Task dispatcher for block storage devices
A system, method, and computer program product are provided for implementing a protocol stack for handling memory access requests associated with one or more block devices. The method includes the steps of allocating a thread to process a task associated with a memory access request for a block device; and executing the thread to process the task, wherein the thread implements a state machine that manages calls to two or more layers of the protocol stack. A task dispatcher is implemented for managing the execution of tasks using at least one thread pool, the tasks managing the calls to the various layers of the protocol stack. The protocol stack may implement abstraction layers for virtual storage devices as well as the real block devices.
1. A method comprising:
allocating a thread, executed by a processor included in a node, to process a task according to a protocol stack for handling memory access requests associated with one or more block devices included in the node;
executing the thread to process the task, wherein the thread implements a state machine that manages calls to two or more layers of the protocol stack; and
allocating a second thread, executed by the processor, to manage the allocation of a thread pool for processing a plurality of tasks associated with a plurality of memory access requests, wherein the second thread is configured to stall the thread processing the task when a direction of the task changes.
2. The method of claim 1 , wherein the processor is multi-threaded and two or more threads are executed by the processor simultaneously.
3. The method of claim 2 , further comprising:
allocating a third thread to process a second task according to the protocol stack; and
executing the third thread to process the second task,
wherein the thread is included in a first thread pool allocated to tasks having a first priority and the third thread is included in a second thread pool allocated to tasks having a second priority that is higher than the first priority.
4. The method of claim 1 , wherein the protocol stack includes a first layer that implements a server for communicating with one or more clients that generate memory access requests for the one or more block devices.
5. The method of claim 4 , wherein the protocol stack includes a second layer associated with one or more virtual storage device objects, and wherein each memory access request includes an address in a virtual storage device address space.
6. The method of claim 5 , wherein the protocol stack includes a third layer associated with one or more real storage device objects corresponding to the one or more block devices.
7. The method of claim 1 , wherein the task is pushed onto a first task lane in a plurality of task lanes.
8. The method of claim 7 , wherein the task is stalled based on a barrier instruction associated with the first task lane.
9. The method of claim 7 , wherein the plurality of task lanes are included in a task lane hierarchy, and wherein a barrier instruction associated with a parent task lane may stall any task associated with a child task lane corresponding to the parent task lane.
10. A non-transitory computer-readable storage medium storing instructions that, when executed by a processor, cause the processor to perform steps comprising:
allocating a thread, executed by a processor included in a node, to process a task according to a protocol stack for handling memory access requests associated with one or more block devices included in the node;
executing the thread to process the task, wherein the thread implements a state machine that manages calls to two or more layers of the protocol stack; and
allocating a second thread, executed by the processor, to manage the allocation of a thread pool for processing a plurality of tasks associated with a plurality of memory access requests, wherein the second thread is configured to stall the thread processing the task when a direction of the task changes.
11. The computer-readable storage medium of claim 10 , the steps further comprising:
allocating a third thread to process a second task according to the protocol stack; and
executing the third thread to process the second task,
wherein the thread is included in a first thread pool allocated to tasks having a first priority and the third thread is included in a second thread pool allocated to tasks having a second priority that is higher than the first priority.
12. The computer-readable storage medium of claim 10 , wherein the protocol stack includes:
a first layer that implements a server for communicating with one or more clients that generate memory access requests for the one or more block devices;
a second layer associated with one or more virtual storage device objects, wherein each memory access request includes an address in a virtual storage device address space; and
a third layer associated with one or more real storage device objects corresponding to the one or more block devices.
13. A system comprising:
one or more block devices;
a memory that includes a protocol stack for handling memory access requests associated with the one or more block devices; and
a processor coupled to the memory and configured to:
allocate a thread to process a task according to the protocol stack,
execute the thread to process the task, wherein the thread implements a state machine that manages calls to two or more layers of the protocol stack, and
allocate a second thread, executed by the processor, to manage the allocation of a thread pool for processing a plurality of tasks associated with a plurality of memory access requests, wherein the second thread is configured to stall the thread processing the task when a direction of the task changes.
14. The system of claim 13 , the steps further comprising:
allocating a second thread to process a second task according to the protocol stack; and
executing the second thread to process the second task,
wherein the thread is included in a first thread pool allocated to tasks having a first priority and the second thread is included in a second thread pool allocated to tasks having a second priority that is higher than the first priority.
15. The system of claim 13 , wherein the one or more block devices, the memory, and the processor are included in a first node, the system further comprising a second node, wherein a processor in the second node generates a memory access request associated with the task.