|
pdsw 2024:9th International
|
DATE: Sunday, November 17, 2024 |
Time: 9:00 AM - 5:30 pm (EST) |
| Program Co-Chairs: The Ohio State University, USA Illinois Institute of Technology, USA Reproducibility Co-Chairs: Lawerence Berkeley National Laboratory, USA RWTH Aachen University, Germany |
General Chair: Microsoft, USA Publicity Chair: Oak Ridge National Laboratory, USA Web & Publications Chair: Carnegie Mellon University, USA |
abstract / cfp / submissions / WIP session / workshop registration / committees
PDSW24 Reproducability Addendum
SUBMISSION DEADLINE EXTENDED: AUG 9, 2024 - final deadline
Any additional genda information, slides and abstracts will be posted here as soon as it becomes available. You will also be able to view the official agenda on the SC workshop page for the latest information and abstracts for each of the talks at a future date.
| 9am-9:10am | PDSW 2024 Welcome Bing Xie, Microsoft Slides |
| INVITED TALK: | |
| 9:10am- 10am | Invited Speaker: Bridging the Data Gaps in Computing for Science, Education and Society Dr. İlkay Altintaş, University of California, San Diego Slides |
| MAIN SESSION: | |
| 10am- 10:30am | Morning Break |
| 10:30am- 11am | Fault-Tolerant Deep Learning Cache with Hash Ring for Load Balancing in HPC Systems Seoyeong Lee, Sogang University Awais Khan, Oak Ridge National Laboratory (ORNL) Yoochan Kim, Sogang University, South Korea Junghwan Park, Sogang University, South Korea Soon Hwang, Sogang University, South Korea Jae-Kook Lee, Korea Inst of Science and Technology Information (KISTI) Taeyoung Hong, Korea Inst of Science and Technology Information (KISTI) Chris Zimmer, Oak Ridge National Laboratory (ORNL) Youngjae Kim, Sogang University, South Korea Paper | Slides |
| 11am- 11:30am | MOSAIC: Detection and Categorization of I/O Patterns in HPC Applications Théo Jolivel, French Institute for Research in Computer Science and Automation (INRIA) François Tessier, INRIA Julien Monniot, INRIA Guillaume Pallez, INRIA Paper | Slides |
| 11:30am- 12pm | Exploring DAOS Interfaces and Performance Nicolau Manubens Gil, European Centre for Medium-Range Weather Forecasts (ECMWF) Johann Lombardi, DAOS Foundation Simon Smart, ECMWF Emanuele Danovaro ECMWF Tiago Quintino, ECMWF Dean Hildebrand, Google Cloud Adrian Jackson, EPCC, The University of Edinburgh Paper | Slides |
| 12pm- 12:05pm | [WiP] Scalable RPC Layer Towards Millions of IOPS per Server Hiroki Ohtsuji, Fujtisu Limited Munenori Maeda, Fujtisu Limited Reika Kinoshita, Fujtisu Limited Masahiro Miwa, Fujtisu Limited Osamu Tatebe, (University of Tsukuba Abstract | Slides |
| 12:05pm- 12:10pm | [WiP] Reducing I/O Bottleneck for Pretraining AI Foundation Models for Climate Gabriele Padovani, University of Trento, Italy Awais Khan, Oak Ridge National Laboratory, USA Sandro Fiore, University of Trento, Italy Valentine Anantharaj, Oak Ridge National Laboratory, USA Abstract | Slides |
| 12:10pm- 12:15pm | [WiP] BULKI - Binary Unified Layout for Key-value Interchange Wei Zhang, Lawrence Berkeley National Laboratory Houjun Tang, Lawrence Berkeley National Laboratory Suren Byna, The Ohio State University Abstract | Slides |
| 12:15pm- 12:20pm | [WiP] Distributed, Resilient and In-Memory Storage of Key-Value Data for HPC Rüdiger Nather, University of Kassel, Germany Mia Reitz, University of Kassel, Germany Claudia Fohry, University of Kassel, Germany Abstract | Slides |
| 12:20pm- 12:25pm | [WiP] A Global In-Memory Cache and Computation Tier for DAOS John Byrne, Hewlett Packard Enterprise, HPE Clarete Crasta, Hewlett Packard Enterprise, HPE Abhishek Dwaraki, Hewlett Packard Enterprise, HPE David Emberson, Hewlett Packard Enterprise, HPE Harumi Kuno, Hewlett Packard Enterprise, HPE Sekwon Lee, Hewlett Packard Enterprise, HPE Sharad Singhal, Hewlett Packard Enterprise, HPE Ramya Ahobala Rao, Hewlett Packard Enterprise, HPE Shreyas Vinayaka Basri K S, Hewlett Packard Enterprise, HPE Amitha C, Chinmay Ghosh, Hewlett Packard Enterprise, HPE Rishi Kesh Rajak, Hewlett Packard Enterprise, HPE Sriram Ravishankar, Hewlett Packard Enterprise, HPE Porno Shome, Hewlett Packard Enterprise, HPE Lance Evans, Hewlett Packard Enterprise, HPE Sherin George, Hewlett Packard Enterprise, HPE Kevan Rehm, Hewlett Packard Enterprise, HPE Myungjun (MJ) Son, Hewlett Packard Enterprise, HPE Taeklim Kim, Hewlett Packard Enterprise, HPE Shiyue (Jason) Hou, Hewlett Packard Enterprise, HPE Abstract | Slides |
| 12:25pm- 12:30pm | [WiP] Are Streaming Engines and Vector Databases Integrated Well? Yeonwoo Jeong, Sogang University, Republic of Korea Sungyong Park, Sogang University, Republic of Korea Abstract | Slides |
| 12:30pm- 2pm | Lunch Break |
| 2pm-2:30pm | Initial Experiences With DAOS Object Storage on Aurora Rob Latham, Argonne National Laboratory Robert Ross, Argonne National Laboratory Phillip Carns, Argonne National Laboratory Shane Snyder, Argonne National Laboratory Kevin Harms, Argonne National Laboratory Kaushik Velusamy, Argonne National Laboratory Paul Coffman, Argonne National Laboratory Gordon McPheeters, Argonne National Laboratory Paper | Slides |
| 2:30pm-3pm | Understanding and Predicting Cross-Application I/O Interference in HPC Storage Systems Chris Egersdoerfer, University of Delaware Hasanur Rashid, University of Delaware Dong Dai, University of Delaware Bo Fang, Pacific Northwest National Laboratory (PNNL) Nathan Tallent, Pacific Northwest National Laboratory (PNNL) Paper | Slides |
| 3pm-3:30pm | Afternoon Break |
| 3:30pm-4pm | Copper: Cooperative Caching Layer for Scalable Data Loading in Exascale Supercomputers Noah Lewis, Ohio State University Kevin Harms, Argonne National Laboratory Kaushik Velusamy, Argonne National Laboratory Huihuo Zheng, Argonne National Laboratory Paper | Slides |
| 4pm-4:05pm | [WiP] Jarvis: Towards a Shared, User- Friendly, and Reproducible, I/O Infrastructure Jaime Cernuda, Luke Logan, Illinois Institute of Technology Noah Lewis, Illinois Institute of Technology Suren Byna, The Ohio State University Xian-He Sun, The Ohio State University Anthony Kougkas, Illinois Institute of Technology Abstract | Slides |
| 4:05pm- 4:10pm | [WiP] DAOS Project Update - One Year in the DAOS Foundation Michael Hennecke, Intel Corporation Johann Lombardi, DAOS Foundation Abstract | Slides |
| 4:10pm- 4:15pm | [WiP] Improving SQL Query Execution of Distributed Query Engines on Object- Based Computational Storage through Multi-Layere... Soon Hwang, Sogang University, Republic of Korea Junhyeok Park, Sogang University, Republic of Korea Junghyun Ryu, Sogang University, Republic of Korea Jungahn Park, Memory System Research, SK hynix Inc. Jeongjin Lee, Memory System Research, SK hynix Inc. Jungki Noh, Memory System Research, SK hynix Inc. Soonyeal Yang, Memory System Research, SK hynix Inc. Woosuk Chung, Memory System Research, SK hynix Inc. Youngjae Kim, Sogang University, Republic of Korea Abstract | Slides |
| 4:15pm- 4:20pm | [WiP] Lustre for Grace Hopper: Current Status Report Sohei Koyama, DataDirect Networks, Japan Shuichi Ihara, DataDirect Networks, Japan Abstract | Slides |
| 4:20pm- 4:25pm | [WiP] Exploring the Proactive Data Containers Runtime System in VAST - A Case Study Jean Luca Bez (Lawrence Berkeley National Laboratory) Suren Byna (The Ohio State University) Abstract | Slides |
| 4:25pm- 4:30pm | [WiP] Silent Errors to Scientific Applications: Impacts of PFS Metadata Corruptions Dong Dai (University of Delaware) Mai Zheng (Iowa State University) Bo Fang (Pacific Northwest National Laboratory (PNNL) Abstract | Slides |
| 4:30pm- 4:35pm | [WiP] When Stream Processing Engine Meets Log-structured Merge-tree as State Store Kyuli Park, Sogang University, Republic of Korea Sungyong Park, Sogang University, Republic of Korea Abstract | Slides |
| 4:35pm - 5:30pm | Panel: Data, Data Everywhere Moderator: Kathryn Mohror, Lawrence Livermore Lab Panelists: Laura Biven, Jefferson Lab Eli Dart, Lawrence Berkeley National Laboratory Sarp Oral, Oak Ridge National Laboratory Manish Parashar, University of Utah Adam Thompson, NVIDIA |
We are excited to announce the 9th International Parallel Data Systems Workshop (PDSW’24), to be held in conjunction with SC24: The International Conference for High Performance Computing, Networking, Storage, and Analysis, in Atlanta, GA. PDSW’24 builds upon the rich legacy of its predecessor workshops, the Petascale Data Storage Workshop (PDSW, 2006–2015) and the Data Intensive Scalable Computing Systems (DISCS, 2012–2015) workshop. Since their successful merger in 2016, the joint workshop has drawn an average of 200 attendees annually.
The increasing importance of efficient data storage and management continues to drive scientific productivity across traditional simulation-based HPC environments and emerging Cloud, AI/ML, and Big Data analysis frameworks. Challenges are compounded by the rapidly expanding volumes of experimental and observational data, the growing disparity between computational and storage hardware performance, and the rise of novel data-driven algorithms in machine learning. This workshop aims to advance research and development by addressing the most pressing challenges in large-scale data storage and processing.
We invite the community to contribute original research manuscripts that introduce and evaluate novel algorithms or architectures, share significant scientific case studies or workloads, or assess the reproducibility of previously published work. We emphasize the importance of community collaboration for problem identification, workload capture, solution interoperability, standardization, and shared tools. Authors are encouraged to provide comprehensive experimental environment details (software versions, benchmark configurations, etc.) to promote transparency and facilitate collaborative progress.
Topics of Interest:
Call for papers available now (pdf).
All submissions to the PDSW’24 will undergo a rigorous double-anonymous peer review process overseen by the workshop program committee. Successful submissions will be published in the SC24 Workshop Proceedings and featured on the workshop website alongside associated talk slides.
Aligned with the SC24 Reproducibility Initiative, we encourage detailed and structured artifact descriptions (AD) using the SC24 format. The AD should include a field for one or more links to data (zenodo, figshare, etc.) and code (Github, GitLab, Bitbucket, etc.) repositories. For the artifacts that will be placed in the code repository, we encourage authors to follow the PDSW 2024 Reproducibility Addendum on how to structure the artifact, as it will make it easier for the reviewing committee and readers of the paper in the future.
Submissions website: https://submissions.supercomputing.org/
Submissions due: EXTENDED DEADLINE - Aug 9th, 2024, 11:59 PM AoE
AD due: EXTENDED DEADLINE - Aug 16th, 2024, 11:59 PM
AoE
Paper Notification: Sep 6th, 2024, 11:59 PM AoE
Camera ready due: Sep 27th, 2024, 11:59 PM AoE
Final AD/AE due: Oct 15, 2024, 11:59 PM AoE
Copyright forms due: TBD
Slides due before workshop: TBD
The WIP session will showcase brief 5-minute presentations on ongoing work that may not yet be ready for a full paper submission. WIP papers will not be included in the proceedings. A one-page abstract is required for participation.
Submissions due: Sept 13th, 2024, 11:59PM AoE
WIP Notification: On or before Sept 21st, 2024
Registration opens July 10, 2024. To allow you to prepare, find further details on registration pricing, and policies affecting registration changes and cancellations.
