The Editors of New York Oracle User Group (NYOUG) publication - TechJournal - chose my article Exadata Demystified as the Article of the Year. Here is the snippet from the Editorial:
And the Award Goes To …
With Exadata version 126.96.36.199.0 came the Unbreakable Linux Kernel for Exadata, which had been the stock EL5 redhat kernel prior to this version (2.6.18). With the unbreakable kernel came the opportunity to run the perf utility. This utility has the opportunity to see which functions are active inside an executable when there’s a symbol table. And the oracle database executable has a symbol table! One reason to do this, is to get a more granular overview of what the Oracle database is doing than the wait interface, especially to get a more detailed overview of what the database is doing in what is visible in the wait interface as ‘on cpu’.
I will be doing a lot of (Exadata) talking and teaching in the coming months. Here’s a list of events where you’ll see me speaking, teaching, hacking, learning and hopefully also drinking beer:
In order to look how Exadata smartscans are different, first let’s have a peek the Oracle full segment/multiblock read evolution as short as possible:
a) Traditional multiblock reads, visible via the event ‘db file scattered read’
The essence is: Multiple adjacent blocks are read from disk, and put in the buffercache. Because every read is sequentially processed, IO latency is a performance penalty for every physical read. This works roughly this way: get a set of adjacent blocks from the segment header, fetch these blocks from disk, process these blocks, then get the next set of adjacent blocks, fetch these blocks from disk, process these blocks, etc.
If you attended my Exadata hacking session today, you saw me using the cellver.sql script which lists some basic configuration info about the currently connected storage cells:
We recently received our 3rd Exadata machine into Enkitec’s exalab. Now we have a V2, X2 and X3 there, in addition to ODA, Big Data Appliance (which comes with a beer-holder built in!) and an Exalytics box! So you understand why Karl Arao is so excited about it :-)
This occasion demands that we hack the hell out of all this kit soon! So, let’s have another (super-secret) hacking session!
This time, let’s see how the Exadata Smart Flash Cache works! (both for reads and writes). Note that we won’t cover Smart Flash Logging in this session (otherwise we’ll end up spending half a day on it :)
This FF is a bit of a follow-up to the one I posted last week on PL/SQL skills and a comment made by Noons on how much knowledge you need to be an OakTable member.
I have a question to answer and I would appreciate other people’s opinion. Should there be more intro talks at conferences? If so, should the experts be giving them?
Thank you for attending the Exadata Snapper (ExaSnapper) hacking session!
I have split the recording of this session into 3 pieces and uploaded to enkitec.tv. The ExaSnapper v0.7 beta that I demoed is also available now at enkitec.com (registration needed). See the links below.
For quick reference, here’s the syntax of running ExaSnapper – there are two modes, one is the before/after capture (think Tom Kyte’s runstats, but for exadata metrics) and the other is more like a DBA-monitoring mode, where you can just measure a few seconds worth of a long-running query runtime and get the IO and efficiency figures from there.
Here’s an excerpt from the install script documentation section:
The promised hacking session about Exadata performance troubleshooting will happen on Thursday 21st February 9am-11am PST (2 hours). I will show a number of examples and demos where the Exadata Snapper shows what was going on in storage cells when running Smart Scans or just doing IO.
Register here (free stuff!):
I will post the scripts on the day of this session.
I was recently involved in an upgrade project to go from 188.8.131.52 to 184.108.40.206 on an Exadata V2. We hit some snags during the upgrade specifically related to OEM 12c Cloud Control. We performed an out-of-place upgrade and OEM 220.127.116.11.0 had some difficulty in dealing with this.
12c Cloud Control is supposed to run a daily check that looks for new targets on each server. When it finds something new, it places this in a queue to wait for admin approval. With a single click, you can promote the newly discovered target to an OEM managed object.