CANOPIE-HPC is a workshop focusing on containerization, virtualization, and other methods to implement user-defined, bring-your-own, or isolated software environments. Submissions will be peer-reviewed, and accepted papers will be published in IEEE Computer Society.
The workshop is held in conjunction with SC24: The International Conference for High Performance Computing, Networking, Storage, and Analysis.


### Important Dates ###

Submission opens: June 4, 2024

Submission closes (hard deadline - no extensions): August 16, 2024

Papers decisions: September 6, 2024

Lightning talks decisions: September 13, 2024

Camera-ready papers deadline: September 27, 2024

Workshop date: November 17, 2024

Note: Items are due at 23:59 “anywhere on Earth” on the specified date. Specifically, this is 23:59 IDLW, i.e., UTC–12:00.


### URLs ###

- Workshop homepage: https://canopie-hpc.org/

- Paper submission link: https://submissions.supercomputing.org/?page=Submit&id=SCWorkshopCANOPIEHPCSubmission&site=sc24

- Lightning talk submission link: https://submissions.supercomputing.org/?page=Submit&id=SCWorkshopCANOPIEHPCLightningTalkSubmission&site=sc24


### Program chairs ###

Alberto Madonna, Swiss National Supercomputing Centre
Claudia Misale, IBM T.J. Watson Research Center
Shane Canon, Lawrence Berkeley National Lab
Andrew Younge, Sandia National Labs


### Overview ###

The ongoing revolution enabled via containerization, virtualization, and new orchestration models has dramatically changed how applications and services are delivered and managed across the computing industry. This revolution has established a new ecosystem of tools and techniques with new, flexible and agile approaches, and continues to gain traction in the HPC community. In addition to HPC-optimized container runtimes, emerging technologies like Kubernetes create a new set of opportunities and challenges. While adoption is growing, questions regarding best practices, foundational concepts, tools, and standards remain.

Our goal is to promote the adoption of these tools and introspect the impact of this new ecosystem on HPC use cases. This workshop serves as a key venue for presenting late-breaking research, sharing experiences and best practices, and fostering collaboration in this field. Our sixth workshop iteration will continue to emphasize real-world experiences and challenges in adopting and optimizing these new approaches for HPC.


### Scope ###

The scope of this workshop is to better understand and improve paradigms around containerization, virtualization and orchestration for HPC use cases. The most well-known approaches are containers and virtualization, but anything to further these goals is welcome.

Topics include but are not limited to:

- Container runtimes, virtual machine (VM) implementations, and other related technologies (e.g. unikernels, WebAssembly)
- System and architecture portability
- Scientific reproducibility, including FAIR considerations
- Experience reports, perspectives from container framework users, developers, and system administrators HPC in the cloud and/or cloud in the HPC
- GPUs, accelerators, proprietary interconnects, and other hardware considerations
Security considerations when adopting container technologies, including zero trust models
- Container and VM image management and curation, including distribution and archiving
- Workflows, including interaction between traditional HPC applications and containerized workflows or edge services
- Performance and scaling studies with containers and/or VMs
- Orchestration, scheduling, and/or resource management of jobs and microservices
- New interaction techniques such as web apps (Jupyter, Rstudio, etc.)
- Community standards, including OCI
- Role of containerization in DevOps (ex: Gitlab CI/CD)
- Perspectives on container outreach - convincing native application users/devs to make the jump


### Workshop format ###

We worked continuously to improve the CANOPIE-HPC program based on feedback from the community. We have made adjustments to the format for SC24 to include more dynamic and engaging content, in addition to the traditional full paper presentations. The content of technical sessions will be driven by the mix of accepted submissions curated by the program chairs. We will offer two submission tracks: one for full paper submissions, and one for lightning talks. Differently from the previous edition, CANOPIE-HPC SC24 is going to be a full day workshop.

Full paper submissions will receive at least 3 peer reviews in the SC24 submission system on the basis of scientific validity, impact to the field, reproducibility, and opportunity for interactive discussion at the workshop. The program committee will discuss the submissions and their reviews over a conference call moderated by the program chairs; final proceedings will be selected by majority vote of the committee. As in all previous years, accepted submissions will be published by the IEEE Computer Society for inclusion in the IEEE Digital Library. Artifacts for reproducibility (digital archives of code, README instructions, images, etc.) will be a significant factor in paper acceptance into the workshop. Further details are available in the "Submission procedure" section below.

For lightning talks, participants can submit an abstract and an extended abstract that will be reviewed by the program committee. Our lightning talk track is meant to showcase demos, exploratory and late-breaking work, pain-points or shortcomings in current container/virtualization technologies, and projects that may not be fleshed-out enough to warrant a paper submission. We especially encourage lightning talk submissions from early career participants and people who are relatively new to the world of containers/virtualization.


### Diversity and inclusivity ###

Similarly to the broader field of HPC, the organizers recognize concerns with diversity and inclusion for CANOPIE-HPC. We continuously work to improve diversity and inclusion at leadership, program committee, and participation levels. We plan early and often targeted recruiting and advice-seeking with diversity organizations (e.g., WHPC). CANOPIE-HPC states explicit inclusion goals in the workshop CFP, includes an optional inclusion plan in paper submission system, and specially considers any submitted inclusion plans during the Program Committee meeting. To open CANOPIE-HPC to more early-career and non-expert participants, we plan to offer an expanded lightning talks session to lower the barrier to entry in the workshop. The workshop organizers will make any space considerations necessary to include those with accessibility requirements and ensure wheelchair-designated seating remains available to those in need. We also plan to include a 30 minute open discussion section to invite additional audience questions and participation. Furthermore, we will encourage attendees to join the Slack channel hpc-containers to have an opportunity to continue to chat, ask questions and make professional connections.


### Reproducibility plan ###

In conjunction with the SC Reproducibility Initiative, submissions are required to provide artifact descriptions as submission appendices, if applicable. Artifacts such as reproduction instructions, source code, build recipes, and/or container images would all be appropriate and considered during peer review.


### Submission procedure ###

Upload your paper in PDF format. Papers must use the IEEE conference proceedings template, in two-columns, U.S. letter (8.5″ x 11″) format. The templates for LaTeX and MS Word provided by IEEE eXpress Conference Publishing are available for download at https://www.ieee.org/conferences/publishing/templates.html. Manuscript length must have a lower limit of 5 pages, including references. There is no upper page limit.

We enthusiastically welcome original, high-quality submissions within the scope above. These may describe complete studies; work-in-progress research; position papers on controversial, emerging, or hot topics; state of the practice; or any other manuscript the authors believe should be included in the CANOPIE-HPC program. We encourage submissions from academia, industry, government, and/or any other type of institution.

Each manuscript will be assessed using peer review by program committee members (or outside reviewers, if needed) on the basis of scientific validity, impact to the field, reproducibility, inclusivity, and opportunity for useful and lively discussion at the workshop. Review will be single-blind; i.e., reviewers will know authors’ identities, but not vice versa. Authors should not anonymize their manuscripts. Each manuscript submission can expect to have at least 3 reviews and will be checked using anti-plagiarism software (see Plagiarism Guidelines: https://www.ieee.org/publications/rights/index.html#plagiarism-guidelines).

The use of artificial intelligence AI–generated text in an article shall be disclosed in the acknowledgements section of any paper submitted to SC. The sections of the paper that use AI-generated text shall have a citation to the AI system used to generate the text. Utilizing Large Language Models (LLMs) is permissible as a general-purpose writing assistance tool. Authors are expected to acknowledge their complete accountability for the contents of their papers, including content generated by LLMs that could be interpreted as plagiarism or scientific misconduct (e.g., fabrication of facts). LLMs are not eligible for authorship.

In conjunction with the SC Reproducibility Initiative, submissions should be as transparent as possible regarding all methods. Submissions are also expected to provide reproducibility artifacts such as reproduction instructions, source code, build recipes, and/or container images. These should be included as part of the submission in the form of Artifact Description/Artifact Evaluation appendices (https://sc24.supercomputing.org/program/papers/reproducibility-appendices-badges/), which adhere to an established format adopted by SC24.

The program committee will discuss the submissions and their reviews and select the program over a video conference meeting. Submissions will be assessed as-is, with no expectation of substantive revision after peer review, though some submissions may be accepted conditional on specified changes ("shepherded").

Rejected paper submissions could be offered the possibility to present as a lightning talk, at the discretion of the program committee.