created 04 October 2009 in links tagged amazon, reliability and scaling.
Scaling is basically an engineering problem, which can be solved with the right architecture and algorithms. Reliability is more like security: it’s a weakest-link problem and requires constant vigilance in all areas, including engineering, testing, operations, and deployment
I’m going to bookmark this and read it every week.
created 23 September 2009 in links tagged api, flickr and scaling.
Paul speaks Truth.
..while adding more “extras” would reduce the number of API calls needed for a specific application, it doesn’t reduce the load on flickr’s back end systems. We’ve found it actually increases the load, as developers are more likely to request extra data they don’t need, and one heavy API request ties up more of our system resources than multiple lighter API requests.
Also, developer time. But that’s more obvious. What I like is the point that, yes, it’s annoying making n+1 calls, but putting that into one call is still going to end up as n+1 database calls at some point. Joins are Bad. Keep them explicit.
created 28 March 2006 in links tagged rails and scaling.
http://poocs.net/articles/2006/03/13/the-adventures-of-sc...
