Comments on: The little program that could http://engineering.wayfair.com/the-little-program-that-could/?utm_source=rss&utm_medium=rss&utm_campaign=the-little-program-that-could A team of engineers building a platform that serves a zillion products to the world. Thu, 10 Jan 2013 08:53:47 +0000 hourly 1 http://wordpress.org/?v=3.4.2 By: Philip K http://engineering.wayfair.com/the-little-program-that-could/#comment-62 Philip K Fri, 04 May 2012 20:13:50 +0000 http://engineering.wayfair.com/?p=605#comment-62 Awesome work! I can imagine the transition from ASP was not an easy feat. Good thing you solved the "having trouble making connections under high traffic" problem before, well, the heaviest expected traffic time of the year. Out of curiosity, do you plan to stick with MSSQL? If so, what benefit does it provide over other alternatives for your environment? Awesome work! I can imagine the transition from ASP was not an easy feat. Good thing you solved the “having trouble making connections under high traffic” problem before, well, the heaviest expected traffic time of the year.

Out of curiosity, do you plan to stick with MSSQL? If so, what benefit does it provide over other alternatives for your environment?

]]> By: Fred Isaacs http://engineering.wayfair.com/the-little-program-that-could/#comment-61 Fred Isaacs Mon, 30 Apr 2012 20:43:47 +0000 http://engineering.wayfair.com/?p=605#comment-61 I like the way that you go at things. Once you find a bottleneck, you think of how to attack THAT and leave the parts that are working alone. You rethink the bottlenecked part and come up with solutions for it which are not just brute force. That's the sort of thing that I like to do. And your solution was tested for robustness with the captured transaction logic before it was put in place. There will always be problems in any implementation. The trick is to hunt them down and solve them with minimal side effects. I like the way that you go at things. Once you find a bottleneck, you think of how to attack THAT and leave the parts that are working alone. You rethink the bottlenecked part and come up with solutions for it which are not just brute force. That’s the sort of thing that I like to do. And your solution was tested for robustness with the captured transaction logic before it was put in place. There will always be problems in any implementation. The trick is to hunt them down and solve them with minimal side effects.

]]>