Call for Papers:
OpenSHMEM and Related Technologies 2022 Workshop
Virtual/Online Event Only
Dates: Sep 20-22, 2022
Important Update
Submissions for ORT 2022 have been accepted for an OpenSHMEM
session at
the UCF
2022 workshop. Authors will be contacted individually to ask
if this is suitable for them.
Call for Papers
The OpenSHMEM workshop is an annual event dedicated to the
promotion and advancement of parallel programming with the
OpenSHMEM programming interface and related technologies. It is
the premier venue to discuss and present the latest developments,
implementation technology, tools, trends, recent research ideas
and results related to OpenSHMEM and its use in applications. The
workshop will emphasize OpenSHMEM and related technologies that
are part of its ecosystem such as PGAS languages, communication
frameworks, tools, and application experiences. We will also focus
on programming model extensions to improve the programmability of
systems that use accelerators and smart networks, as well as
extensions that improve interconnect efficiency. Although this is
an OpenSHMEM specific workshop, we welcome ideas used for other
PGAS languages/APIs and additional network programming models that
may be applicable to topics of interest for workshop submissions
include (but are not limited to):
- Potential enhancements to the current OpenSHMEM Specification
- Accelerator-spec version of OpenSHMEM
- Fault Tolerance
- Runtime checking
- Performance analysis tools
- Low-level communications frameworks and libraries
- PGAS languages and emerging programming models
- Julia, Chapel, Rust, etc.
- Utilizing OpenSHMEM/UCX/Communications libraries in Computational Storage
- Scalability or performance issues (of current networks and implementations)
- Advanced memory technologies that impact on PGAS (asymmetric memory)
- Extensions to support non-volatile, high-bandwidth, and other types of memory
- OpenSHMEM/PGAS and their value to ML/AI/Big Data/(I/O) support
- Applications experiences with OpenSHMEM, Rust and PGAS languages
- Experiences with applications from any domain, especially dynamic and irregular applications
- Hybrid programming models: OpenSHMEM combined with task-based models (e.g. OCR, HPX, ParSEC) or heterogeneous models (e.g. OpenCL, OpenACC, CUDA, OpenMP)
- Experiences implementing OpenSHMEM on new architectures
- Power / Energy studies that improves network utilization
- Validation and verification suites for OpenSHMEM and network software stacks
- Auto-tuning or optimization strategies
- Integration with runtime environments and schedulers
- Benchmarks (e.g. OpenHPCA), performance evaluation, and validation suites
- Unified Communication X (UCX), UCX-Py, UCX-Java, UCX-Go
- Unified Communication Collectives (UCC)
- Data Processing Units (DPUs) / SmartNIC APIs
- Machine Learning and data sciences frameworks implemented on top of UCX
- Integration with Apache Spark, Apache Arrow, BlazingSQL, Dask/RAPIDS, etc.
- Network offloading of scientific libraries, FFTs, etc.
Important Dates
"OpenSHMEM and Related Technologies" will take place during
the UCF
2022 workshop, Sep 20-22, 2022.
Paper Submission Details
The proceedings will be published in a LNCS Springer volume and
will be peer-reviewed by the program committee. All authors must
first submit a 250-word abstract to register their papers. Once
the abstract is accepted, we will encourage the authors to submit
full or short papers. We will accept full papers of 12-15 pages
and short papers of 6-8 pages. Preference will be given to full
papers. Papers need to be formatted according to s single-column
style. Please use the paper templates available for LaTeX and Word
(https://www.springer.com/gp/authors-editors/conference-proceedings/conference-proceedings-guidelines). The
copyright will need to be transferred to Springer. A copyright
form will be provided to authors which allows users to
self-archive.
Abstracts and papers should be uploaded here:
https://easychair.org/conferences/?conf=openshmem2022.
General Chair
Steering Programming Committee
- Oscar Hernandez, NVIDIA
- Pavel Shamis, ARM
- Mathew Baker, Voltron Data
- Manjunath Gorentla Venkata, NVIDIA
- Michael Raymond, HPE
- Bryant Lam, DoD
- Nick Park, DoD
Technical Committee Members
- Pavel Shamis, ARM, Chair
- Naveen Namashivayam, HPE
- Michael Raymond, HPE
- James Dinan, NVIDIA
- Gilad Shainer, NVIDIA
- Manjunath Gorentla Venkata, NVIDIA
- Khaled Hamidouche, AMD
- Camile Cotte, UoO/Paris
- Sameer Shende, UoO
- Brody Williams, TTU
- Matthew Baker, Voltron Data
- Swen Boehm, ORNL
- Oscar Hernandez, NVIDIA, Co-chair
- Swaroop Pophale, ORNL
- Min Si, Meta
- Dhabaleswar Panda, OSU
- Tony Curtis, SBU
- Pavan Balaji, Meta
- Stephen Poole, LANL
- Howard Pritchard, LANL
- Wendy Poole, LANL
- Nick Park, DoD
- Bryant Lam, DoD
- Marcel Fallet, DoD
- Thomas Rolinger, DoD
- Curtis Hughey, DoD
- Jessica Steffy, DoD
- Alex Margolin, NextSilicon
- Max Grossman, GTRC
- Mitsuhisa Sato, Riken
- Jeff Kuehn, AMD
- Yangei Guo, ANL
- Mathew Dosanjh, SNL
- Perry Schmidt, IBM
Logistics
- Web Chair
- Valerie Hartman, LANL
- Beth Kaspar, LANL