Re:Announcing the Facebook 2011 Hacker Cup
by pedalpete on Friday December 10, 2010 @12:10AM
(#1990714)
I don't have any knowledge of the inside workings of FB, but I think some of the issues you are referring to could be caused by
1) ajax errors, packets get dropped, it is going to happen
2) single server instances that somehow didn't get updated. Again, if you think of the number of servers they are running, and the update process across multiple datacenters, I'm sure it is entirely possible that one or two servers take longer to update, or something gets missed and isn't recognized until a later update.
I don't think we can call the fb code 'buggy'. Their ability to scale and manage the volume of data along with how quickly they are constantly making major updates to their entire system is mind boggling. To me anyway, but I'm sure you are a much better programmer than I am.
1) ajax errors, packets get dropped, it is going to happen
2) single server instances that somehow didn't get updated. Again, if you think of the number of servers they are running, and the update process across multiple datacenters, I'm sure it is entirely possible that one or two servers take longer to update, or something gets missed and isn't recognized until a later update.
I don't think we can call the fb code 'buggy'. Their ability to scale and manage the volume of data along with how quickly they are constantly making major updates to their entire system is mind boggling. To me anyway, but I'm sure you are a much better programmer than I am.