Proximity communication-based off-chip cache memory architectures
View Patent ↗A proximity interconnect module includes a plurality of off-chip cache memories. Either disposed external to the proximity interconnect module or on the proximity interconnect module are a plurality of processors that are dependent on the plurality of off-chip cache memories for servicing requests for data. The plurality of off-chip cache memories are operatively connected to either one another or to one or more of the plurality of processors by proximity communication. Each of the plurality of off-chip cache memories may cache certain portions of the physical address space.
1. A system, comprising:
a first processor;
a first cache overlapping a first corner of the first processor;
a second cache overlapping a second corner of the first processor;
a third cache overlapping a third corner of the first processor;
a fourth cache overlapping a fourth corner of the first processor, wherein the first cache, the second cache, the third cache, and the fourth cache belong to a plurality of off-chip cache memories and are operatively connected to the first processor by proximity communication; and
a physical memory comprising a physical address space,
wherein the first cache is configured to cache data for the first processor only from a first portion of the physical address space,
wherein the second cache is configured to cache data for the first processor only from a second portion of the physical address space,
wherein the third cache is configured to cache data for the first processor only from a third portion of the physical address space, and
wherein the fourth cache is configured to cache data for the first processor only from a fourth portion of the physical address space.
2. The system of claim 1 , further comprising:
a proximity interconnect module, wherein the first processor and the plurality of off-chip cache memories are disposed on the proximity interconnect module.
3. The system of claim 1 , further comprising:
a proximity interconnect module, wherein the first processor is disposed external to the proximity interconnect module, wherein the proximity interconnect module includes functionality to cause the first processor to communicate with at least one of the plurality of off-chip cache memories.
4. The system of claim 3 , wherein at least one of the plurality of off-chip cache memories is disposed on the proximity interconnect module.
5. The system of claim 1 , wherein the proximity communication comprises at least one of capacitive coupling, inductive coupling, magnetic coupling, and optical coupling.
6. The system of claim 1 , wherein the first portion, the second portion, the third portion, and the fourth portion are each one-fourth of the physical address space.
7. The system of claim 1 , further comprising:
a second processor;
a fifth cache overlapping a first corner of the second processor; and
a sixth cache overlapping a second corner of the second processor,
wherein the fifth cache and the sixth cache belong to the plurality of off-chip cache memories and are operatively connected by proximity communication to the second processor,
wherein the third cache further overlaps a third corner of the second processor,
wherein the fourth cache further overlaps a fourth corner of the second processor,
wherein the third cache and the fourth cache are further operatively connected by proximity communication to the second processor,
wherein the fifth cache is configured to cache data for the second processor only from the first portion of the physical address space,
wherein the sixth cache is configured to cache data for the second processor only from the second portion of the physical address space,
wherein the third cache is further configured to cache data for the second processor only from the third portion of the physical address space, and
wherein the fourth cache is further configured to cache data for the second processor only from the fourth portion of the physical address space.
8. A system, comprising:
a first processor;
a proximity interconnect module operatively connected to main memory, wherein the main memory comprises a physical address space;
a first cache overlapping a first corner of the first processor;
a second cache overlapping a second corner of the first processor;
a third cache overlapping a third corner of the first processor; and
a fourth cache overlapping a fourth corner of the first processor,
wherein the first cache, the second cache, the third cache, and the fourth cache belong to a plurality of off-chip cache memories disposed on the proximity interconnect module,
wherein the first cache is configured to cache data for the first processor only from a first portion of the physical address space,
wherein the second cache is configured to cache data for the first processor only from a second portion of the physical address space,
wherein the third cache is configured to cache data for the first processor only from a third portion of the physical address space, and
wherein the fourth cache is configured to cache data for the first processor only from a fourth portion of the physical address space.
9. The system of claim 8 , wherein the first processor is disposed on the proximity interconnect module.
10. The system of claim 8 , wherein the proximity communication comprises at least one of capacitive coupling, inductive coupling, magnetic coupling, and optical coupling.
11. The system of claim 8 , wherein the first portion, the second portion, the third portion, and the fourth portion are each one-fourth of the physical address space.
12. The system of claim 8 , further comprising
a second processor;
a fifth cache overlapping a first corner of the second processor; and
a sixth cache overlapping a second corner of the second processor,
wherein the fifth cache and the sixth cache belong to the plurality of off-chip cache memories and are operatively connected by proximity communication to the second processor,
wherein the third cache further overlaps a third corner of the second processor,
wherein the fourth cache further overlaps a fourth corner of the second processor,
wherein the third cache and the fourth cache are further operatively connected by proximity communication to the second processor,
wherein the fifth cache is configured to cache data for the second processor only from the first portion of the physical address space,
wherein the sixth cache is configured to cache data for the second processor only from the second portion of the physical address space,
wherein the third cache is further configured to cache data for the second processor only from the third portion of the physical address space, and
wherein the fourth cache is further configured to cache data for the second processor only from the fourth portion of the physical address space.