HPCW 2019: 5th High Performance Containers Workshop - Call For Papers
In conjunction with ISC High Performance 2019, Frankfurt, Germany
Linux Containers continue to gain momentum within data centers all over the world. They are able to benefit legacy infrastructures by leveraging the lower overhead compared to traditional, hypervisor-based virtualization. But there is more to Linux Containers, which this workshop will explore. Their portability, reproducibility and distribution capabilities outclass all prior technologies and disrupt former monolithic architectures, due to sub-second life cycles and self-service provisioning.
This workshop will outline the current state of Linux Containers in HPC/AI, what challenges are hindering the adoption in HPC/BigData and how containers can foster improvements when applied to the field of HPC, Big Data and AI in the mid- and long-term. By dissecting the different layers within the container ecosystem (runtime, supervision, engine, orchestration, distribution, security, scalability) this workshop will provide a holistic and a state-of-the-container overview, so that participants can make informed discussions on how to start, improve or continue their container adoption.
The workshop will follow the concepts of the last 4 successful workshops at ISC by inviting leading members within the container industry (Mellanox, UberCloud, rCUDA, NVIDIA, Docker, Singularity/Sylabs), end-users from Bioinformatics labs, Enterprise Performance Computing and participants of the Student cluster competition.
The workshop will solicit research papers from the broader research community, reviewed by renowned container experts within HPC/AI, and will comprise of invited speakers and research paper slots, with the general goal to provide a holistic view of the containerized ecosystem and compare different parts within this ecosystem with regards to what the provide and spark panel discussions about highly debated topics.
Submitted papers must be written in English, following the Springer LNCS Style for conference proceedings. Authors are requested to first register their submissions and then submit their manuscripts in PDF format at the Easychair page. The workshop has two submission tracks:
Full paper submissions should be structured as technical papers (6-12 pages). Submitted papers must represent original unpublished research that is not currently under review for any other conference or journal. The submission is open to academic as well as industrial authors.
Short Industrial papers
We encourage submissions on innovative solutions and applications related to commercial or industrial-strength software. At least one of the authors needs to have a non-academic affiliation. Authors affiliated to technical departments within academic institutions will be considered non-academic. Industrial submissions can be 4-8 pages. Camera- ready versions of accepted industrial papers need to be at least 6 pages in length in order to be included in the proceedings.
All submitted manuscripts should include author names and affiliations. Authors of accepted papers will be requested to provide source files.
Accepted ISC 2019 workshop papers will be published in the Springer’s Lecture Notes in Computer Science (LNCS) series.
The proceedings will appear as post-conference workshop proceedings. The idea behind post-conference proceedings is that they give the authors the flexibility to adjust their contributions slightly based on the feedback given during the workshops.
Abstract submission deadline:
April 20(extended to May 02)
Paper submission deadline:
April 25(extended to June 02 for industrial papers)
Author notification: May 08
- Conference-ready deadline: June 10 (To be shared during the workshop)
- Workshop Day: June 20
- Camera-ready deadline: July 22
Topics on Interest
Topics of interest include but are not limited to:
- Provisioning and building containers for HPC systems.
- Building portable and hardware optimized containers
- Pre-build container images in contrast to runtime decisions on what specific system to run
- Security of provenance and the build process; building from “untrusted” sources
- Efficient management and distribution of images across large number of nodes
- Container/image access control
- Distribution and transport security
- Container deployment, monitoring, and automation
- Performance benchmarking of containerised workloads
- Containers inside lightweight VMs
- Comparison of different container platforms in terms of security and performance
- Container runtime environments
- Container operating systems
HPC Ecosystem specifics
- Host agnostic vs host specific
- Data, network, and hardware access
- Providing a secure, easy to use RDMA networking to accelerate MPI applications
- Container isolation
- Container access control
- Container audits and compliance
- Container vulnerability scanning
- Application whitelisting
- Security best practices
- Scheduling and orchestration of container workloads
- Integration with existing HPC/HTC schedulers, e.g. Slurm and HTCondor
- Eligibility of container orchestration platforms (e.g. Kubernetes and Swarm) to manage HPC workloads
- Containers in Big Data, Modeling and Simulation, Life Sciences and AI/ML applications
- Containers for sensitive data analysis
- Containerised RDMA, GPU, and MPI applications
- High availability systems for containerized distributed workloads
- Auxiliary Services tied to HPC/AI workflows
- IoT/Edge computing if related to HPC/AI workloads
- Containers in cloud computing
Other related topic
- The impact of containers on software development in HPC and technical computing.
- Monitoring (logs, events, metrics) and auditing in the area of containers
- Other topics relevant to containers in HPC
Christian Kniep, Docker Inc.
Program and Publications Chairs
Abdulrahman Azab, University of Oslo, Norway / Partnership for Advanced Computing in Europe (PRACE) / Nordic e-Infrastructure Collaboration (NeIC)
Shane Canon, National Energy Research Scientific Computing Center (NERSC), USA
Technical Program Committee (tentative)
Thomas Röblitz, University of Oslo, Norway
Björn Grüning, University of Freiburg, Germany
Bob Killen, Lawrence Livermore National Laboratory (LLNL), USA
Burak Yenier, UberCloud, USA
Wolfgang Gentzsch, UberCloud, USA
CJ Newburn, NVIDIA, USA
Carlos E. Arango, Sylabs, USA
Michael W. Bauer, Sylabs, USA
Giuseppa Muscianisi, CINECA, Italy
Hisham Kholidy, SUNY Polytechnic Institute, USA
Martial Michel, Data Machines Corp, USA
Michael Jennings, Los Alamos National Lab (LANL), USA
Paolo Di Tommaso, Centre for Genomic Regulation (CRG), Spain / Seqera Labs, Spain
Parav Pandit, Mellanox, USA
Rosa M Badia, Barcelona Supercomputing Center, Spain
Lucas Benedicic, CSCS, Switzerland
Carlos Fernandez, CESGA, Spain
Gilles Wiber, CEA, France
Diakhate Francois, CEA, France
The following organizations are represented in the steering committee: