Distributed resource allocation

In a computing environment comprising a plurality of computing devices having capacity that is logically segmented into first resource slots of a plurality of resource types, wherein the first resource slots are configured to be associated with at least one of a plurality of entities, a method comprising:

• receiving, from at least one entity of the plurality of entities, a request to fulfill a computational task and at least one constraint for fulfilling the computational task, the request being independent of a specified quantity or type of resource slot to fulfill the computation task;
• determining, by one or more processors of a computing system, a quantity or type of the first resource slots sufficient to fulfill the request in accordance with the at least one constraint;
• allocating, by the one or more processors of the computing system, the determined quantity or type of the first resource slots to a plurality of geographically separate computing environments as allocated resource slots;
• and
• allocating, by the one or more processors of the computing system, master resource slots in the plurality of geographically separate computing environments based on the at least one constraint and the quantity or type, wherein two or more of the master resource slots are allocated to a first one of the plurality of geographically separate computing environments, and wherein the two or more of the master resource slots manage two or more of the allocated resource slots within the first one of the plurality of geographically separate computing environments.

A method for allocating computing resources to customers of a multi-tenant web services platform, the method comprising:

• receiving, from one of a plurality of entities, a request comprising one or more parameters indicative of a computing use and at least one criterion for fulfilling the computing use without requiring specification of a quantity of units of computing instances for fulfilling the computing use;
• determining a quantity of first units of computing instances sufficient to fulfill the computing use in accordance with the at least one criterion;
• allocating the quantity of the first units of computing instances to a plurality of geographically separate computing environments as allocated units of computing instances;
• allocating a ratio of master computing instances to the allocated units of computing instances based on the at least one criterion, wherein two or more of the master computing instances are allocated to a first one of the plurality of geographically separate computing environments, and wherein the two or more of the master computing instances manage two or more of the allocated units of computing instances within the first one of the plurality of geographically separate computing environments;
• and
• modifying the master computing instances as the allocated units of computing instances are redistributed over the plurality of geographically separate computing environments.