US 10,698,392 B2 · Using graphics processing unit for substrate routing and throughput modeling

Abstract

Methods, systems, and non-transitory computer readable medium for scheduling a substrate processing sequence in an integrated substrate processing system are disclosed herein. A processing device generates a processing model for a batch of semiconductor substrates. The processing model defines a corresponding start time for each semiconductor substrate in each processing chamber in the integrated substrate processing system. One or more graphics processing units (GPUs) generates parallel inputs based on the processing model and concurrently processes, by a plurality of cores of the one or more GPUs, the parallel inputs to generate parallel outputs for the batch of semiconductor substrates. Each of the parallel inputs is processed on a distinct core of the plurality of cores of the one or more GPUs to generate a corresponding parallel output. The processing device causes the batch of semiconductor substrates to be processed in the integrated substrate processing system based on the parallel outputs.

Claims20

Dependent claims

A method of scheduling a semiconductor substrate processing sequence in an integrated substrate processing system, the method comprising:

generating a processing model for a batch of semiconductor substrates, wherein the processing model comprises a matrix and defines a corresponding start time for each semiconductor substrate in each processing chamber in the integrated substrate processing system;
generating parallel inputs based on the processing model, wherein the generating of the parallel inputs comprises generating, by a first graphics processing unit (GPU) of one or more GPUs, a plurality of matrices based on the matrix;
concurrently processing, by a plurality of cores of the one or more GPUs, the parallel inputs to generate parallel outputs for the batch of semiconductor substrates, wherein each of the parallel inputs is processed on a distinct core of the plurality of cores of the one or more GPUs to generate a corresponding parallel output;
and
causing the batch of semiconductor substrates to be processed in the integrated substrate processing system based on the parallel outputs.

2 more elements

A system comprising:

a memory;
and
a processor, coupled to the memory, to generate a processing model for a batch of semiconductor substrates, wherein the processing model comprises a matrix and defines a corresponding start time for each semiconductor substrate in each processing chamber in an integrated substrate processing system;
and
a graphics processing unit (GPU), coupled to the memory and the processor, to generate parallel inputs based on the processing model and concurrently process, by a plurality of cores of the GPU, the parallel inputs to generate parallel outputs for the batch of semiconductor substrates, wherein to generate the parallel inputs, the GPU is to generate a plurality of matrices based on the matrix, wherein each of the parallel inputs is processed on a distinct core of the plurality of cores of the GPU to generate a corresponding parallel output, wherein the processor is to cause the batch of semiconductor substrates to be processed in the integrated substrate processing system based on the parallel outputs.

2 more elements

A non-transitory computer readable medium having instructions stored thereon, which, when executed by a graphics processing unit (GPU), cause the GPU to:

receive a processing model for a batch of semiconductor substrates, wherein the processing model comprises a matrix and defines a corresponding start time for each semiconductor substrate in each processing chamber in an integrated substrate processing system;
generate parallel inputs based on the processing model, wherein to generate the parallel inputs, the GPU is to generate a plurality of matrices based on the matrix;
and
concurrently process, by a plurality of cores of the GPU, the parallel inputs to generate parallel outputs for the batch of semiconductor substrates, wherein each of the parallel inputs is processed on a distinct core of the plurality of cores of the GPU to generate a corresponding parallel output, wherein the batch of semiconductor substrates are to be processed in the integrated substrate processing system based on the parallel outputs.

1 more element

Dependencies are inferred from the claim text.

Using graphics processing unit for substrate routing and throughput modeling

Life of this patent

Using graphics processing unit for substrate routing and throughput modeling

Life of this patent

Abstract

Claims20