Apollo: Scalable and Coordinated Scheduling for Cloud-Scale Computing

Amazon also has a cloud deployment tool called Apollo (internally only I believe - the public version is called &quot;CodeDeploy&quot;): <a href="http:&#x2F;&#x2F;www.allthingsdistributed.com&#x2F;2014&#x2F;11&#x2F;apollo-amazon-deployment-engine.html" rel="nofollow">http:&#x2F;&#x2F;www.allthingsdistributed.com&#x2F;2014&#x2F;11&#x2F;apollo-amazon-de...</a>

This looks really great honestly. Including IO as a consumable resource is just as important as allocating CPU or memory.

Request to the mods: please drop the &quot;Microsoft&quot; from the title. This is just a Microsoft Research paper. MSR is almost completely separate from the rest of the company.<p>(edit: whoops, looks like MS actually uses this system internally. sorry, should have read the abstract)