Cooperative resource management

1. A system for deleting content, comprising:

a processor configured to:

receive, at a first node of a plurality of nodes, a delete request for a file, a primary copy of which is stored on each of at least a subset of the plurality of nodes, wherein at least one other node of the plurality of nodes has received the delete request for the file;

determine, at the first node, that the first node does not have a record of the file;

determine, at the first node, that at least some other nodes in the plurality of nodes have replaced their respective copies of the file with a first delete token for the file;

add, at the first node, the first delete token for the file in records associated with the first node in response to determining that the at least some other nodes in the plurality of nodes have replaced their respective copies of the file with the first delete token for the file;

determine, at the first node, that one or more of the at least some other nodes in the plurality of nodes have replaced their respective first delete token with a second delete token; and

not add the second delete token in the records associated with the first node in response to determining that the one or more of the at least some other nodes in the plurality of nodes have replaced their respective first delete token with the second delete token; and

a memory coupled to the processor and configured to store at least the records associated with the first node.

2. The system of claim 1 , wherein the first delete token for the file includes a list of other nodes in the plurality of nodes, wherein the list of other nodes is usable to find all other nodes in the plurality of nodes that should have the first delete token.

3. The system of claim 1 , wherein the delete request is generated by a user.

4. The system of claim 1 , wherein the delete request is generated by an application.

5. The system of claim 1 , wherein the processor is further configured to:

determine that a second node in the subset of the plurality of nodes has a cached copy of the file in addition to the primary copy; and

instruct the second node to disable the cached copy of the file.

6. The system of claim 1 , wherein:

a second node in the subset of the plurality of nodes that have the primary copy of the file stored thereon is configured to remove an entry associated with the file from a distributed index in response at least in part to at least some other nodes in the subset of the plurality of nodes having replaced their respective copies of the file with the first delete token.

7. The system of claim 6 , wherein:

at least some of the subset of the plurality of nodes are configured to store a portion of the distributed index;

the second node is configured to remove the entry associated with the file from the portion of the distributed index stored at the second node in response to the portion of the distributed index stored at the second node including the entry associated with the file; and

the second node is configured to send a request to remove the entry associated with the file from the distributed index to another node at which the portion of the distribution that includes the entry associated with the file is stored.

8. A method for deleting content, comprising:

receiving, at a first node of a plurality of nodes, a delete request for a file, a primary copy of which is stored on each of at least a subset of the plurality of nodes, wherein at least one other node of the plurality of nodes has received the delete request for the file;

determining, at the first node, that the first node does not have a record of the file;

determining, at the first node, that at least some other nodes in the plurality of nodes have replaced their respective copies of the file with a first delete token for the file;

adding, at the first node, the first delete token for the file in records associated with the first node in response to determining that the at least some other nodes in the plurality of nodes have replaced their respective copies of the file with the first delete token for the file;

determining, at the first node, that one or more of the at least some other nodes in the plurality of nodes have replaced their respective first delete token with a second delete token; and

not adding the second delete token in the records associated with the first node in response to the determining that the one or more of the at least some other nodes in the plurality of nodes have replaced their respective first delete token with the second delete token.

9. The method of claim 8 , wherein the first delete token for the file includes a list of other nodes in the plurality of nodes, wherein the list of other nodes is usable to find all other nodes in the plurality of nodes that should have the first delete token.

10. The method of claim 8 , wherein the delete request is generated by a user.

11. The method of claim 8 , wherein the delete request is generated by an application.

12. The method of claim 8 , further comprising:

determining that a second node in the subset of the plurality of nodes has a cached copy of the file in addition to the primary copy; and

instructing the second node to disable the cached copy of the file.

13. The method of claim 8 , wherein:

a second node in the subset of the plurality of nodes that have the primary copy of the file stored thereon is configured to remove an entry associated with the file from a distributed index in response at least in part to at least some other nodes in the subset of the plurality of nodes having replaced their respective copies of the file with the first delete token.

14. The method of claim 13 , wherein:

at least some of the subset of the plurality of nodes are configured to store a portion of the distributed index;

the second node is configured to remove the entry associated with the file from the portion of the distributed index stored at the second node in response to the portion of the distributed index stored at the second node including the entry associated with the file; and

the second node is configured to send a request to remove the entry associated with the file from the distributed index to another node at which the portion of the distribution that includes the entry associated with the file is stored.

15. A non-transitory computer readable storage medium for deleting content, the non-transitory computer readable storage medium comprising computer instructions for:

receiving, at a first node of a plurality of nodes, a delete request for a file, a primary copy of which is stored on each of at least a subset of the plurality of nodes, wherein at least one other node of the plurality of nodes has received the delete request for the file;

determining, at the first node, that the first node does not have a record of the file;

determining, at the first node, that at least some other nodes in the plurality of nodes have replaced their respective copies of the file with a first delete token for the file;

adding, at the first node, the first delete token for the file in records associated with the first node in response to determining that the at least some other nodes in the plurality of nodes have replaced their respective copies of the file with the first delete token for the file;

determining, at the first node, that one or more of the at least some other nodes in the plurality of nodes have replaced their respective first delete token with a second delete token; and

not adding the second delete token in the records associated with the first node in response to determining that the one or more of the at least some other nodes in the plurality of nodes have replaced their respective first delete token with the second delete token.

16. The computer program product of claim 15 , wherein the first delete token for the file includes a list of other nodes in the plurality of nodes, wherein the list of other nodes is usable to find all other nodes in the plurality of nodes that should have the first delete token.

17. An apparatus, comprising:

means for receiving a delete request for a file, a primary copy of which is stored on each of at least a subset of a plurality of nodes, wherein at least one other node of the plurality of nodes has received the delete request for the file;

means for determining that the apparatus does not have a record of the file;

means for determining that at least some other nodes in the plurality of nodes have replaced their respective copies of the file with a first delete token for the file;

means for adding the first delete token for the file in records associated with the apparatus in response to determining that the at least some other nodes in the plurality of nodes have replaced their respective copies of the file with the first delete token for the file;

means for determining that one or more of the at least some other nodes in the plurality of nodes have replaced their respective first delete token with a second delete token; and

means for not adding the second delete token for the file in the records associated with the apparatus in response to determining that the one or more of the at least some other nodes in the plurality of nodes have replaced their respective first delete token with the second delete token.

18. The apparatus of claim 17 , wherein the first delete token for the file includes a list of other nodes in the plurality of nodes, wherein the list of other nodes is usable to find all other nodes in the plurality of nodes that should have the first delete token.