Self-organizing storage system for asynchronous storage service

A non-transitory machine-readable medium comprising machine executable instructions for a self-organizing storage system of a cloud storage service, the instructions to:

• maintain a first set of one or more structures to indicate storage service requests received in a frontend of the cloud storage service;
• maintain a second set of one or more structures for a backend of the cloud storage service, wherein each of the second set of one or more structures indicates selection from the first set of structures of storage service requests for fulfillment by the backend;
• update the first set of structures to indicate requests to be fulfilled by the backend;
• periodically communicate the first set of structures from the frontend to the backend;
• select storage service requests to fulfill in the backend from the storage service requests indicated in the first set of structures based, at least in part, on a set of one or more backend storage indexes;
• and
• indicate in the second set of structures selected storage service requests.

A storage system comprising:

• a plurality of servers, each of the plurality of servers comprising a processor and a machine-readable medium with instructions stored thereon, the instructions executable by the processor to cause the server to,
• maintain a first set of one or more structures to indicate storage service requests received at the server;
• maintain a second set of one or more structures received from backend storage nodes, wherein each of the second set of one or more structures indicates storage service requests selected, by a corresponding of the backend storage nodes, from storage service requests indicated in the first set of structures;
• update the first set of structures to remove indications of storage service requests indicated in the second set of structures;
• the backend storage nodes, each of the backend storage nodes comprising a plurality of storage medium devices and a network interface, each of the backend storage nodes programmed to,
• for different ones of the plurality of servers,
• obtain, periodically via the network interface, the first set of structures from the server;
• select storage service requests to fulfill from the storage service requests indicated in the first set of structures, the selection based, at least in part, on a storage index of the backend storage node;
• indicate selected storage service requests in a second structure;
• communicate the second structure to the server;
• and
• attempt to fulfill the selected storage service requests.