Time to Consider a new WordPress Host

by Brian Ritchie


Last Updated: Mar 22, 2013 13:23
620 words · 3 minutes read

I have been an avid customer of MediaTemple.net for a long time. I am currently on their cheapest Grid-Server service and while this was fun may years ago when I signed up with them, having leveled up my skills I can’t see myself continuing to host with them much longer.

MediaTemple.net is a great solution for the WordPress newbie (though I’d probably use ASmallOrange now) but my primary issue with them is their intermittent downtime that I constantly face. This blog by no means is a high-traffic blog visited by millions daily but at the same time, its something extremely personal. Its where I occasionally write my thoughts down – although I’ll admit I’ve been a tad bit lazy recently and have been drowned with deeper philosophies of life. MediaTemple.net ‘s frequent downtime, sometimes for a couple of hours or more (based on Pingdom reports) doesn’t seem to be the best solution for anything long term.

All in all, I am looking to shift hosting soon and decided its time to embrace my inner Sysadmin. Here’s the brief requirements I have come up with for myself –

Requirements :

  • Version Control of Filesystem to prevent security issues

This is a very simple way of saying I’d love to use Git to maintain my blog. WordPress as everyone knows tends to have a ton of vulnerabilities when badly written plugins are deployed so having a Version Control powering the backed might be nice to do damage control if and when needed. Also, version control definitely helps with problems that tend to rise “between the keyboard and chair”.

  • Allow frontend user to still interact via media/plugin upload/update

Now I added this as an option because I do host a few friends on my MediaTemple.net account for free just because I had the space and they needed to run some tests or host their own personal thoughts online. After all, that’s what Techie friends are for.

  • FTP is needed for Theme uploads and stuff

Again, this is an option for point 2 mentioned above. This is because some themes tend to be rather large and uploading via the WordPress upload panel would be impossible.

  • Blazing fast caching

Speed. Faster. Must get faster and better. This is pretty self-explanatory.

  • Possibly modular enough for scaling later – DB vs Filesystem vs CDN

OK, this was something that cropped up based on experience. WordPress by default has a huge database bottleneck when dealing with a sudden surge of traffic and growth of traffic over time in general. Its not the smartest kid in the block when it comes to managing this and writing scripts to manage each component modularly as and when I need would be kinda epic.

  • Creating new empty WordPress Sites as and when needed (bonus)

This is not really a requirement right now but again, its a test of skill. If I can do it, why the hell not 😉

Now some of you are wondering why WordPress in this day and age – when we have awesome static site generators like Jekyll and OctoPress or their Python equivalent like Pelican ?

Simple. While I am a Techie, I also have had some experience running digital marketing projects as well from the ground up. WordPress currently still dominates this scene due to the numerous social and SEO features that it comes prebuilt in with or vastly available through its humongous plugin repository.

I am considering open-sourcing the scripts when I get it done so if you are interested, definitely reply with your preference – Bash OR Chef OR Puppet OR Capistrano.

I love a good challenge so why not. I’ve written in all four before so it shouldn’t take me too much trouble.