Short Term Pantheon Roadmap

I felt like it might be a good idea to post some news about upcoming PANTHEON development for those of you who are curious where the project is headed next. We've been a bit quiet over the past few weeks as we prepare for the commercial release of our Mercury on-demand product (live by DrupalCon or bust!) and evaluate the next steps for the platform. This post should give you some insight into what we're tracking and where we are headed.

Your feedback on any of these points is humbly requested.

Short Term

In the short term, we're getting set to release our on-demand service, as well as tracking a few developments that could result in a 1.1 iteration of Mercury with a few improvements:

• (Update: thanks for reminding me of this in the comments) More Formats: Now that Mercury is stable, we'll be looking to release for more free clouds as we can — Amazon ESB, Linode Stackscript, TurnkeyLinux, VMWare, etc.
• Ubuntu 10.04, aka "Lucid Lynx": Lucid is designated a Long Term Support (LTS) release, it will be maintained by the Ubuntu project for five years. We believe that kind of backing is important for people looking to make a platform switch to Mercury and to Drupal.
• Varnish 2.1: which may not make it into Lucid, but since we base a lot of the value around Mercury on Varnish, we feel it's in our interest to stay up to date with the latest and greatest.
• Monitoring: We're evaluating a lot of possible answers here, from a homebrewed Munin solution to integration with third-parties (like Server Density) that can provide some cool value-adds like iPhone apps, alerts and more.
• Easy Site Import: A lot of people understandably want to take an existing Drupal site and put it on Mercury. We aim to make that easier through documentation and an automated import process in the on-demand service.
• Multisite: One of our most requested features, this is unfortunately pretty complex. However, we want to document and tweak the system so those who are comfortable configuring Drupal to run a multisite under normal circumstances can do the same with confidence while using Mercury.

Medium Term

In the medium term, we want to start looking at taking on more of the Development/Deployment toolchain:

• Drupal 7: Since D7 contains the critical improvements to leverage Varnish, we're planning on having a version of the system pre-bundled with D7 as soon as its stable. :)
• Other Drupals: with more and more "Drupal Products" coming on line all the time, it makes perfect sense to offer these install profiles or drush_make scripts as part of the initial Mercury rollout.
• Version Control: effortless best-practice version control for Drupal, using tried and true (svn) and cutting edge (git/bzr) systems. There's no excuse for not using this now!
• Issues: if you've worked on a moderately-sized project, you know that an issue-tracker of some sort is a must. Again, nobody should be without these.
• Deployment: once you've got code in version control how do you (safely) roll it out on your production hosting? We'll make that easy too.

Long Term

The long-term vision for PANTHEON is to give people a complete end-to-end system, from your desktop to the cloud, that makes it easy to do Drupal right. This means keeping up with the latest in cutting-edge performance and functionality.

• Developer Tools: In addition to a desktop development environment, we want to provide all the project tools that we've been developing internally over the years, so it's easy to start up a new site, clone an existing one, re-sync your databases, etc.
• Testing/Profiling: We're huge fans of Continuous Integration and we think everyone should use it. Automatically catch security holes, style violations, and get critical performance profiling information. Test Drupal continuously; sleep better at night.
• NoSQL: Mongo, Cassandra and Couch. There's a lot of exciting work happening in this space, and as soon as the best practices begin to emerge, we'll be all over it.
• High-Availability Hosting: For a lot of use-cases, the "one box" architecture just won't cut it. You need to be bulletproof, and you need to be ready for enormous floods of traffic. We aim to (eventually) provide a high-performance, high-availability hosting option with integrated CDN.