Demo Data from EDBT 2026 / reviewers in the wild / expert
Samuel Wasmundt
dblp:189/1199
· DBLP profile ↗
2ranked-venue papers
0as first author
1since 2021 · last 2023
0000-0002-2021-1329ORCID · corroborated
Domains — the database's venue-derived domains; a paper can count in several
Systems, architecture and hardware · 2 · 1 since 2021Software engineering, systems software and programming languages · 1 · 1 since 2021
Expertise — from the expertise taxonomy: the topics of the expert's papers under the CCF categories. A weight counts papers with recency: 1 for a paper about the topic, 0.3 when the topic is its context, halved every five years.
| Computer architecture, parallel and distributed computing, and storage systems
2 papers |
High-performance computing · 55% GPUs and heterogeneous computing · 15% Interconnection networks and networks-on-chip · 15% |
Topics — the 5 heaviest of 6, each with the papers that count most for it
| Topic | Weight | Papers | Last | Evidence papers |
|---|---|---|---|---|
High-performance computing › supercomputing
exascale computing |
0.7 | 1 | 2023 | A Research Retrospective on AMD's Exascale Computing Journey · ISCA 2023 |
Parallel and multicore computing › parallel programming runtimes
active messages |
0.2 | 1 | 2016 | Extended task queuing: active messages for heterogeneous systems · SC 2016 |
Interconnection networks and networks-on-chip
remote direct memory access |
0.2 | 1 | 2016 | Extended task queuing: active messages for heterogeneous systems · SC 2016 |
High-performance computing
supercomputing |
0.2 | 1 | 2023 | A Research Retrospective on AMD's Exascale Computing Journey · ISCA 2023 |
High-performance computing › collective communication
MPI collective communication |
0.1 | 1 | 2016 | Extended task queuing: active messages for heterogeneous systems · SC 2016 |
Methods — techniques the papers use, named apart from their topics
task queuing · 0.2active messages · 0.2RDMA · 0.2
| Year | Publication | Venue | Position |
|---|---|---|---|
| 2023 | A Research Retrospective on AMD's Exascale Computing JourneyabstractThe pace of advancement of the top-end supercomputers historically followed an exponential curve similar to (and driven in part by) Moore's Law. Shortly after hitting the petaflop mark, the community started looking ahead to the next milestone: Exascale. However, many obstacles were already looming on the horizon, such as the slowing of Moore's Law, and others like the end of Dennard Scaling had already arrived. Anticipating significant challenges for the overall high-performance computing (HPC) community to achieve the next 1000x improvement, the U.S. Department of Energy (DOE) launched the Exascale Computing Program to enable and accelerate fundamental research across the many technologies needed to achieve exascale computing. Gabriel H. Loh, Michael J. Schulte, Mike Ignatowski, Vignesh Adhinarayanan, Shaizeen Aga, Derrick Aguren, Varun Agrawal, Ashwin M. Aji, Johnathan Alsop, Paul T. Bauman, Bradford M. Beckmann, Majed Valad Beigi, Sergey Blagodurov, Travis Boraten, Michael Boyer, William C. Brantley, Noel Chalmers, Shaoming Chen, Michael L. Chu, David Cownie, Nicholas Curtis, Joris Del Pino, Nam Duong, Alexandru Dutu, Yasuko Eckert, Christopher Erb, Chip Freitag, Joseph L. Greathouse, Sudhanva Gurumurthi, Anthony Gutierrez, Khaled Hamidouche, Sachin Hossamani, Wei Huang 0004, Mahzabeen Islam, Nuwan Jayasena, John Kalamatianos, Onur Kayiran, Jagadish Kotra, Alan Lee, Daniel Lowell, Niti Madan, Abhinandan Majumdar, Nicholas Malaya, Srilatha Manne, Susumu Mashimo, Damon McDougall, Elliot Mednick, Michael Mishkin, Mark Nutter, Indrani Paul, Matthew Poremba, Brandon Potter, Kishore Punniyamurthy, Sooraj Puthoor, Steven E. Raasch, Karthik Rao, Gregory Rodgers, Marko Scrbak, Mohammad Seyedzadeh, John Slice, Vilas Sridharan, René van Oostrum, Eric Van Tassell, Abhinav Vishnu, Samuel Wasmundt, Mark Wilkening, Noah Wolfe, Mark Wyse, Adithya Yalavarti, Dmitri Yudanov |
ISCA | 66 |
| 2016 | Extended task queuing: active messages for heterogeneous systemsabstractAccelerators have emerged as an important component of modern cloud, datacenter, and HPC computing environments. However, launching tasks on remote accelerators across a network remains unwieldy, forcing programmers to send data in large chunks to amortize the transfer and launch overhead. By combining advances in intra-node accelerator unification with one-sided Remote Direct Memory Access (RDMA) communication primitives, it is possible to efficiently implement lightweight tasking across distributed-memory systems. This paper introduces Extended Task Queuing (XTQ), an RDMA-based active messaging mechanism for accelerators in distributed-memory systems. XTQ's direct NIC-to-accelerator communication decreases inter-node GPU task launch latency by 10-15% for small-to-medium sized messages and ameliorates CPU message servicing overheads. These benefits are shown in the context of MPI accumulate, reduce, and allreduce operations with up to 64 nodes. Finally, we illustrate how XTQ can improve the performance of popular deep learning workloads implemented in the Computational Network Toolkit (CNTK). Michael LeBeane, Brandon Potter, Abhisek Pan, Alexandru Dutu, Vinay Agarwala, Wonchan Lee, Deepak Majeti, Bibek Ghimire, Eric Van Tassell, Samuel Wasmundt, Brad Benton, Maurício Breternitz, Michael L. Chu, Mithuna Thottethodi, Lizy Kurian John, Steven K. Reinhardt |
SC | 10 |