SPECjEnterprise2010 benchmark – IBM beats Oracle on performance and cost


In his revolutionary book “On the Origin of Species”, Charles Darwin described the process of evolution and proposed his theory of the “survival of the fittest” (although he did not explicitly use the word “evolution”, nor did he use the term “survival of the fittest”). My guess is that most of the financial analysts of our day have not heard of that theory since many of them maintain “buy” or “neutral” rating for Oracle. My rating is “sell, sell, sell”. Too bad I am obligated to keep some of Oracle stock as part of my Total US Market Index Fund. Can someone create a fund with no Oracle stock there? Put me in line for that one.

You see, I just can’t see why would anyone buy Oracle software at these ridiculous prices they charge. Yes, I know that list prices get discounted, but sooner or later, Oracle gets you on maintenance. And guess what? Maintenance is tied to list prices. Why not use better software that costs less? IBM is one such example.

It has been quite some time since I wrote about the SPECj battles between IBM and Oracle (see previous Article 1 and Article 2). Red Hat JBoss still has no results published… Today I would like to discuss the rare case of an “apples to apples” comparison between IBM and Oracle on almost identical hardware. It is not often that we get to see results published by different vendors on the identical processor types on servers with very similar configurations. Such rare comparison point became possible thanks to IBM publishing a result last year that compares very well to the earlier Oracle publish. Yeah, I know I am three months late with this article… But better be late, than never.

Here is the side by side comparison of results:

specj_2012_table

Here are some observations based on the table above:

  1. IBM and Oracle used the same number of identical models of Intel processors for app server and for database machines (16 cores for each machine).
  2. IBM performance per core was 14.3% faster than Oracle result (since Oracle result was published earlier in the year, we can cut them some slack on this one).
  3. Despite being faster, IBM cost was almost twice as low as Oracle result for the non-clustered configuration (comparing WebLogic Standard vs. WebSphere App Server).
  4. IBM cost was almost two and a half times lower than Oracle’s for the clustered configuration (comparing WebLogic Enterprise to WAS ND).

These are some impressive numbers. I find it interesting that Oracle keeps telling that their Exalogic hardware runs WebLogic several times faster than any other equipment does, but provides no proof of these claims.

What does my common sense tell me about these findings?

  1. Well, for one thing I am not surprised that the difference in performance is so small. After all, both IBM and BEA (opps, I should say Oracle) have been around for a long time in Java-years (similar to dog-years). There are many companies running mission critical applications on WebSphere and WebLogic. Those are rock solid application platforms and the rivalry over years have boosted performance of both products (does Red Hat pay any attention?).
  2. Secondly, the dramatic difference in cost is even less surprising as IBM hardware historically beaten Sun (opps, Oracle I should say) hardware. To be more specific, Oracle hardware cost is 91.27% higher than IBM’s in this benchmark.
  3. Thirdly, the cost gap is widened even further by the lower license cost of the IBM software compared to Oracle’s. This is true for both WebSphere and DB2 as compared to Oracle WebLogic and Oracle DBMS.

Perhaps this explains why companies are moving away from Oracle to IBM to save money. Here are some examples – D+H story, TBC Corporation, Huntington Bank, Bauer Media Group, and others.

Do you want to be next? Migrate your software and hardware from Oracle to IBM to save!

PS. If you are interested in comparing the cost of IBM and Oracle middleware, you may want to review these articles:



Categories: Migration

Tags: , , , , , ,

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s

%d bloggers like this: