« How did Google Instant become Faster with 5-7X More Results Pages? | Main | 4 General Core Scalability Patterns »

6 Scalability Lessons

Jesper Söderlund not only put together a few interesting scalability patterns, he also came up with a few interesting scalability lessons:

  • Lesson #1. Put Smarty compile and template caches on an active-active DRBD cluster with high load and your servers will DIE!
  • Lesson #2. Don't use out-of-the-box configurations.
  • Lesson #3. Single points of contention will eventually become a bottleneck.
  • Lesson #4. Plan in advance. 
  • Lesson #5. Offload your databases as much as possible.
  • Lesson #6. File systems matter and can run out of space / inodes.

For more details and explanations see the original post.

Reader Comments (3)

Not sure about offload your databases. That suggests the databases aren't scaled properly so you are trying to avoid them. If you follow that pattern long enough, it gets you into a trap where your caching layer is huge and complicated and any cache miss that hits the database yields a bad experience. I think it is much better to examine why your databases suck and make them not suck.

September 9, 2010 | Unregistered CommenterGreg Linden

nice but very short!

Need to understand realtime usage of application along with application life to design Scalability patterns. above points are very valid (like archival of data (offload database)), file system.


September 20, 2010 | Unregistered Commenterrastogis

This is kind of old and the original Article has many comments correcting the wrong statements which were mentioned.

E.g. Lession #1 should be "ReiserFS on a DRBD Active-Active Cluster is a bad idea, cause its not a ClusterFS"


October 1, 2010 | Unregistered CommenterJan

PostPost a New Comment

Enter your information below to add a new comment.
Author Email (optional):
Author URL (optional):
Some HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>