The 5th HMEM workshop serves as a forum to present and discuss ongoing research around heterogeneous memory systems. The scope of the workshop encompasses all the layers of system and software stack, from computer architectures, operating system, middleware, programming models, runtime systems, tools, to applications.
Heterogeneous memory infrastructure design is becoming an emerging trend in today’s HPC scene. Different memory technologies are maturing, such as HBM, NVM/persistent memories, and also emerging, such as those based on CXL. These aim at helping mitigate the memory bottleneck by offering opportunities for benefiting applications that process large amounts of data, e.g., offering small but fast memory subsystems to place frequently accessed data structures. Such opportunities may be realized across all the layers of system and software stack: from computer architectures, operating system, middleware, programming models, runtime systems, tools, up to applications.
As in previous years, the Workshop on Heterogeneous Memory Systems (HMEM) will bring together different research efforts and expertise to the end of integrating different approaches and democratizing the use of heterogeneous memory systems to benefit applications not only in terms of performance, but also energy efficiency and cost tradeoffs. The main goal of the workshop is to push the research frontiers forward by exchanging knowledge and debating ideas through featured talks, technical paper presentations, and interactive discussions. Overall, topics of interest include, but are not limited to:
September 24th, 2024
15:00 Welcome
15:00-15:30 Understanding Aurora’s Heterogeneous Memory Architecture, Brice Videau, Argonne National Laboratory, USA
15:30-16:00 Understanding the Composability of Heterogeneous Memory for Workload-Optimized System Design, Sudharshan Vazhkudai, Micron, USA
16:00-16:30 Embracing Heterogeneous Memory Systems in HPC and Cloud: from Unified to Disaggregated, Ivy Peng, KTH Royal Institute of Technology, Sweden
16:30-16:45 Coffee break
16:45-17:15 Enabling cooperation between object and page management for improved tiering on heterogeneous memory systems, Maciej Maciejewski, Huawei, Poland
17:15-17:45 A couple use cases for heterogeneous memory systems from the Barcelona Supercomputing Center, Antonio Peña, Barcelona Supercomputing Center, Spain
17:45-18:15 H2M: Heuristics for Heterogeneous Memory, Clément Foyer, Université de Reims Champagne-Ardenne, France
The authors of accepted submissions will give a talk at the workshop and participate in the closing discussion panel.
The accepted submissions will be included in the IEEE Cluster 2024 proceedings. The authors of accepted submissions can also opt out of publication, and only give a talk.
Submissions must use the template for the IEEE Cluster 2024 proceedings. Three kinds of submissions are possible:
Full papers: 8 pages + 2 additional pages (including references), must not include author names (double-blind review).
Short papers: 4 pages + 2 additional pages (including references), must not include author names (double-blind review).
Abstracts of previous publications: 2 pages, which summarize recently accepted/published at top-tier conferences/journals. In this case, the author names and references to the published works should be included in the abstract. This kind of submission will not be published in the proceedings.
Submit your paper here: submission link
The use of content generated by artificial intelligence (AI) in a paper (including but not limited to text, figures, images, and code) shall be disclosed in the acknowledgments section of any paper submitted to an IEEE publication. The AI system used shall be identified, and specific sections of the paper that use AI-generated content shall be identified and accompanied by a brief explanation regarding the level at which the AI system was used to generate the content. The use of AI systems for editing and grammar enhancement is common practice and, as such, is generally outside the intent of the above policy. In this case, disclosure as noted above is recommended.
Time Zone: AOE (Anywhere One Earth)
[to be completed]