Nvidia Corporation

Prakash Bangalore Prabhakar

Shirish C Gadre

Jack H Choquette

Lacky V Shah

John H Edmondson

Ronny Krashinsky

Manan Patel

Ravi Kiran Manyam

Wish Gandhi

Apoorv Parle

Steve Heinrich

Jerome Francis Duluk, Jr

James M Van Dyke

Olivier Giroux

Luke Durant

Gentaro Hirota

Brian S Pharris

Jeff Tuckey

Ze Long

Kun Fang

Rajballav Dash

Greg Palmer

Sandeep Navada

Eric Werness

Naman Govil

Shayani Deb

Alexander Lev Minkin

A new level(s) of hierarchy—Cooperate Group Arrays (CGAs)—and an associated new hardware-based work distribution/execution model is described. A CGA is a grid of thread blocks (also referred to as cooperative thread arrays (CTAs)). CGAs provide co-scheduling, e.g., control over where CTAs are placed/executed in a processor (such as a GPU), relative to the memory required by an application and relative to each other. Hardware support for such CGAs guarantees concurrency and enables applications to see more data locality, reduced latency, and better synchronization between all the threads in tightly cooperating collections of CTAs programmably distributed across different (e.g., hierarchical) hardware domains or partitions.

Cooperative group arrays

Distributed shared memory (DSMEM) comprises blocks of memory that are distributed or scattered across a processor (such as a GPU). Threads executing on a processing core local to one memory block are able to access a memory block local to a different processing core. In one embodiment, shared access to these DSMEM allocations distributed across a collection of processing cores is implemented by communications between the processing cores. Such distributed shared memory provides very low latency memory access for processing cores located in proximity to the memory blocks, and also provides a way for more distant processing cores to also access the memory blocks in a manner and using interconnects that do not interfere with the processing cores' access to main or global memory such as hacked by an L2 cache. Such distributed shared memory supports cooperative parallelism and strong scaling across multiple processing cores by permitting data sharing and communications previously possible only within the same processing core.

Distributed shared memory

This specification describes a programmatic multicast technique enabling one thread (for example, in a cooperative group array (CGA) on a GPU) to request data on behalf of one or more other threads (for example, executing on respective processor cores of the GPU). The multicast is supported by tracking circuitry that interfaces between multicast requests received from processor cores and the available memory. The multicast is designed to reduce cache (for example, layer 2 cache) bandwidth utilization enabling strong scaling and smaller tile sizes.

Programmatically controlled data multicasting across multiple compute engines

A system in having M memory controllers between a first memory and a second memory having N operative memory slices, where N and M are not evenly divisible, includes logic to operate the M memory controllers to linearly distribute addresses of the second memory across the N operative memory slices. The system may be utilized in commercial applications such as data centers, autonomous vehicles, and machine learning.

Addressing cache slices in a last level cache

An addressing scheme in systems utilizing a number of operative memory slices in a last level cache that is not evenly divisible by a number of memory channels utilizes the operative slices exposes the full last level cache bandwidth and capacity to data processing logic in a high-performance graphics system.

Growing community of inventors

Santa Clara, CA, United States of America

Prakash Bangalore Prabhakar