
I am a Data Architect for Shutterfly Inc. in Redwood Shores, CA.
LinkedIn Profile-
Github activity
2010-09-07 20:31:11: kgorman created gist: 5690412010-09-07 20:18:04: kgorman created gist: 5690182010-08-28 01:48:18: kgorman updated gist: 5545412010-08-28 01:48:05: kgorman updated gist: 5545412010-08-28 01:46:34: kgorman updated gist: 554541 Categories
- Data Architecture (5)
- Database Engineering (12)
- Mongodb (12)
- MySQL (1)
- Oracle (5)
- PostgreSQL (18)
- Python (7)
- Random (6)
Recent Comments
- jametong on Data clustering in MongoDB using embedded docs
- Data clustering in MongoDB using embedded docskennygorman.com | kennygorman.com on Cluster data, save cash
- kgorman on WordPress 3.0
- Disk I/O: PCI Based SSDs « makeitfaster on Fusion-io SSD
- Log Buffer #189, A Carnival of the Vanities for DBAs « So Many Oracle Manuals, So Little Time on Wayback Machine: snapshots still valid technique
Tags
acid architecture auto-increment backups cap theorem clustering database Database Engineering direct i/o durability fusion-io gnuplot hp Mongodb mongosf mongostat MySQL nosql open source Oracle performance pgstat pg_reorg PostgreSQL postgresql Python replica sets replication rrdtool rss scalability sequence Shutterfly slony snapshots Solaris SSD statspack strace Sun tools vxfs WordpressBlogroll
Archives
Tag Archives: database
Memory sizing for PostgreSQL
I needed to make an assessments of a linux server’s memory footprint consumed by postgresql, however, calculating the exact memory footprint of a postgresql database is not as straightforward as one might think. Since unix utilities like top, and ps … Continue reading
Performance Tuning: Be Proactive
I see many articles out there talking about performance tuning using the Oracle waits interface. While I am not contending that technique; I believe it to be sound, and use this technique every day. What I am saying, is performance … Continue reading
The benefits of testing
I was recently involved on a problem for a critical production database that had slow response time for inserts. The batch job inserting was running at full tilt, and adding more processes (concurrency) was not helping. Ah, there must be … Continue reading