The 2nd Minisymposium on Applications and Benefits of UPMEM commercial Massively Parallel Processing-In-Memory Platform

August 26, 2024

Full day minisymposium at EURO-PAR 2024

Motivation and scope of the minisymposium

In the rapidly evolving landscape of data-intensive applications across diverse fields such as genomics, analytics, and artificial intelligence (AI), traditional compute-centric architectures are increasingly reaching their limits. The bottleneck often lies in the communication between main memory and CPUs, which is constrained by a narrow bus that suffers from high latency and limited bandwidth, with a significant portion of energy consumption attributed to DRAM data movement. A promising solution to these challenges is the integration of robust computing capabilities directly onto the DRAM memory die, known as Processing-in-Memory (PIM) DRAM.

The field of PIM is experiencing dynamic progress, highlighted by efforts such as SAMSUNG’s HPM-PIM and SK Hynix’s AiM Accelerator. These developments underscore the growing momentum in PIM, although it’s important to note that these products are not yet commercialized in real hardware. In this context, UPMEM stands out as a pioneer with the first commercially available PIM architecture. UPMEM’s PIM modules, which seamlessly integrate in place of standard DIMMs, bring massively parallel computing capabilities to the table. Each DRAM chip is equipped with 8 general purpose processors (DPUs) that provide fast access to DRAM banks. In a standard server configuration, 2560 DPUs can accelerate applications by an order of magnitude.

The ABUMPIMP Symposium provides a unique platform to delve into the use of this cutting-edge technology. It aims to showcase how different applications can use PIM to their advantage and the intricacies involved in developing PIM applications. As we prepare for the second edition of the Symposium, we’re committed to enhancing the Symposium experience. Our program begins with two compelling keynotes. The first provides an insightful overview of the UPMEM architecture and delves into the application development process. While similar to last year’s keynote, it will highlight the innovations in the latest version of our product. The second keynote will take a deep dive into the new design and will provide a glimpse into future versions of the UPMEM products. Building on the success of last year, where we explored a wide range of applications, this year we want to spotlight programmability. We have already lined up three speakers who will cover various aspects of this important topic. In addition, the symposium will serve as a platform to showcase our internal research efforts to accelerate Lucene with our PIM DRAM.

Target audience

This event is a unique opportunity for anyone looking to accelerate data-intensive applications, offering insights from both industrial and academic researchers who have first-hand experience with UPMEM technology. Attendees will leave with a deeper understanding of this technology’s value and practical insights into its application potential.

Tentative agenda

09:00 – 09:15Session welcome and aimsYann FALEVOZ (UPMEM)
09:15 – 10:00Keynote: UPMEM PIM platform for Data-Intensive ApplicationsSylvan BROCARD (UPMEM)
10:00 – 10:30Keynote: UPMEM PIM DRAM new generationCristobal ORTEGA (UPMEM)
10:30 – 11:00Coffee break
11:00 – 11:22uPIMulator: A Flexible and Scalable Simulation Framework for General-Purpose Processing-In-Memory (PIM) ArchitecturesBongjoon Hyun (KAIST)
11:23 – 11:45Invited talk: Processing in Memory VirtualizationDufy TEGUIA (UGA / Orange Innovation) / Jiaxuan CHEN (McGill University)
11:46 – 12:07Research paper: SimplePIM: A Software Framework for Productive and Efficient Processing-in-MemoryGeraldo F. OLIVEIRA (ETHZ)
12:08 – 12:30Research paper: High-level programming abstractions and compilation for near and in-memory computing.Jeronimo CASTRILLON (TU Dresden)
12:30 – 14:00Lunch Break
14:00 – 14:22Research Paper: PID-Comm: A Fast and Flexible Collective Communication Framework for Commodity Processing-in-DIMMsSi Ung Noh (Seoul National University)
14:23 – 14:45Keynote: PIM LuceneSylvan BROCARD (UPMEM)
14:46 – 15:07Research Paper: PIM-tree: A Skew-resistant Index for Processing-in-MemoryHongbo Kang (Tsinghua University)
15:08 – 15:30Research Paper: Enhancing Personalized Recommender Systems with PIM-Rec: Leveraging Processing-In-Memory Technology for Efficient AINiloofar Zarif (University of British Columbia)
15:30 – 16:00Coffee break
16:00 – 16:22Research Paper: BIMSA: Accelerating Long Sequence Alignment Using Processing-In-MemoryAlejandro Alonso-Marín (BSC)
16:23 – 16:45Research Paper: Compression of genomic dataDominique LAVENIER
(Univ. Rennes, CNRS-IRISA & Inria)
16:46 – 17:00ClosingUPMEM


  • Sylvan BROCARD (UPMEM) – Sylvan Brocard is an application engineer at UPMEM. He works on the SDK as well as analytics and machine learning applications. He previously worked on HPC and distributed computing for semiconductor physics simulations at the CEA-Leti and Minatec.
  • Cristobal ORTEGA (UPMEM) – Cristobal Ortega received the Ph.D. degree in computer architecture from the Universitat Politecnica de Catalunya (UPC) in 2022. He is currently a CPU architect at UPMEM. He has worked on past generations of PIM chips and he is now focused on the next generation of the AI-PIM chip.
  • Bongjoon HYUN (KAIST) – Bongjoon Hyun is a PhD student at KAIST. He is majoring in Electrical Engineering and his primary research interests are computer architecture, and hardware/software support for Processing-In-Memory devices under the guidance of Professor Minsoo Rhu.
  • Dufy TEGUIA (UGA / Orange Innovation) – Dufy TEGUIA is a Ph.D. candidate at Université Grenoble Alpes, specializing in Virtualization and virtual machine security. Currently pursuing a CIFRE thesis with Orange Innovation, his research focuses on building a framework for virtual machine observation. With prior experience in distributed hypervisors, including contributions to GiantVM and Scalevisor, Dufy holds a Master’s Degree in Computer Science from École Normale Supérieure de Lyon. His expertise includes virtualizing novel Processing-In-Memory (PIM) hardware for cloud integration, showcasing a strong commitment to advancing virtualization technologies.
  • Jiaxuan CHEN (McGill University) – Jiaxuan Chen is  a first year PhD student in Computer Science at McGill University. His research interests include hardware virtualization, edge computing and computer storage systems.
  • Geraldo F. OLIVEIRA (ETH Zurich) – Geraldo F. Oliveira received a B.S. degree in computer science from the Federal University of Viçosa, Viçosa, Brazil, in 2015, and an M.S. degree in computer science from the Federal University of Rio Grande do Sul, Porto Alegre, Brazil, in 2017. Since 2018, he has been working toward a Ph.D. degree with Onur Mutlu at ETH Zürich, Zürich, Switzerland. His current research interests include system support for processing-in-memory and processing-using-memory architectures, data-centric accelerators for emerging applications, approximate computing, and emerging memory systems for consumer devices. He has several publications on these topics.
  • Jeronimo CASTRILLON (TU Dresden) – Jeronimo Castrillon is a professor in the Department of Computer Science at the TU Dresden, where he is also affiliated with the Center for Advancing Electronics Dresden (CfAED), the Center for Scalable Data Analytics and Artificial Intelligence Dresden/Leipzig (ScaDS.AI), the 6G-life Hub, and the Barkhausen Institut. He is the head of the Chair for Compiler Construction, with research focus on methodologies, languages, tools and algorithms for programming complex computing systems. He received the Electronics Engineering degree from the Pontificia Bolivariana University in Colombia in 2004, his masters degree from the University of Lugano in Switzerland in 2006 and his Ph.D. degree (Dr.-Ing.) with honors from the RWTH Aachen University in Germany in 2013. In 2014, Prof. Castrillon co-founded Silexica GmbH/Inc, a company that provides programming tools for heterogeneous architectures, now with Xilinx/AMD.
  • Si Ung Noh (Seoul National University) – Si Ung Noh is a first year PhD student at Seoul National University, majoring in Electrical Engineering and computer science under the guidance of Professor Jinho Lee. His research interests lie in the area of systems for AI and Processing-In-Memory.
  • Hongbo KANG (Tsinghua University) – Hongbo Kang is a Ph.D. candidate from Tsinghua University. He works on algorithm design for processing-in-memory systems, especially indexes and other memory-intensive components in database management systems.
  • Niloofar ZARIF (University of British Columbia) – Niloofar Zarif is passionate about advancing machine learning efficiency. With a master’s degree from the University of British Columbia, she has hands-on experience with in UPMEM’s Processing-In-Memory technology, aiming to enhance ML performance. With industry experience at NVIDIA and Torc Robotics, she’s skilled in leveraging GPUs and memory for faster ML models. Her research background includes stints at the National University of Singapore and the Sharif University of Technology. Overall, Niloofar is dedicated to making machine learning more efficient and accessible.
  • Alejandro ALONSO-MARIN (Barcelona Supercomputing Center) – Alejandro Alonso-Marín earned his BSc in Computer Engineering from the Universitat Autònoma de Barcelona (UAB) in 2019. Over the next two years, he worked at the ALBA Synchrotron in Barcelona. He completed his MSc in High-Performance Computing at the Universitat Politècnica de Catalunya (UPC) in 2023. Since then, he has been pursuing his PhD in the Computer Science Department at the Barcelona Supercomputing Center (BSC).
  • Dominique LAVENIER (Univ. RennesCNRS-IRISA & Inria) – Dominique Lavenier is a senior CNRS (French National Center for Scientific Research) researcher and a member of the GenScale bioinformatics group at IRISA/Inria, Rennes, he created in 2012. His current research interests include bioinformatics, data structures, genomics, parallelism, processing-in-memory, and DNA storage. He is currently leading the French GenoPIM project whose goal is to investigate the parallelization of the main genomic algorithms on the UPMEM Processing-in-Memory architecture. He also participates in the European BioPIM project.

Workshop organizers

  • Yann FALEVOZ – Tech marketing project manager and product manager, in charge of lab relationship management and collaborative projects at UPMEM.
  • Denis MAKOSHENKO – Software team leader at UPMEM.


For any information, please contact Yann FALEVOZ: yfalevoz@upmem.com