Application startup page fault management in a hardware multithreading environment
A method, system and computer-usable medium are disclosed for startup page fault management improves application startup performance by assigning startup tasks to a hardware thread 0 across plural processing cores in a simultaneous multithreading environment to provide more rapid processing of processor bound page faults. I/O bound page faults are flagged to associated with predetermined cache locations to improve data and text first reference page-in I/O response.
1. A system comprising:
a processor;
a data bus coupled to the processor; and
a computer-usable medium embodying computer program code, the computer-usable medium being coupled to the data bus, the computer program code used for application startup page fault management and comprising instructions executable by the processor and configured for:
detecting of an application startup for an application designated to receive the application startup page fault management;
determining plural tasks associated with the startup of the application;
scheduling the plural tasks to a hardware thread 0 of each of plural processor cores; executing application startup to a predetermined steady state, wherein the predetermined steady state comprises at least a predetermined number of dispatches of each of the plural tasks on their respective thread 0 ; and
responsive to the predetermined steady state, scheduling tasks associated with the application to hardware threads in addition to thread 0 .
2. The system of claim 1 , wherein the instructions executable by the processor are further configured for:
caching plural first reference page-in I/Os of the application in a predetermined cache in association with a flag;
detecting a first reference page-in I/O flag during the executing application startup; and
in response to detecting, retrieving the first reference page-in I/O from the predetermined cache.
3. The system of claim 2 , wherein caching plural first reference page-in I/Os of the application further comprises:
monitoring startup of the application to detect page faults; and
in response to detecting a page fault, caching the page-in I/O of the fault in association with the first reference page-in I/O flag.
4. The system of claim 3 , wherein the predetermined steady state comprises retrieval of predetermined cached first reference page-in I/Os.
5. The system of claim 1 , wherein detecting of an application startup for an application designated to receive the application startup page fault management further comprises detecting one of a fork, exec, or thread create_system calls.
6. A non-transitory, computer-readable storage medium embodying computer program code, the computer program code comprising computer executable instructions configured for:
detecting of an application startup for an application designated to receive the application startup page fault management;
determining plural tasks associated with the startup of the application;
scheduling the plural tasks to a hardware thread 0 of each of plural processor cores; executing application startup to a predetermined steady state, wherein the predetermined steady state comprises at least a predetermined number of dispatches of each of the plural tasks on their respective thread 0 ; and
responsive to the predetermined steady state, scheduling tasks associated with the application to hardware threads in addition to thread 0 .
7. The non-transitory, computer-readable storage medium of claim 6 , wherein the instructions executable by the processor are further configured for:
caching plural first reference page-in I/Os of the application in a predetermined cache in association with a flag;
detecting a first reference page-in I/O flag during the executing application startup; and
in response to detecting, retrieving the first reference page-in I/O from the predetermined cache.
8. The non-transitory, computer-readable storage medium of claim 7 , wherein caching plural first reference page-in I/Os of the application further comprises:
monitoring startup of the application to detect page faults; and
in response to detecting a page fault, caching the page-in I/O of the fault in association with the first reference page-in I/O flag.
9. The non-transitory, computer-readable storage medium of claim 8 , wherein the predetermined steady state comprises retrieval of predetermined cached first reference page-in I/Os.
10. The non-transitory, computer-readable storage medium of claim 6 , wherein detecting of an application startup for an application designated to receive the application startup page fault management further comprises detecting one of a fork, exec, or thread_create system calls.
11. The non-transitory, computer-readable storage medium of claim 6 , wherein the computer executable instructions are deployable to a client system from a server system at a remote location.
12. The non-transitory, computer-readable storage medium of claim 6 , wherein the computer executable instructions are provided by a service provider to a user on an on-demand basis.