Memory centric computing storage controller system
A deaggregated computing system having a memory centric computing storage controller can transfer data from a source to a destination node while dynamically updating a transfer route between them. The transfer route can be recalculated based on the current conditions of the routing nodes between the source and destination. Recalculating the transfer route can be based on power status, bandwidth, in-use status, current capacity, or failure conditions. The deaggregated computing system can include one or more processor units coupled to one or more storage and memory units all connected by the memory centric computing storage controller that can route control and data packets between the processor units and the storage and memory units. The processor units and the storage units can be connected by a combination of serial data communication links and a data storage fabric network.
1. A method of operation of a memory centric computing system comprising:
calculating a transfer route based on a transfer request having a transfer source, a transfer destination, a transfer type, and a data target, the transfer route including one or more routing nodes between the transfer source and the transfer destination, wherein the transfer destination is a memory device;
recalculating the transfer route to exclude a target node of the routing nodes having an exclusion status less than or equal to a routing threshold, the exclusion status based on a power metric of one of the routing nodes less than or equal to a power threshold;
detecting a reservation for a target address within an address range of the transfer destination; and
transferring the target data to the transfer destination along the transfer route after the reservation indicates the target address is available.
2. The method as claimed in claim 1 , wherein calculating the exclusion status includes calculating the exclusion status based on one of the routing nodes having an in-use flag indicating active use.
3. The method as claimed in claim 1 , wherein calculating the exclusion status includes calculating the exclusion status based on a performance metric of one of the routing nodes less than or equal to a performance threshold.
4. The method as claimed in claim 1 , wherein detecting the reservation includes detecting the reservation in a memory map cache in one of the routing nodes in the transfer route.
5. The method as claimed in claim 1 , wherein transferring the target data includes transferring the target data to a processor unit, a memory unit, or a storage unit.
6. The method as claimed in claim 1 , wherein recalculating the transfer route includes recalculating the transfer route dynamically while transferring the target data to the transfer destination.
7. A deaggregated computing system comprising:
a processor unit coupled to a storage device; and
a memory centric computing storage controller configured to transfer target data between the processor unit and the storage device, the memory centric computing storage controller including:
a configuration module configured to calculate a transfer route based on a transfer request having a transfer source, a transfer destination, a transfer type, and a data target, the transfer route including one or more routing nodes between the transfer source and the transfer destination, for recalculating the transfer route to exclude a target node of the routing nodes having an exclusion status less than or equal to a routing threshold, and calculating the exclusion status based on a power metric of one of the routing nodes less than or equal to a power threshold, and wherein the transfer destination is a memory device,
a protocol engine configured to detect a reservation for a target address within an address range of the transfer destination, and
a fabric controller configured to transfer the target data to the transfer destination along the transfer route after the reservation indicates the target address is available.
8. The system as claimed in claim 7 , wherein the configuration manager is configured for calculating the exclusion status based on one of the routing nodes having an in-use flag indicating active use.
9. The system as claimed in claim 7 , wherein configuration manager is configured for calculating the exclusion status based on a performance metric of one of the routing nodes less than or equal to a performance threshold.
10. The system as claimed in claim 7 , wherein the protocol engine is configured for detecting the reservation in a memory map cache in one of the routing nodes in the transfer route.
11. The system as claimed in claim 7 , wherein the fabric controller is configured for transferring the target data to a processor unit, a memory unit, or a storage unit.
12. The system as claimed in claim 7 , wherein the configuration manager is configured for recalculating the transfer route dynamically while transferring the target data to the transfer destination.
13. One or more non-transitory computer-readable media storing instructions that, when executed by one or more computing devices, cause:
calculating a transfer route based on a transfer request having a transfer source, a transfer destination, a transfer type, and a data target, the transfer route including one or more routing nodes between the transfer source and the transfer destination, wherein the transfer destination is a memory device;
recalculating the transfer route to exclude a target node of the routing nodes having an exclusion status less than or equal to a routing threshold, the exclusion status based on a power metric of one of the routing nodes less than or equal to a power threshold;
detecting a reservation for a target address within an address range of the transfer destination;
transferring the target data to the transfer destination along the transfer route after the reservation indicates the target address is available.
14. The non-transitory computer-readable media of claim 13 , wherein the instruction when executed by the one or more computing devices, further cause calculating the exclusion status based on one of the routing nodes having an in-use flag indicating active use or the exclusion status based on a performance metric of one of the routing nodes less than or equal to a performance threshold.
15. The non-transitory computer-readable media of claim 13 , wherein the instruction when executed by the one or more computing devices, further cause detecting the reservation in a memory map cache in one of the routing nodes in the transfer route.
16. The non-transitory computer-readable media of claim 13 , wherein the instruction when executed by the one or more computing devices, further cause transferring the target data to a processor unit, a memory unit, or a storage unit.
17. The non-transitory computer-readable media of claim 13 , wherein the instruction when executed by the one or more computing devices, further cause recalculating the transfer route dynamically while transferring the target data to the transfer destination.