Memory extensible chip
A memory extensible chip ( 200 ) is provided. The chip ( 200 ) includes a substrate ( 240 ), and a processor ( 230 ), a first memory module set ( 210 ), and a second memory module set ( 220 ) that are integrated on the substrate ( 240 ). The processor ( 230 ) communicates with at least one memory module in the first memory module set ( 210 ) using a first communications interface ( 250 ), and the processor ( 230 ) communicates with at least one memory module in the second memory module set ( 220 ) using a second communications interface ( 260 ). A memory module in the first memory module set ( 210 ) communicates with a memory module in the second memory module set ( 220 ) using a substrate network, where the substrate network is a communications network located inside the substrate ( 240 ). In this way, the processor ( 230 ) can access a memory module in the first memory module set ( 210 ) by using the second memory module set ( 220 ).
1. A memory extensible chip, comprising: a substrate, and a processor, a first memory module set, and a second memory module set that are integrated on the substrate of the chip, wherein
the processor communicates with at least one memory module in the first memory module set by using a first communications interface, and the processor communicates with at least one memory module in the second memory module set by using a second communications interface; and
a memory module in the first memory module set communicates with a memory module in the second memory module set by using a substrate network, wherein the substrate network is a communications network located inside the substrate of the chip, wherein a first processor core of the processor is configured to determine, from a plurality of communication paths that are from the first processor core to a first memory module of the first memory module set, a communication path that has a minimum hop count as an access path for accessing the first memory module in the first memory module set.
2. The chip according to claim 1 , wherein:
the processor comprises a plurality of processor cores, the plurality of processor cores communicate with each other by using a network on chip, and the network on chip is a communications network located outside the substrate; and
the first memory module set and the second memory module set each comprise a plurality of memory modules.
3. The chip according to claim 2 , wherein:
any two memory modules in the first memory module set communicate with each other by using the substrate network; and
any two memory modules in the second memory module set communicate with each other by using the substrate network.
4. The chip according to claim 2 , wherein any one memory module in the first memory module set communicates with any one memory module in the second memory module set by using the substrate network.
5. The chip according to claim 2 , wherein the first communications interface and the second communications interface are located on different processor cores.
6. The chip according to claim 1 , wherein a second processor core of the processor is configured to determine, from a plurality of communication paths that are from the second processor core to a second memory module of the first memory module set, a communication path that has a minimum access latency as an access path for accessing the second memory module in the first memory module set.
7. The chip according to claim 6 , wherein the second processor core is configured to:
determine a substrate network latency based on a memory latency and a memory hop count, wherein the memory latency is average time required for data transmission between any two adjacent memory modules in the chip, and the memory hop count is a quantity of memory modules passed during data transmission in the plurality of communication paths that are from the second processor core to the second memory module;
determine a network-on-chip latency based on a core latency and a core hop count, wherein the core latency is average time required for data transmission between any two adjacent processor cores in the processor, and the core hop count is a quantity of processor cores passed during data transmission in the plurality of communication paths that are from the second processor core to the second memory module;
determine, based on the substrate network latency and the network-on-chip latency, access latencies of the plurality of communication paths that are from the second processor core to the second memory module; and
select a communication path that has a minimum access latency from the plurality of communication paths, as the access path.
8. The chip according to claim 7 , wherein the second processor core is further configured to:
determine the substrate network latency based on the memory latency, the memory hop count, and a substrate network load parameter, wherein the substrate network load parameter is used to indicate a load amount of the substrate network.
9. The chip according to claim 7 , wherein the second processor core is further configured to:
determine the network-on-chip latency based on the core latency, the core hop count, and a network-on-chip load parameter, wherein the network-on-chip load parameter is used to indicate a load amount of the network on chip.
10. A computing device, comprising:
a storage;
a chip coupled to the storage and comprising:
a substrate;
a processor, a first memory module set, and a second memory module set that are integrated on the substrate of the chip;
wherein the processor communicates with at least one memory module in the first memory module set by using a first communications interface, and the processor communicates with at least one memory module in the second memory module set by using a second communications interface; and
a memory module in the first memory module set communicates with a memory module in the second memory module set by using a substrate network, wherein the substrate network is a communications network located inside the substrate of the chip, wherein a first processor core of the processor is configured to determine, from a plurality of communication paths that are from the first processor core to a first memory module of the first memory module set, a communication path that has a minimum hop count as an access path for accessing the first memory module in the first memory module set.
11. The computing device according to claim 10 , wherein:
the processor comprises a plurality of processor cores, the plurality of processor cores communicate with each other by using a network on chip, and the network on chip is a communications network located outside the substrate; and
the first memory module set and the second memory module set each comprise a plurality of memory modules.
12. The computing device according to claim 11 , wherein:
any two memory modules in the first memory module set communicate with each other by using the substrate network; and
any two memory modules in the second memory module set communicate with each other by using the substrate network.
13. The computing device according to claim 11 , wherein any one memory module in the first memory module set communicates with any one memory module in the second memory module set by using the substrate network.
14. The computing device according to claim 11 , wherein the first communications interface and the second communications interface are located on different processor cores.
15. The computing device according to claim 10 , wherein a second processor core of the processor is configured to determine, from a plurality of communication paths that are from the second processor core to a second memory module of the first memory module set, a communication path that has a minimum access latency as an access path for accessing the second memory module in the first memory module set.
16. The computing device according to claim 15 , wherein the second processor core is configured to:
determine a substrate network latency based on a memory latency and a memory hop count, wherein the memory latency is average time required for data transmission between any two adjacent memory modules in the chip, and the memory hop count is a quantity of memory modules passed during data transmission in the plurality of communication paths that are from the second processor core to the second memory module;
determine a network-on-chip latency based on a core latency and a core hop count, wherein the core latency is average time required for data transmission between any two adjacent processor cores in the processor, and the core hop count is a quantity of processor cores passed during data transmission in the plurality of communication paths that are from the second processor core to the second memory module;
determine, based on the substrate network latency and the network-on-chip latency, access latencies of the plurality of communication paths that are from the second processor core to the second memory module; and
select a communication path that has a minimum access latency from the plurality of communication paths, as the access path.
17. The computing device according to claim 16 , wherein the second processor core is further configured to:
determine the substrate network latency based on the memory latency, the memory hop count, and a substrate network load parameter, wherein the substrate network load parameter is used to indicate a load amount of the substrate network.
18. The computing device according to claim 16 , wherein the second processor core is further configured to:
determine the network-on-chip latency based on the core latency, the core hop count, and a network-on-chip load parameter, wherein the network-on-chip load parameter is used to indicate a load amount of the network on chip.