Fast cache purge optimization handling of unavailable nodes
Processing a purge request is disclosed. In an embodiment, the purge request is received from a node, where the purge request is for a next purge instruction and the node has an associated queue of purge instruction(s) with associated timestamps. In response to receiving the purge request, providing an unprocessed purge instruction having a time stamp before a threshold time. After processing the purge instruction having a timestamp before the threshold time, processing the remaining purge instructions as follows: indicating an availability state of the node as transitional, storing a current time value as a reference time value, processing in chronological order those remaining purge instructions in the queue with a time value chronologically before the reference time value, and indicating an availability state of the node as available.
1. A method for processing a purge request, comprising:
receiving the purge request from a node, wherein the purge request is for a next purge instruction and the node has an associated queue of one or more purge instructions with associated timestamps;
in response to receiving the purge request, providing an unprocessed purge instruction having a timestamp before a threshold time; and
after processing the purge instruction having a timestamp before the threshold time, processing remaining purge instructions as follows:
indicating an availability state of the node as transitional;
storing a current time value as a reference time value;
processing, in chronological order, those remaining purge instructions in the queue with a time value chronologically before the reference time value; and
indicating an availability state of the node as available.
2. The method of claim 1 , wherein the next purge instruction is provided by a queue of a node unavailable to serve/distribute content to users.
3. The method of claim 2 , wherein the unavailable node is a node with a status of being offline.
4. The method of claim 2 , wherein an availability status of the unavailable node is tracked by a content distribution manager.
5. The method of claim 2 , wherein the unavailable node sends the purge request to begin processing a backlog of one or more purge instructions.
6. The method of claim 1 , wherein the providing the unprocessed purge instruction having a timestamp before the threshold time includes determining that a time value associated with the next purge instruction is chronologically before a threshold time.
7. The method of claim 1 , wherein the node is included in a content distribution network having a plurality of nodes, and each node in the content distribution network has a respective associated queue of one or more purge instructions.
8. The method of claim 1 , further comprising determining that there is at least one unprocessed purge instruction having a timestamp before a threshold time, and wherein the threshold time is a time when the determination is made.
9. The method of claim 1 , wherein the threshold time is a dynamic value determined by subtracting a predetermined amount of a time period from a current time.
10. The method of claim 1 , wherein the threshold time is selected based at least in part on an estimated time for verifying that first-in-first-out processing of queued purge instructions has reached a state of the queue where any purge instruction still waiting in the queue to be processed has been added to the queue.
11. The method of claim 1 , wherein the threshold time is selected based at least in part on a time allotted for purge instructions to remain queued in the queue before the node transitions to a transitional state.
12. The method of claim 1 , wherein the threshold time is the time when a first purge instruction was requested from the queue of the node after the node became unavailable.
13. The method of claim 1 , wherein in a transitional state, the node is not currently serving/distributing requested content as part of a content distribution network.
14. The method of claim 1 , wherein in the transitional state, the node does not provide a process completion confirmation to a purge request.
15. The method of claim 1 , wherein the reference time value is a time when the node enters a transitional state.
16. A system for processing a purge request, comprising:
a communication interface configured to receive the purge request from a node, wherein the purge request is for a next purge instruction and the node has an associated queue of one or more purge instructions with associated timestamps
a processor configured to:
in response to receiving the purge request, providing an unprocessed purge instruction having a timestamp before a threshold time; and
after processing the purge instruction having a timestamp before the threshold time, processing remaining purge instructions as follows:
indicating an availability state of the node as transitional;
storing a current time value as a reference time value;
processing, in chronological order, those remaining purge instructions in the queue with a time value chronologically before the reference time value; and
indicating an availability state of the node as available.
17. A computer program product embodied in a non-transitory computer readable storage medium and comprising computer instructions for:
receiving a purge request from a node, wherein the purge request is for a next purge instruction and the node has an associated queue of one or more purge instructions with associated timestamps;
in response to receiving the purge request, providing an unprocessed purge instruction having a timestamp before a threshold time; and
after processing the purge instruction having a timestamp before the threshold time, processing remaining purge instructions as follows:
indicating an availability state of the node as transitional;
storing a current time value as a reference time value;
processing, in chronological order, those remaining purge instructions in the queue with a time value chronologically before the reference time value; and
indicating an availability state of the node as available.