Boston Linux & Unix (BLU) Home | Calendar | Mail Lists | List Archives | Desktop SIG | Hardware Hacking SIG
Wiki | Flickr | PicasaWeb | Video | Maps & Directions | Installfests | Keysignings
Linux Cafe | Meeting Notes | Blog | Linux Links | Bling | About BLU

BLU Discuss list archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

PostgreSQL vs MySQL Throughdown! Teams?

 It is beginning to sound like people have a few things to say about the 
matter. There are lots of benchmarks out there and no one knows what to 
trust, so here's my proposal. 

We have a MySQL team and a PostgreSQL team. The two teams will develop a 
single benchmark that will function exactly the same on both systems. 
Something with a lot of concurrency and a balance between complex and 
simple queries. 

Once we get the benchmark written, the two teams will optimize their query 
set and database, and we'll run the benchmark with one process, then two,. 
then four, then eight, up to about 128 processes. We can publish all the 
code and results on the BLU web site. 

I think this can be done pretty easily, I have a lot of infrastructure 
already in place to build the benchmark and I my development machine has a 
good/fast SATA drive, dual core 64 AMD with 4G RAM. I have a few other 32 
bit machines that can act as clients. 

If anyone has a larger infrastructure and is interesting in participating, 
that would be awesome as well. 

Wouldn't we all like to have this question answered in a way we can all 
agree upon? Anyone interested? 

Here's my general thoughts: 

Use the TIGER U.S. database to find the location of an address. If it is 
found, a table of locations will be updated to include the address. 

Query the locations table to find all the addresses with 25 mile radius of 
the address. 

Query the locations table to find all the addresses with 50 mile radius of 
the address. 

The source data for the client will not be 100% random, most will be 
generated from the TIGER database itself so we have a large percentage of 
controlled input. 

About 10% of the addresses will be randomly generated addresses. 

Anyone have any comments/additions? 

This message has been scanned for viruses and 
dangerous content by MailScanner, and is 
believed to be clean. 

Discuss mailing list 
[hidden email]

BLU is a member of BostonUserGroups
BLU is a member of BostonUserGroups
We also thank MIT for the use of their facilities.

Valid HTML 4.01! Valid CSS!

Boston Linux & Unix /