Data access path optimization
Aspects disclosed herein relate to a method comprising: obtaining a list of data paths to at least one persistent storage device through a plurality of NUMA nodes; associating with each data path, access performance information; receiving a request to access one of the at least one persistent storage device; calculating a preferred data path to the one of the at least one persistent storage device using the access performance information; and accessing the one of the at least one persistent storage device using the preferred data path.
1. A method comprising:
obtaining a list of data paths from at least one central processing unit (CPU) core to at least one persistent storage device;
associating access performance information with each data path;
receiving a request to access a first persistent storage device of the at least one persistent storage device;
calculating a preferred data path from the at least one CPU core to the first persistent storage device using the access performance information;
accessing the first persistent storage device using the preferred data path; and
updating the access performance information of the used data path,
wherein the access performance information of each data path includes at least a last update time of the access performance information.
2. The method of claim 1 , further comprising:
identifying ones of the data paths for testing;
testing the identified data paths; and
updating the access performance information of the tested data paths.
3. The method of claim 2 , wherein identifying ones of the data paths for testing comprises:
calculating an age of the access performance information using the last update time; and
identifying ones of the data paths for which the age of the access performance information exceeds an age threshold.
4. The method of claim 2 , wherein identifying ones of the data paths for testing comprises selecting a percentage of all data paths for testing.
5. The method of claim 2 , wherein identifying ones of the data paths for testing comprises selecting all the data paths that have an error for testing.
6. The method of claim 1 , wherein the access performance information of each data path includes at least a latency of the data path.
7. The method of claim 6 , wherein the preferred data path is calculated based on at least the latency of each data path.
8. The method of claim 1 , wherein the access performance information of each data path includes at least an average bandwidth indicator of the data path.
9. The method of claim 8 , wherein the preferred data path is calculated based on at least the average bandwidth indicator of each data path.
10. The method of claim 1 , wherein the access performance information of each data path includes at least a load on the data path.
11. The method of claim 10 , wherein the preferred data path is calculated based on at least the load on each data path.
12. The method of claim 1 , wherein the access performance information of each data path includes at least a reliability indicator of the data path.
13. The method of claim 12 , wherein the preferred data path is calculated based on at least the reliability indicator on each data path.
14. The method of claim 1 , wherein accessing the first persistent storage device comprises read access.
15. The method of claim 1 , wherein accessing the first persistent storage device comprises write access.
16. A system comprising:
a memory;
a plurality of non-uniform memory access (NUMA) nodes, each NUMA node comprising:
at least one central processing unit (CPU) core; and
at least a portion of the memory attached to the at least one CPU core,
wherein the memory stores instructions that, when executed by the plurality of NUMA nodes, configures the system to:
obtain a list of data paths from a first one of the at least one CPU core to at least one persistent storage device;
associate access performance information with each data path;
receive a request to access a first persistent storage device of the at least one persistent storage device;
calculate a preferred data path from the first CPU core to the first persistent storage device using the access performance information;
access the first persistent storage device using the preferred data path; and
update the access performance information of the used data path,
wherein the access performance information of each data path includes at least a last update time of the access performance information.
17. A non-transitory computer readable medium comprising instructions which, when executed by one or more processors, cause a computing device to perform steps comprising:
obtaining a list of data paths from at least one central processing unit (CPU) core to at least one persistent storage device;
associating access performance information with each data path;
receiving a request to access a first persistent storage device of the at least one persistent storage device;
calculating a preferred data path from the at least one CPU core to the first persistent storage device using the access performance information;
accessing the first persistent storage device using the preferred data path; and
updating the access performance information of the used data path,
wherein the access performance information of each data path includes at least a last update time of the access performance information.