Database

Project Update 12th February 2017

Hey hey! It’s been a month since our last update and you all deserved to know what’s going on.

The Community

Our tight keeper crew have been busy sifting through the many great contributions from the community and the database is growing very nicely in both quality and quantity.

We’ve seen a bump in registrations/site usage and people are more willing to engage compared to just a few months ago. With a little extra polish on the social features we have, we will really start to see our niche community reach it’s hands out a little further – very exciting!

The feed was originally put out there as a quick project to see what kind of other game information people were interested in. When we started the experiment, we weren’t really sure what a feed should be but after many hours of discussion and analysis of services and data we now know the direction to take. This can be reflected in the rejuvenated feed design and functionality that was introduced a week or so ago. We will continue to evolve the feed, making it more and more relevant to each type of gamer whether you are a data nerd or just looking to be entertained.

The Tech

We kicked off the year with a couple of weeks hard focus on performance and stability. Jonas initiated neo mode and installed a memory management tool called jemalloc to help with some memory issues. That and other trickery lead to some nice results, especially when it comes to dealing with our ever increasing traffic.

A tweak to the way images and video are loaded has resulted in a front-end performance increase on almost every page around the site.

Despite various odd behaviour creeping up on us these last few weeks, our continuous integration is working well and we are adding more tests every day to make sure old bugs don’t return and new bugs never make it to the live site.

We are looking for a new image hosting solution and spotted a very promising open source project being developed by @h2non – so promising in fact that we decided to back the open collective. In the future, this could not only save us money, but also give us some freedom with how we use and generate images.

We have ongoing issues with our search as the latest elasticsearch is much more memory oriented than before. We have some-what stabilised it but will continue to work on this issue over the next couple of weeks.

Oh, and check out your user settings, there is a new section called “Site Settings” where you can disable autoplaying of videos (at last!).

TL;DR

Data count is up, Traffic is up, User count and site usage is up, API consumer count is up, Coffee consumption is up. We have a bright year ahead. Let’s rock n’ roll.

 

You may also like