Kevin Sala joined the Center for Applied Scientific Computing (CASC) in November 2024 as a postdoctoral researcher. His research focuses on compiler and runtime support for accelerating the execution of code GPUs, particularly in the HPC context. He has also been involved in the LLVM/OpenMP offload runtime since 2022 and the OpenMP language standard since 2024.

Kevin is passionate about parallel programming models, parallel architectures, and HPC in general. As computing systems continue to grow in complexity and capability, he believes it's essential to provide the HPC community with robust, user-friendly software solutions to fully leverage available computational resources.

He received his Ph.D. in Computer Architecture from the Universitat Politècnica de Catalunya (UPC) in 2024 in Barcelona, Spain. During his Ph.D., also working in the Barcelona Supercomputing Center (BSC), he focused on improving the performance and interoperability between distributed-memory programming models and shared-memory tasking models in hybrid applications (e.g., MPI+OpenMP).

Ph.D. Computer Architecture, Universitat Politècnica de Catalunya (UPC), Barcelona, Spain

Master’s Degree Innovation and Research in Informatics, Universitat Politècnica de Catalunya (UPC), Barcelona, Spain

Bachelor’s Degree Computer Engineering, Universitat Politècnica de Catalunya (UPC), Barcelona, Spain

Sala, K., Álvarez, D., Peñacoba, R., Mallo, R.A., Navarro, A.C., Roca, A., Beltran, V.: Alpi: Enhancing portability and interoperability of task-aware libraries. In: WAMTA. pp. 142–153 (2024)

Sala, K., Macià, S., Beltran, V.: Combining one-sided communications with task-based programming models. In: 2021 IEEE International Conference on Cluster Computing (CLUSTER). pp. 528–541. IEEE (2021)

Sala, K., Rico, A., Beltran, V.: Towards data-flow parallelization for adaptive mesh refinement applications. In: 2020 IEEE International Conference on Cluster Computing (CLUSTER). pp. 314–325. IEEE (2020)

Sala, K., Teruel, X., Perez, J.M., Peña, A.J., Beltran, V., Labarta, J.: Integrating blocking and non-blocking mpi primitives with task-based programming models. Parallel Computing 85, 153–166 (2019)

Sala, K., Bellón, J., Farré, P., Teruel, X., Perez, J.M., Peña, A.J., Holmes, D., Beltran, V., Labarta, J.: Improving the interoperability between mpi and task-based programming models. In: Proceedings of the 25th European MPI Users’ Group Meeting. p. 6. ACM (2018)

Álvarez, D., Sala, K., Beltran, V.: nos-v: Co-executing hpc applications using system-wide task scheduling. In: 2024 IEEE International Parallel and Distributed Processing Symposium (IPDPS). pp. 312–324. IEEE (2024)

Álvarez, D., Sala, K., Maroñas, M., Roca, A., Beltran, V.: Advanced synchronization techniques for task-based runtime systems. In: Proceedings of the 26th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming. pp. 334–347 (2021)

Maroñas, M., Sala, K., Mateo, S., Ayguadé, E., Beltran, V.: Worksharing tasks: An efficient way to exploit irregular and fine-grained loop parallelism. In: 2019 IEEE 26th International Conference on High Performance Computing, Data, and Analytics (HiPC). pp. 383–394. IEEE (2019)

Castelló, A., Mayo, R., Sala, K., Beltran, V., Balaji, P., Peña, A.J.: On the adequacy of lightweight thread approaches for high-level parallel programming models. Future Generation Computer Systems (2018)