entry as I haven’t figured out a good way to enable comments in the google sites page I’m using…

This entry was posted in Oracle and tagged , , , , . Bookmark the permalink.

8 Responses to Exadata v2 Smart Scan Performance Troubleshooting article

  1. Hi, Tanel.
    Great article on Exadata issues. Interesting, where you could get chance to test it? Did you bought Exadata machine? :)

    Thanks.

  2. Hey Tanel,

    Interesting issue with the Exadata. What are your initial impressions? We’re having lot’s of fun figuring out what works well and what doesn’t, but so far haven’t seen any real glaring holes.

    And how long is this secret startup going to remain a secret? Once you start talking about a secret it can’t stay hidden for long.

    Kerry

  3. Fairlie Rego says:

    Bad performance is one thing… Wrong results ..explaining that to business people is a bit tricky..
    I have been using the V2 machine since Dec last year.
    One of the issues I hit 3 months ago was
    9388886 – WRONG RESULTS ON EHCC TABLES WITH MORE THAN 255 COLUMNS
    Unfortunately due to IP reasons I cannot talk about some of the other more interesting issues..

  4. null says:

    Tanel is very busy. And He can not answer questions. (Tanel busy wait. TBW :) )

  5. Tanel Poder says:

    @null

    Ha! The TBW just ended for a while! ela=3 days when I was really busy :-)

    @Teymur
    Nope, I’m not that rich yet, but some of my clients are ;-)

    @Kerry
    My exadata experience is the same what I’ve also heard from multiple other Exadata users – it does rock performance wise, it is fast, but the downside is that there are lots of bugs (partially due being on db version 11.2 instead of the usual 10.2 and 9i ;-) and also it seems that people have experienced more hardware failures than with normal Oracle database setups. That of course may be because in a usual SAN storage environment, the DBAs rarely hear about disk failures, but with Exadata, the hardware is closer and more visible to DBAs (which from the performance perspective is a good thing). The bugs & hardware issues aren’t again a design limitation but rather the usual issues what you’ll get with any fresh product with radically changed new features.

    @Fairlie
    Yes, that’s why I mentioned that the bug I troubleshooted in my article wasn’t the most evil one :-)
    There are quite a few wrong data bugs listed in the BP notes, which makes me think that as part of the Exadata migration validation/testing all the reports and jobs should also be run with all these new features disabled so one could compare the query outputs. There’s another article coming on that topic ;-)

  6. Tanel Poder says:

    @Kerry Osborne

    The secret’s gonna be revealed soon ;-)

  7. maclean says:

    Fresh experience! Oracle should denote more Exadata Machine to expert!

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>