| 主权项 |
1. A method for managing a size for an issue queue in a system including a plurality of host computers, wherein each host computer includes an issue queue, the method comprising:
determining a latency value for an input/output IO request in a host computer of the plurality of host computers based on an amount of time the IO request spent in the host computer's issue queue, the issue queue of the host computer used to transmit IO requests from the host computer to a storage system shared by the plurality of host computers; determining a host specific value assigned to the host computer based in proportion on a number of shares assigned to the host in a quality of service policy for IO requests, wherein other host computers in the plurality of host computers are assigned other host specific values based on a number of shares assigned to each respective host computer in the quality of service policy; and determining the size for the host computer's issue queue based on the latency value and the host specific value to control a number of IO requests that are added to the host computer's issue queue, wherein other hosts in the plurality of hosts independently determine respective sizes for respective issue queues based on each host computer's host specific value and a respective latency value for each host computer. |
