Using graphics processing unit for substrate routing and throughput modeling
AssigneeApplied Materials, Inc.
InventorShyam Sunder Emani
Life of this patent
Estimating the term…
Abstract
A method includes receiving a matrix including start times associated with substrate operations in a substrate processing system. The method further includes generating, by a first graphics processing unit (GPU) of one or more GPUs, a plurality of matrices based on the matrix. The method further includes concurrently processing, by a plurality of cores of the one or more GPUs, the plurality of matrices to generate parallel outputs. A schedule for processing substrates in the substrate processing system is to be generated based on the parallel outputs.
Claims20
A method comprising:
- receiving a matrix comprising a plurality of start times associated with substrate operations in a substrate processing system;
- generating, by a first graphics processing unit (GPU) of one or more GPUs, a plurality of matrices based on the matrix;
- and
- concurrently processing, by a plurality of cores of the one or more GPUs, the plurality of matrices to generate parallel outputs, wherein a schedule for processing substrates in the substrate processing system is to be generated based on the parallel outputs.
A system comprising:
- a memory;
- and
- a graphics processing units (GPU), coupled to the memory, wherein the GPU is to:
- receive a matrix comprising a plurality of start times associated with substrate operations in a substrate processing system;
- generate a plurality of matrices based on the matrix;
- and
- concurrently process, by a plurality of cores of the GPU, the plurality of matrices to generate parallel outputs, wherein a schedule for processing substrates in the substrate processing system is to be generated based on the parallel outputs.
A non-transitory computer readable medium having instructions stored thereon, which, when executed by a graphics processing unit (GPU), cause the GPU perform operations comprising:
- receiving a matrix comprising a plurality of start times associated with substrate operations in a substrate processing system;
- generating a plurality of matrices based on the matrix;
- and
- concurrently processing, by a plurality of cores of the GPU, the plurality of matrices to generate parallel outputs, wherein a schedule for processing substrates in the substrate processing system is to be generated based on the parallel outputs.
Dependencies are inferred from the claim text.