Quality of service for memory devices using weighted memory access operation types

A method comprising:

• providing a plurality of functions for accessing a memory device, wherein a function of the plurality of function receives I/O operations from a host computing system;
• selecting a first function of the plurality of functions to service;
• computing, by measuring a number of I/O operations serviced in one unit of time using a timer, an average latency of the memory device;
• determining a first operation weight and a second operation weight, wherein the first operation weight and the second operation weight are each computed by comparing a target latency to the average latency;
• assigning the first operation weight to a first I/O operation type of I/O operations received at the first function and the second operation weight to a second I/O operation type of I/O operations received at the first function, wherein the first operation weight indicates a first number of I/O operations having the first I/O operation type to be executed during an execution cycle of the memory device, and wherein the second operation weight indicates a second number of I/O operations having the second I/O operation type to be executed during the execution cycle of the memory device;
• assigning a maximum value and a minimum value to one or more parameters of the first function pertaining to the first I/O operation type and the second I/O operation type based on the assigned first operation weight and the assigned second operation weight, wherein the one or more parameters comprise a number of I/O operations per second received at the first function;
• selecting, for execution, a first number of operations of the first I/O operation type of the I/O operations received at the first function according to the first operation weight and a second number of operations of the second I/O operation type of the I/O operations received at the first function according to the second operation weight;
• and
• executing the first number of operations and the second number of operations at the memory device.

A non-transitory computer-readable storage medium comprising instructions that, when executed by a processing device, cause the processing device to:

• execute a selection scheme to select I/O operations from a plurality of functions for accessing a memory device during an execution cycle;
• select a function of the plurality of functions to service;
• compute, by measuring a number of I/O operations serviced in one unit of time using a timer, an average latency of the memory device;
• determine a first operation weight and a second operation weight, wherein the first operation weight and the second operation weight are each computed by comparing a target latency to the average latency;
• determine a number of I/O operations allocated to the function;
• retrieve, from a first queue, a first number of a first I/O operation type operations based on the first operation weight assigned to the first I/O operation type and the number of I/O operations allocated to the function, wherein the first operation weight indicates a first number of I/O operations having the first I/O operation type to be executed during the execution cycle of the memory device;
• retrieve, from a second queue, a second number of a second I/O operation type operations based on the second operation weight assigned to the second I/O operation type and the number of I/O operations allocated to the function, wherein the second operation weight indicates a second number of I/O operations having the second I/O operation type to be executed during the execution cycle of the memory device;
• assign a maximum value and a minimum value to one or more parameters of the function pertaining to the first I/O operation type and the second I/O operation type based on the assigned first operation weight and the assigned second operation weight, wherein the one or more parameters comprise a number of I/O operations per second received at the function;
• and
• execute the first number of operations and the second number of operations during the execution cycle.