Dynamic multitasking for distributed storage systems by detecting events for triggering a context switch
View Patent ↗A method is disclosed for dynamic multitasking in a storage system, the storage system including a first storage server configured to execute a first I/O service process and one or more second storage servers, the method comprising: detecting a first event for triggering a context switch; transmitting to each of the second storage servers an instruction to stop transmitting internal I/O requests to the first I/O service process, the instruction including an identifier corresponding to the first I/O service process, the identifier being arranged to distinguish the first I/O service process from other first I/O service processes that are executed by the first storage server concurrently with the first I/O service process; deactivating the first I/O service process by pausing a frontend of the first I/O service process, and pausing one or more I/O providers of the first I/O service process; and executing a first context switch between the first I/O service process and a second process.
1. A method for dynamic multitasking in a storage system, the storage system including a first storage server configured to execute a first I/O service process and one or more second storage servers, the method comprising:
detecting a first event for triggering a context switch;
transmitting to each of the second storage servers an instruction to stop transmitting internal I/O requests to the first I/O service process, the instruction including an identifier corresponding to the first I/O service process, the identifier being arranged to distinguish the first I/O service process from other first I/O service processes that are executed by the first storage server concurrently with the first I/O service process;
deactivating the first I/O service process by pausing a frontend of the first I/O service process, and pausing one or more I/O providers of the first I/O service process; and
after the first I/O service process is deactivated, executing a first context switch between the first I/O service process and a second process, the first I/O service process and the second process being executed on a same processor core of the first storage server, and the first context switch causing the processor core to stop executing the first I/O service process and begin executing the second process.
2. The method of claim 1 , wherein pausing the frontend of the first I/O service process includes transmitting to the first I/O service process an instruction, which when received by the first I/O service process causes the first I/O service process to set a synchronization barrier that is arranged to prevent the frontend of the first I/O service process from retrieving I/O requests from a pool of I/O requests after any on-going retrieval of I/O requests is completed.
3. The method of claim 1 , wherein pausing the one or more I/O providers of the first I/O service process includes transmitting to the first I/O service process an instruction, which when received by the first I/O service process causes the first I/O service process to set a synchronization barrier that is arranged to prevent the I/O providers from beginning execution of new I/O tasks after completing any in-flight tasks that are being executed by the I/O providers.
4. The method of claim 1 , wherein disabling the first I/O service process includes disabling one or more background components of the first I/O service process.
5. The method of claim 1 , further comprising:
detecting a second event for triggering a context switch;
transmitting to each of the second storage servers an instruction to resume transmitting internal I/O requests to the first I/O service process, the instruction including an identifier corresponding to the first I/O service process;
activating the first I/O service process by resuming a frontend of the first I/O service process, and resuming one or more I/O providers of the first I/O service process; and
after the first I/O service process is activated, executing a second context switch between the second process and the first I/O service process, the second context switch causing the processor core to stop executing the second process and begin executing the first I/O service process.
6. The method of claim 1 , wherein pausing the frontend of the first I/O service process includes transmitting to the first I/O service process an instruction, which when received by the first I/O service process causes the first I/O service process to release a synchronization barrier that is arranged to prevent the frontend of the first I/O service process from retrieving I/O requests from a pool of I/O requests after any on-going retrieval of I/O requests is completed.
7. The method of claim 1 , wherein pausing the one or more I/O providers of the first I/O service process includes transmitting to the first I/O service process an instruction, which when received by the first I/O service process causes the first I/O service process to release a synchronization barrier that is arranged to prevent the I/O providers from beginning execution of new I/O tasks after completing any in-flight tasks that are being executed by the I/O providers.
8. An apparatus for use in a storage system, the storage system including a first storage server configured to execute a first I/O service process and one or more second storage servers, the apparatus comprising:
a memory; and
at least one processor operatively coupled to the memory, the at least one processor being configured to:
detect a first event for triggering a context switch;
transmit to each of the second storage servers an instruction to stop transmitting internal I/O requests to the first I/O service process, the instruction including an identifier corresponding to the first I/O service process, the identifier being arranged to distinguish the first I/O service process from other first I/O service processes that are executed by the first storage server concurrently with the first I/O service process;
deactivate the first I/O service process by pausing a frontend of the first I/O service process, and pausing one or more I/O providers of the first I/O service process; and
after the first I/O service process is deactivated, execute a first context switch between the first I/O service process and a second process, the first I/O service process and the second process being executed on a same processor core of the first storage server, and the first context switch causing the processor core to stop executing the first I/O service process and begin executing the second process.
9. The apparatus of claim 8 , wherein pausing the frontend of the first I/O service process includes transmitting to the first I/O service process an instruction, which when received by the first I/O service process causes the first I/O service process to set a synchronization barrier that is arranged to prevent the frontend of the first I/O service process from retrieving I/O requests from a pool of I/O requests after any on-going retrieval of I/O requests is completed.
10. The apparatus of claim 8 , wherein pausing the one or more I/O providers of the first I/O service process includes transmitting to the first I/O service process an instruction, which when received by the first I/O service process causes the first I/O service process to set a synchronization barrier that is arranged to prevent the I/O providers from beginning execution of new I/O tasks after completing any in-flight tasks that are being executed by the I/O providers.
11. The apparatus of claim 8 , wherein disabling the first I/O service process includes disabling one or more background components of the first I/O service process.
12. The apparatus of claim 8 , wherein the one or more processors are further configured to:
detect a second event for triggering a context switch;
transmit to each of the second storage servers an instruction to resume transmitting internal I/O requests to the first I/O service process, the instruction including an identifier corresponding to the first I/O service process;
activate the first I/O service process by resuming a frontend of the first I/O service process, and resuming one or more I/O providers of the first I/O service process; and
after the first I/O service process is activated, execute a second context switch between the second process and the first I/O service process, the second context switch causing the processor core to stop executing the second process and begin executing the first I/O service process.
13. The apparatus of claim 8 , wherein pausing the frontend of the first I/O service process includes transmitting to the first I/O service process an instruction, which when received by the first I/O service process causes the first I/O service process to release a synchronization barrier that is arranged to prevent the frontend of the first I/O service process from retrieving I/O requests from a pool of I/O requests after any on-going retrieval of I/O requests is completed.
14. The apparatus of claim 8 , wherein pausing the one or more I/O providers of the first I/O service process includes transmitting to the first I/O service process an instruction, which when received by the first I/O service process causes the first I/O service process to release a synchronization barrier that is arranged to prevent the I/O providers from beginning execution of new I/O tasks after completing any in-flight tasks that are being executed by the I/O providers.
15. A non-transitory computer-readable medium storing one or more processor-executable instructions, which when executed by at least one processor cause the at least one processor to execute the operations of:
detecting a first event for triggering a context switch in a first storage server, the first server being configured to execute a first I/O service process for servicing I/O requests in a distributed storage system;
transmitting to each of a plurality of second storage servers that are part of the distributed storage system an instruction to stop transmitting internal I/O requests to the first I/O service process, the instruction including an identifier corresponding to the first I/O service process, the identifier being arranged to distinguish the first I/O service process from other first I/O service processes that are executed by the first storage server concurrently with the first I/O service process;
deactivating the first I/O service process by pausing a frontend of the first I/O service process, and pausing one or more I/O providers of the first I/O service process; and
after the first I/O service process is deactivated, executing a first context switch between the first I/O service process and a second process, the first I/O service process and the second process being executed on a same processor core of the first storage server, and the first context switch causing the processor core to stop executing the first I/O service process and begin executing the second process.
16. The non-transitory computer-readable medium of claim 15 , wherein pausing the frontend of the first I/O service process includes transmitting to the first I/O service process an instruction, which when received by the first I/O service process causes the first I/O service process to set a synchronization barrier that is arranged to prevent the frontend of the first I/O service process from retrieving I/O requests from a pool of I/O requests after any on-going retrieval of I/O requests is completed.
17. The non-transitory computer-readable medium of claim 15 , wherein pausing the one or more I/O providers of the first I/O service process includes transmitting to the first I/O service process an instruction, which when received by the first I/O service process causes the first I/O service process to set a synchronization barrier that is arranged to prevent the I/O providers from beginning execution of new I/O tasks after completing any in-flight tasks that are being executed by the I/O providers.
18. The non-transitory computer-readable medium of claim 15 , wherein the one or more processor-executable instructions are further configured to cause the at least one processor to perform the operations of:
detecting a second event for triggering a context switch;
transmitting to each of the second storage servers an instruction to resume transmitting internal I/O requests to the first I/O service process, the instruction including an identifier corresponding to the first I/O service process;
activating the first I/O service process by resuming a frontend of the first I/O service process, and resuming one or more I/O providers of the first I/O service process; and
after the first I/O service process is activated, executing a second context switch between the second process and the first I/O service process, the second context switch causing the processor core to stop executing the second process and begin executing the first I/O service process.
19. The non-transitory computer-readable medium of claim 15 , wherein pausing the frontend of the first I/O service process includes transmitting to the first I/O service process an instruction, which when received by the first I/O service process causes the first I/O service process to release a synchronization barrier that is arranged to prevent the frontend of the first I/O service process from retrieving I/O requests from a pool of I/O requests after any on-going retrieval of I/O requests is completed.
20. The non-transitory computer-readable medium of claim 15 , wherein pausing the one or more I/O providers of the first I/O service process includes transmitting to the first I/O service process an instruction, which when received by the first I/O service process causes the first I/O service process to release a synchronization barrier that is arranged to prevent the I/O providers from beginning execution of new I/O tasks after completing any in-flight tasks that are being executed by the I/O providers.