Systems, methods, and apparatus for memory access in storage devices
A method for memory access may include receiving, at a device, a first memory access request for a parallel workload, receiving, at the device, a second memory access request for the parallel workload, processing, by a first logical device of the device, the first memory access request, and processing, by a second logical device of the device, the second memory access request. Processing the first memory access request and processing the second memory access request may include parallel processing the first and second memory access requests. The first logical device may include one or more first resources. The method may further include configuring the first logical device based on one or more first parameters of the parallel workload. The method may further include allocating one or more first resources to the first logical device based on at least one of the one or more first parameters of the parallel workload.
1. A method for memory access, the method comprising:
receiving, at a device, a first memory access request for a parallel workload;
receiving, at the device, a second memory access request for the parallel workload;
processing, by a first logical device of the device, with a first bandwidth, the first memory access request; and
processing, by a second logical device of the device, with a second bandwidth, the second memory access request, where the first bandwidth and second bandwidth are different and are assigned based on a number of lookups of a data structure that is part of a computational model.
2. The method of claim 1 , wherein processing the first memory access request and processing the second memory access request comprises parallel processing the first memory access request and the second memory access request.
3. The method of claim 1 , wherein the first logical device comprises one or more first resources.
4. The method of claim 3 , wherein the second logical device comprises one or more second resources.
5. The method of claim 3 , wherein:
the first logical device has a device physical address (DPA); and
the DPA is mapped to a physical block address (PBA) of the one or more first resources.
6. The method of claim 1 , wherein the first memory access request comprises a tag corresponding to the first logical device.
7. The method of claim 1 , further comprising configuring the first logical device based on one or more first parameters of the parallel workload.
8. The method of claim 7 , wherein configuring the first logical device is performed by a fabric manager.
9. The method of claim 7 , further comprising allocating one or more first resources to the first logical device based on at least one of the one or more first parameters of the parallel workload.
10. The method of claim 9 , further comprising mapping the one or more first resources to the first logical device.
11. The method of claim 7 , further comprising configuring the second logical device based on one or more second parameters of the parallel workload.
12. A device comprising:
a communication interface;
a first logical device having a first bandwidth, configured to:
receive, through the communication interface, a first memory access request; and
process, using one or more first resources, the first memory access request; and
a second logical device having a second bandwidth, configured to:
receive, through the communication interface, a second memory access request; and
process, using one or more second resources, the second memory access request, where the first bandwidth and second bandwidth are different and are assigned based on a number of lookups of a data structure that is part of a computational model.
13. The device of claim 12 , wherein the first logical device and the second logical device are configured to parallel process the first memory access request and the second memory access request.
14. The device of claim 12 , further comprising:
support logic configured to allocate the one or more first resources to a first logical device based on first configuration information received through the communication interface;
and mapping logic configured to map the one or more first resources to the first logical device.
15. The device of claim 14 , wherein:
the support logic is configured to allocate the one or more second resources to a second logical device based on second configuration information received through the communication interface; and
the mapping logic is configured to map the one or more second resources to the second logical device.
16. The device of claim 14 , wherein the support logic is configured to receive the first configuration information by an application programming interface.
17. The device of claim 12 , wherein the communication interface comprises a memory coherent interface.
18. A device comprising:
a communication interface;
one or more first memory resources;
one or more second memory resources;
support logic configured to allocate the one or more first memory resources to a first logical device based on first configuration information received through the communication interface, where the first logical device has a first bandwidth;
the support logic is configured to allocate the one or more second memory resources to a second logical device, where the second logical device has a second bandwidth and the first bandwidth and second bandwidth are different and the first and second bandwidth are assigned based on a number of lookups of a data structure that is part of a computational model; and
mapping logic configured to map the one or more first memory resources to the first logical device.
19. The device of claim 18 , wherein:
the support logic is configured to allocate the one or more second memory resources to the second logical device based on second configuration information received through the communication interface; and
the mapping logic is configured to map the one or more second memory resources to the second logical device.
20. The device of claim 18 , wherein the support logic is configured to receive the first configuration information by an application programming interface.