As TBB readership continues to grow — we just crossed the 50,000 threshold for fans of our Facebook page — our poor old server is starting to crack at the seams. Rather than the hour or two every few months (which still works out to 99.9% uptime), we’ve had four or five separate downtime incidents over this past month alone, including about 7 hours earlier today. For those of us who want to give our readers the best possible experience, it’s incredibly frustrating, and we apologize for the inconvenience.
Our main challenge is that the TBB team is entirely volunteer-run, including administrators like me. We run TBB in our spare time because we love the LEGO building community and enjoy bringing our readers the best of LEGO creations and news from around the world. But doing something well for 11 years means gaining enough readers that it’s begun placing a strain on our infrastructure.
Back when we first spun up Brothers-Brick.com 10 years ago, we were on a virtual machine on a shared server, which we quickly outgrew. In 2008, we migrated to a dedicated physical server. Bear in mind that this was years before Infrastructure as a Service offerings like Amazon Web Services, Microsoft Azure, and Google Cloud. We’ve done our best with caching, content distribution networks, and offsite image hosting to minimize server strain, but none of that is enough today.
We now face the choice of migrating to another physical server from our provider (an easy enough endeavor) or creating a more sustainable, scalable infrastructure on an IaaS solution. If you’re involved in running high-traffic websites, there’s an obvious “right” answer. And if you have experience with running WordPress sites on AWS, we’d love to hear from you via our Facebook page (since one of the things that’s broken is email notifications from our contact form).