Copyright (C) 2021 Washington University in St. Louis. All rights reserved.
MERCATOR is licensed under the Apache License, Version 2.0 (the "License"); you may not use this software except in compliance with the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.
Detailed documentation on the use of MERCATOR can be found in the doc subdirectory. For an overview of Mercator's goals and paradigm, please see the references below.
Development of MERCATOR has been sponsored by NSF CISE awards CNS-1500173 and CNS-1763503.
When using MERCATOR, please cite the following publications:
[The full system]
S. V. Cole and J. Buhler. "MERCATOR: A GPGPU framework for irregular streaming applications." 2017 Conference on High Performance Computing & Simulation (HPCS), Genoa, 2017, pp. 727-736.
S. Timcheck and J. Buhler, "Reducing queueing impact in streaming applications with irregular dataflow." Parallel Computing 109:102863, 2022.
[The node scheduling algorithm]
T. Plano and J. Buhler. "Scheduling irregular dataflow pipelines on SIMD architectures." 6th Wkshp. on Programming Models for SIMD/Vector Processing, 1-9, San Diego, CA, Feb 2020.
[The signaling mechanism and region-based state extensions]
S. Timcheck and J. Buhler. "Streaming computations with region-based state on SIMD architectures." 13th Int'l Wkshp. on Programmability and Architectures for Heterogeneous Multicores, 1, Bologna, Italy, Jan 2020.