Data Migration Blogs
Is custom really better?
- Font size: Larger Smaller
- Hits: 1121
- 5 Comments
- Subscribe to this entry
- Bookmark
Custom. Proprietary. Exclusive. All words we associate with quality and better performance – custom suits fit better, proprietary software meets our specific needs, exclusive access grants us a social edge.
But is that always true? For vintage cars, probably. But, for an RDBMS? Not always. We put PostgreSQL 9.1 against a proprietary RDBMS and the results are in.
Using the Open Source load test tool, Hammerora, we simulated 500 users, each performing 10,000 transactions against 800 data warehouses and measured sustained throughput. The same platform was used for all tests and repeated multiple times (after reset) ensuring the repeatability of the results.
So, how did PostgresSQL do? In short, consistently better. During our testing, the average Proprietary RDBMS sample showed throughput of 53824 Transactions per Minute (TPM) while PostgreSQL samples averaged 64374 TPM, - an almost 20% higher throughput.
For the full results, (click here).
Contact us if you'd like to discuss your project in more detail, and to find out more about our database migration services and PostgreSQL offerings.
Comments
-
Thursday, 25 October 2012
Which proprietary databases did you include to find your average? It would be interesting to see the results split out against the specific databases you tested against.
-
Thursday, 25 October 2012
Unfortunately, we are not at liberty to identify the specific RDBMS we used for legal reasons.
-
Thursday, 25 October 2012
Great to see this. Please try out 9.2!
-
Thursday, 25 October 2012
We're working on it. As you may know, PostgreSQL Release 9.2 came out on September 10. You can read about it at: PostgreSQL 9.2 Released: http://www.postgresmigrations.com/about-bull-services/press-releases/pr-09-10-12
I love the way you wrote this article. This is wonderful. I do hope you intend to write more of these types of articles. Thank you for this interesting content!