Welcome to your super fast, highly scalable, fully managed and monitored Drupal hosting environment, with built-in database cluster configuration, and easy to use control panel. Now available in five locations and three options: Lite, Pro and Ice. Most importantly, it just works! You don't need your own sysadmin to manage and secure the servers for you. Just manage and update your Drupal sites. We take care about everything else. You can read more about many included by default features further below.

You should seriously consider this option if you host a few dozens, or maybe hundreds of Drupal sites, but only some of them are very active and as a result generate most of the traffic and system load. How to pay only for what you really use in such a scenario?

Meet Cluster Pro! It comes with new feature which adds a concept of RAM SQL Working Set, or in human language, it allows you to basically ignore the size of your databases on disk, and care only about the size of databases actually loaded to server memory, because only the visited sites have a good reason to load their databases from disk to memory, to generate the web pages to display.

With Cluster Pro we have made the databases limits much more flexible, so it's now possible to have even 150 GB of databases on disk in total, while up to 10 GB of memory is used by the most active databases at any given moment, on an average. And there is still a giant extra memory and disk space ready for scaling your system resources for load and traffic spikes, on the fly. And that is Cluster Pro!

The center of the cluster -- its Core -- holds all your files and runs your web server equipped with super fast web cache. We call it a MicroCache, because by default it stores the pages cached for anonymous visitors and logged in users only for a few seconds -- just enough to speed up things considerably, while keeping everything fresh in the browser.

The Core runs also a ProxySQL server, which transparently manages load balanced connections between your web server and all connected database nodes.

The default cluster configuration comes with three database nodes, shown above as head and two hands. These nodes talk also to each other directly, and sync data automatically, as members of their private Galera Cluster DB Ring. The cluster can be turned into real monster by adding more DB heads and hands!

Thanks to the ProxySQL server running on the Cluster Core, your Drupal sites don't need to figure out how to connect to all these database nodes in a Galera Cluster DB Ring. It's important, because no matter how many database nodes are connected to the cluster, Drupal will still see them all as a single database server, while ProxySQL will transparently handle and optimize all connections and send/receive data between your sites and their databases.

There is something special about this configuration -- all database nodes are equal! There is no single 'master' or 'primary' database node, and the system can survive as long as there is at least one database node available online. This helps to reduce, or in fact eliminate any downtime normally associated with database servers upgrades. Your visitors and clients will love you for keeping them always online!

Now let's talk about Cluster Legs. While database servers are much faster and more reliable when running in a cluster configuration, they are still not fast enough for reading and writing temporary data. That is why all your Drupal sites' cache tables are outsourced and all their caching needs are handled by the crazy fast Redis Memory Cache Server leg. This helps tremendously to speed up everything, especially for logged in sites users.

Finally, the second Leg runs a Solr Search Server, which opens a wide array of advanced search based content presentation methods to use, not just the flexible search box for your sites. It is a free of charge add-on for all sites you host, and you can easily add Solr instances to any hosted sites you wish.

Now, how about a little bit of Ice for the summer time? Would you like to have an absolute peace of mind, like, when you are on vacation with friends or family, and that feeling of true relax, because we can easily restart, or even recreate your cluster, if needed, on another machine? And all that without paying exorbitant price for complicated, slow and hard to properly manage classic High Availability system with duplicate web load balancers, web nodes, and very slow network filesystem? Meet Cluster Ice!

The Cluster Ice maintains a fresh, frozen and constantly updated copy of all your files and entire system running Cluster Core and its both Legs, ready to start and connect again to both Hands and the Head, all running together as your perfect database cluster on separate machines.

Note that Cluster Ice can be added as your optional 'hosting insurance policy' to your live Cluster Pro or Cluster Lite, but can't be used independently.

Our pricing is based on Engines – virtual units of servers resources bundled into single package with simple and transparent pricing. Each Aegir instance can have multiple Engines attached. We don’t limit the number of Drupal sites you can host per Aegir instance and it can automatically scale up both the RAM and CPU for short load peaks. Since the beginning in 1997 and especially since 2009 we like to use motorsport naming convention in our hosting options, because we always focus on the top performance, top speed and greatest security of your Drupal sites hosted with us.