Accelerators and Communications for HPC (AccelCom)

Primary tabs

Our team drives leading-edge research and development around 3 main pillars: (1) Accelerators/coprocessors in HPC, (2) Programmability of heterogeneous memory systems, and (3) Inter-node communications. We collaborate closely with the major accelerator/coprocessor and network vendors for HPC: NVIDIA, Intel, and Mellanox. We run the BSC/UPC NVIDIA GPU Center of Excellence and collaborate with the Intel-BSC Exascale Lab. activities. We organize locally international events such as the PUMPS+AI Summer School, PATC Courses on CUDA/OpenACC and use of heterogeneous memory systems, and the MareNostrum hackathon, and collaborate in the organization of related international conferences and workshops such as HMEM or AsHES.

Objectives

Our broad mission is driving top-notch research and development around accelerators and communications technology for HPC. This can be framed in three main points:

  • Helping vendors advance their hardware architecture and software stack.
  • Designing and developing programming models, runtime systems, and libraries supporting accelerated computing and networking.
  • Developing and porting coprocessor-accelerated and distributed-memory applications.

Group News

[June 2021] AccelCom just released JACC (Just In Time OpenACC framework). Big kudos to its developer Kazuaki Matsumura and his mentor Simon Garcia de Gonzalo!

[June 2021] Don't miss our participation at ISC'21:

[Apr. 2021] Check out our paper "Enabling Homomorphically Encrypted Inference for Large DNN Models" accepted for publication in IEEE Transactions on computers

[Apr. 2021] Congrats to our Team Lead Antonio J. Peña for being named ACM Sr. Member

[Apr. 2021] Don't miss our team's participation in the GPU Technology Conference 2021:

[Dec. 2020] AccelCom in the news: Boosting memory capacity and performance while saving megawatts, by The Next Platform

[Nov. 2020] Congrats to our Postdoc Simon Garcia de Gonzalo for coauthoring SC20's Best Paper

[Oct. 2020] Congrats to our former Researcher Pedro Valero-Lara for receiving an IEEE-CS TCHPC Early Career Researchers Award for Excellence in HPC

[Jul. 2020] AccelCom in the news: Building an ecosystem for heterogeneous memory supercomputing, by The Next Platform

[Jul. 2020] AccelCom in the news: BSC research accelerates HPC workloads with less power-hungry DRAM, by HPCwire
 
 

[Apr. 2020] Antonio J. Peña's interview on the EPEEC project, by insideHPC

[Nov. 2019] Welcome our new Postdoc Simon Garcia de Gonzalo joining on a STARS Fellowship

[Oct. 2018] The EPEEC H2020 Project, led by our Team Lead Antonio J. Peña starts