CPC H04L 47/6245 (2013.01) [H04L 43/106 (2013.01); H04L 47/6225 (2013.01)] | 32 Claims |
1. A method for managing requests in a distributed computing system comprising a plurality of computing nodes, the method comprising:
receiving, by one or more of the plurality of computing nodes in the distributed computing system, a plurality of requests;
assigning a respective timestamp to each of the plurality of requests;
assigning a first subset of the plurality of requests to a first epoch of a plurality of epochs, wherein each of the plurality of epochs comprises a respective predetermined closing time of a plurality of predetermined closing times, and wherein the first epoch corresponds to a first interval of time comprising the respective timestamp of each request of the first subset of the plurality of requests;
comparing a present time to the predetermined closing time of the first epoch, wherein the predetermined closing time of the first epoch is after the first interval of time and based on (i) a duration of the first interval of time and (ii) a grace period of time; and
queuing, based on the comparison and when the present time is equivalent to or greater than the predetermined closing time of the first epoch, the first subset of the plurality of requests for execution in a waiting data structure by last-in, first-out (LIFO) ordering based on the respective timestamp of each request of the first subset.
|