Moving our data warehousing from Hive/EMR to Redshift has the potential to speed up our queries and reduce our costs. Here’s how we are implementing our product recommendations system in Redshift.
We’ve been running Amazon Redshift in production at Monetate for almost a year. Over that time we’ve learned a bit about what slows queries down and have built some useful views that allow insight into the key contributors to query performance. What follows is a quick run through of those queries.
One thing I’ve never liked about Hacker News comments is that you have to click through to a user’s profile page to learn anything about them. I thought it might be helpful to see some of that information attached to the comments themselves, so I whipped up this user script.
Monetate engineers are big fans of Hacker News. Not only is it on just about everybody’s daily-read list, but it’s also been very helpful in attracting new developers (several of our hires, including me, learned…
Codeswarm is an organic software visualization of project repositories. Below is a version I created for Monetate’s code, which starts from the first commit and goes to the point when we reached 15,000 commits. Each dot represents an individual commit, and the color code corresponds to the different sections of the codebase.