hmem-workshop.github.io

4th Workshop on Heterogeneous Memory Systems (HMEM 2023)

In conjunction with SC’23, Denver, CO, USA, November 17th, 2023

Overview and scope

The 4th 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 has become a dominating trend in today’s HPC scene. Different memory technologies are emerging such as: NVM, HBM, Persistent memories, and CXL, that can help mitigate the memory bottleneck by offering opportunities for benefiting applications that process large amounts of data. Such opportunities can 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 keynote speeches, technical paper presentations, and interactive discussions. Overall, topics of interest include, but are not limited to:

Program

November 17th, 2023 (MST time zone)

8:30-8:40 Welcome

8:40-9:30 Keynote: Empowering Large AI Models based on Heterogeneous Memory, Dong Li (University of California, Merced)

9:30-10:00 Persistent snapshot isolation with unlimited reads on commodity hardware transactional memory, Alexandro Baldassin (São Paulo State University (Unesp)), João Barreto, Daniel Castro, Miguel Figueiredo, Paolo Romano (INESC-ID, IST, Universidade de Lisboa) [slides]

10:00-10:30 Break

10:30-11:00 DAOS beyond Persistent Memory: Architecture and Initial Performance Results, Michael Hennecke, Jeff Olivier, Tom Nabarro, Liang Zhen, Yawei Niu, Shilong Wang, Xuezhao Liu (Intel Corporation) [slides]

11:00-11:30 CachedArrays: API and Framework to Optimize Data Movement for Heterogeneous Memory Systems, Mark Hildebrand, Jason Lowe-Power, Venkatesh Akella (University of California, Davis) [slides]

11:30-12:00 Evaluating the latest Optane memory: A glorious swansong?, Adrian Jackson (University of Edinburgh) [slides] [paper]

12:00 Closing Remarks

Submissions

This is a traditional-style workshop without formal proceedings. The authors of accepted submissions will give a talk at the workshop and participate in the closing discussion panel. Additionally, authors will be invited to (optionally) upload their submitted paper (PDF) to be shared on the workshop website. A paper accepted to the HMEM workshop does not preclude its future publication at a major conference.

Submissions must use the ACM proceedings template (for Latex users, version 1.90 (last update April 4, 2023) is the latest template, and please use the “sigconf” option).

We accept two types of submissions.

A first type of submission includes position papers as well as papers that describe completed or early-stage work. Such submissions are limited to 12 pages including references and figures.
Extra pages can be included in a clearly marked appendix (to be read at the discretion of the reviewers). Submitted papers must not include author names (double-blind review).

We also welcome 2-page abstracts that 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.

Submit your paper here: https://submissions.supercomputing.org

Important dates

Time Zone: AOE (Anywhere One Earth)

Organization committee

Program commitee

Acknowledgements

This workshop has received funding from the European High Performance Computing Joint Undertaking (JU) under Framework Partnership Agreement No 800928 and Specific Grant Agreement No 955606 (DEEP-SEA). The JU receives support from the European Union’s Horizon 2020 research and innovation programme and from Croatia, France, Germany, Greece, Italy, Netherlands, Portugal, Spain, Sweden, and Switzerland.

DEEP-SEA