A public copy of things I've worked on.
Y. H. Temuçin, A. B. Sedeh, W. Schonbein, R. Grant, and A. Afsahi, "Utilizing Network Hardware Parallelism for MPI Partitioned Collective Communication", 33rd Euromicro International Conference on Parallel, Distributed, and Network-Based Processing (PDP 2025), Turin, Italy, Mar 12 - 14, 2025
Y. H. Temuçin, M. Gazimirsaeed, R. E. Grant and A. Afsahi, "ROCm-Aware Leader-based Designs for MPI Neighbourhood Collectives" in ISC High Performance 2024 Research Paper Proceedings (39th International Conference), Hamburg, Germany, 2024 pp. 1-12. doi: 10.23919/ISC.2024.10528923
Y. H. Temuçin, S. Levy, W. Schonbein, R. Grant and A. Afsahi, "A Dynamic Network-Native MPI Partitioned Aggregation Over InfiniBand Verbs", in 2023 IEEE International Conference on Cluster Computing (CLUSTER), Santa Fe, NM, USA, 2023 pp. 259-270. doi: 10.1109/CLUSTER52292.2023.00029 (Best Paper Award)
P. Alizadeh, A. Sojoodi, Y. H. Temuçin, and A Afsahi, "Efficient Process Arrival Pattern Aware Collective Communication for Deep Learning", in Proceedings of the 29th European MPI Users' Group Meeting (EuroMPI/USA'22), 68–78. doi: 10.1145/3555819.3555857
Y. H. Temuçin, R. E. Grant, and A. Afsahi, "Micro-Benchmarking MPI Partitioned Point-to-Point Communication", in Proceedings of the 51st International Conference on Parallel Processing (ICPP '22), Article 64, 1–12. doi: 10.1145/3545008.3545088
[paper] [slides] [bibtex] [code]
Y. H. Temuçin, A. Sojoodi, P. Alizadeh and A. Afsahi, "Efficient Multi-Path NVLink/PCIe-Aware UCX based Collective Communication for Deep Learning", 2021 IEEE Symposium on High-Performance Interconnects (HOTI), Santa Clara, CA, USA, 2021, pp. 25-34, doi: 10.1109/HOTI52880.2021.00018
Y. H. Temuçin, W. Schonbein, S. Levy, A. Sojoodi, R. Grant and A. Afsahi, "Design and Implementation of MPI-Native GPU-Initiated MPI Partitioned Communication" in 12th Workshop on Extreme Scale MPI (ExaMPI), Atlanta, GA, USA, Nov. 17, 2024.
A. Sojoodi, Y. H. Temuçin, and A. Afsahi, "Enhancing Intra-node GPU-to-GPU Performance in MPI+UCX through Multi-Path Communication", 3rd International Workshop on Extreme Heterogeneity Solutions (ExHET) 2024, Edinburgh, UK, Mar 3, 2024. ACM (Best Paper Award)
Y. H. Temuçin, A. Sojoodi, P. Alizadeh, B. Kitor and A. Afsahi, "Accelerating Deep Learning Using Interconnect-Aware UCX Communication for MPI Collectives", in IEEE Micro, vol. 42, no. 2, pp. 68-76, 1 March-April 2022, doi: 10.1109/MM.2022.3148670.
Y. H. Temuçin, "Benchmarking MPI for Deep Learning and HPC Workloads", in Benchmarking in the Data Center: Expanding to the Cloud (BID), workshop held in conjunction with Principles and Practice of Parallel Programming (PPoPP), Montréal, Canada, Feb. 2023