US 11,809,978 B2
Neural network scheduling mechanism
Liwei Ma, Beijing (CN); Nadathur Rajagopalan Satish, Santa Clara, CA (US); Jeremy Bottleson, Rancho Cordova, CA (US); Farshad Akhbari, Chandler, AZ (US); Eriko Nurvitadhi, Hillsboro, OR (US); Chandrasekaran Sakthivel, Sunnyvale, CA (US); Barath Lakshmanan, Chandler, AZ (US); Jingyi Jin, Folsom, CA (US); Justin E. Gottschlich, Santa Clara, CA (US); and Michael Strickland, Sunnyvale, CA (US)
Assigned to Intel Corporation, Santa Clara, CA (US)
Filed by Intel Corporation, Santa Clara, CA (US)
Filed on Apr. 18, 2022, as Appl. No. 17/723,074.
Application 17/723,074 is a continuation of application No. 16/918,220, filed on Jul. 1, 2020, granted, now 11,315,007.
Application 16/918,220 is a continuation of application No. 15/482,793, filed on Apr. 9, 2017, granted, now 10,719,760, issued on Jul. 21, 2020.
Prior Publication US 2022/0327357 A1, Oct. 13, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G06N 3/04 (2023.01); G06N 3/044 (2023.01); G06F 9/50 (2006.01); G06N 3/063 (2023.01); G06N 3/084 (2023.01); G06N 3/045 (2023.01)
CPC G06N 3/044 (2023.01) [G06F 9/5038 (2013.01); G06N 3/045 (2023.01); G06N 3/063 (2013.01); G06N 3/084 (2013.01); G06F 2209/5021 (2013.01)] 20 Claims
OG exemplary drawing
 
1. An apparatus to facilitate workload scheduling comprising:
an interconnect fabric; and
one or more general purpose graphics processing units coupled with the interconnect fabric, the one or more general purpose graphics processing units to processes general purpose graphics workloads received from one or more clients via the interconnect fabric, the one or more general purpose graphics processing units including:
hardware resources;
a scheduler to schedule direct access to the hardware resources to process the workloads on behalf of the one or more clients, wherein the one or more clients are each associated with a precompiled neural network (NN) kernel; and
a gather unit to bypass zero data values and gather non-zero data values associated with the one or more clients, the non-zero data values stored sparsely in memory.