cloud

Looks like I generated a little buzz when I mentioned in my SxSw talk that I'd recently done a proof of concept scaling Open Atrium to 200+ requests/second using EC2. Since a number of people have asked about it, here's the skinny:

Disclaimer

First of all, I need to point out that this was a proof of concept build designed to let me know if the use-case I was pursuing (1000s of logged in users in a "burst" scenario) was even feasible or not. I cannot guarantee these results for your business case; your mileage may vary, etc.

The Basic Setup

I've posted updated slides from my Drupal in the Cloud presntation here at SxSw interactive. Check 'em out!

Just a heads up; as we move towards more and more stable builds of the Mercury stack, we are starting to look at deploying it on other infrastructure besides EC2. This week, we set it up and tuned for a modest (512MB of ram) VPS. These tests were successful. We were able to simulate a mix of non-cached traffic along side the simple ApacheBench battering, and the system held up well, even without gigabytes of ram to support it.

I'm happy to announce the 0.6 Alpha release of the Mercury AMI, now including ApacheSolr as the search backend! This is the last piece of major infrastructure we want to integrate into the stack for scalability purposes. You can now move from a single-server install based on Mercury to a best-practice vertically scaled architecture with separate hardware to run front-end cache, application, back-end cache, search and database!

The quickest way to find it is by searching Amazon EC for "Pantheon" or "Mercury". The manifest path for the latest release (in 32bit and 64 bit flavors) is:

• chapter3-storage/PANTHEON-pressflow-mercury-alpha-6.manifest.xml
• chapter3-storage/PANTHEON-pressflow-mercury64-alpha-6.2.manifest.xml (back!)

If you'd like to "roll your own" we've updated the wiki instructions page with a new set of instructions for getting Solr up and running as part of the process. Feel free to improve that documentation, as it's definitely a community process.

This will likely be one of the last releases before we move the project into the Beta phase, at which point we'll be focusing on fine tuning and stability as well as portabilty onto non EC2 systems moreso than new features. If you have ideas for additional things you'd like to see integrated in the stack, please chime in. We're also going to be documenting real-world "how to" use-cases — e.g. "how do I put my existing site on Mercury" in user-friendly detail — so stay tuned for that.

As always, let us know what you think of the release, what you'd like to see in future iterations, and how your experience is in using the stack. There's plenty more to come.

Hello,

I wanted to announce a new module, just released today: Node Cloud. If you're familiar with tag clouds this module should look pretty familiar: it themes the output of a view like a tag cloud.

The primary ordering of the view is from the first sort. The sizing of each item is based on the second sort order. This makes the plugin very versatile for making clouds of popular content, highly rated content (say with the voting api), highly commented content, etc.