The 60% Allocation “Rule.” Oracle TPC-H Proves Hard Drives Are Still Round!

I recently blogged about the phenomenal Oracle10g TPC-H result with HP’s Itanium2 based Superdome. I just took another look at the Full Disclosure Report to see what percentage of the gross disk capacity was used for Oracle tablespaces. When allocating space from each spindle, it is always good practice to use no more than about the outer most 60% of the platters. The sectors on the outside of each platter have higher capacity than the sectors closer to the center. I know not all storage arrays allow administrators to choose the disk geometry that derives a LUN, but if it is supported, it is good practice.

The 60% Rule Lives On
This audacious TPC-H result used 3072 36GB hard drives. Yes, folks, unfortunately for databases like Oracle more small drives are better—yet most of today’s storage arrays are shipping maximum capacity with minimum spindles. Yikes! Anyway, a storage configuration with 3072 36GB drives yields a gross capacity of 108TB. As I discussed in my first post about this TPC-H results, the ASM diskgroup consisted of 256 “disks” which were actually 138GB LUNs—an ASM diskgroup of 34.5TB. Since ASM was used with external redundancy, it is safe to presume that the LUNs were mirrored so the 108TB gross yields a RAID net of 54TB. The ASM space, therefore, consumed 63.8% of the drives’ gross capacity.

Some Things Never Change
Good fundamental principles such as preferring the outer portions of those round, brown spinning things generally stand the test of time. The storage subsystem configured for this TPC-H result prices out at nearly USD $3 million and yet the same fundamental storage allocation rules are still followed.

Audited TPC results are a wealth of information. I sure used to loathe doing them though!

6 Responses to “The 60% Allocation “Rule.” Oracle TPC-H Proves Hard Drives Are Still Round!”

  1. 1 Noons December 18, 2006 at 12:14 am

    Heh! Good to know some things will never change!
    Good old “use only outer disk cylinders” is as good nowadays as it was 30 years ago.

    3072 disk drives?… Man, my head is still spinning!

  2. 2 kevinclosson December 18, 2006 at 12:35 am

    yep, 3072… ok, here’s one for you…try to picture physically identifying drive number 1042 if it crashes 🙂 Even back in the 1000 drive Sequent days we always wished there was a red light on top of the cabinet that let the operations staff know there was a dead drive in a particular cab… hmmm …

    Thanks for stopping by Noons.

  3. 3 Chris M Evans December 20, 2006 at 11:10 pm

    Unfortunately on enterprise class arrays, this type of configuration simply isn’t possible. 3072 drives on 16 drives per shelf, 12 shelves per rack means reserving 16 racks for this allocation of 108TB. Using 500GB SATA drives, this could be stored in 1 rack of 216 drives. HDDs typically use around 12w of power each. Using 216 drives means saving at least 34Kw of power and the associated cooling.

    SATA is not the solution for high performance systems however an enterprise class array should be able to cache a significant amount of I/Os and mask the need to have to place data at specific points on spindles.

  4. 4 kevinclosson December 20, 2006 at 11:41 pm

    Thanks for stopping by, Chris, but I’m not sure I get your point. Are you saying that HP did not in fact use 3072 disk drives for this benchmark?

  5. 5 Noons December 21, 2006 at 1:54 am


    I’ve got a feeling the folks involved in getting these amazing numbers off the tpc-c benchmarks will not settle for “should be able to cache a significant amount of I/Os”:

    they simply can’t afford “should”.

    That’s why they use these incredible large numbers of disk drives and only the outer disk surfaces: no cache can ensure 100% hit rate or the sustained transfer rates of these configurations.

    That’s also why you won’t find enterprise class arrays being widely used in tpc-c: if they could sustain the high performance, they would have certainly been selected, these folks are after the absolute tops!

    As for the 3072 drives, I can see your point about total power requirements. But don’t forget this type of problem is usually not an issue in benchmark centres.

    Hey: no one said tpc-c demonstrates practical systems! But when someone demands the best, they are usually a good source of hints as to what works fast and why.

    Refer also to Kevin’s earlier notes on unstructured versus structued data storage as to why the two device architecture aproaches will never meet at the table of top performance.

  1. 1 Go-faster stripes « Oracle Scratchpad Trackback on February 5, 2007 at 11:01 am

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.


I work for Amazon Web Services. The opinions I share in this blog are my own. I'm *not* communicating as a spokesperson for Amazon. In other words, I work at Amazon, but this is my own opinion.

Enter your email address to follow this blog and receive notifications of new posts by email.

Join 746 other subscribers
Oracle ACE Program Status

Click It

website metrics

Fond Memories


All content is © Kevin Closson and "Kevin Closson's Blog: Platforms, Databases, and Storage", 2006-2015. Unauthorized use and/or duplication of this material without express and written permission from this blog’s author and/or owner is strictly prohibited. Excerpts and links may be used, provided that full and clear credit is given to Kevin Closson and Kevin Closson's Blog: Platforms, Databases, and Storage with appropriate and specific direction to the original content.

%d bloggers like this: