wolfsREIGN

MySpace and Scalability

Over the weekend, MySpace’s data-center was hit with a power outage. On top of having the power go out, the backup generators failed, and caused the entire website to come to a halt. With a profile count somewhere near the 96 million mark, one would think that there would be some redundancy in the system, to prevent such outages to occur. Also, one would expect that time would be taken to refactor the application, to allow for fewer problems, and ultimately make a much tighter system for the phenomenal number of users MySpace currently has.

Suggestions for Improvement

1. First things first, I would take time and rewrite the application from the ground up. Start over. Go through, and make sure that the database models are as tight, and normalized as possible, and then start importing the existing data over to the new models.
2. Create a system with redundant data-centers, located throughout at least the United States, if not globally. Make sure that they are kept as synced as realistically needed. While it might be a good idea to keep all of the United States in sync all the time, with data-centers throughout the globe, data syncs can lag a little while without causing too much trouble for the international users.
3. Optimize the markup to reduce strain on the servers. While MySpace can’t control the data that users place on their pages, they can certainly improve upon the base code that is used throughout the site. Reduce the page weight by using semantic markup and css. I’ve already done this for them before. They just chose not to do it.

So there you have it… All of my ideas on what would make MySpace better… or what could be used to make a competitor to them. Hopefully they will take this most recent power outage, and the outcries of users and nonusers alike, and make some much needed changes to their system.