Call for Papers

Overview

The ACM International Symposium on High-Performance Parallel and Distributed Computing (HPDC) is the premier annual conference for presenting the latest research on the design, implementation, evaluation, and use of parallel and distributed systems for high-end computing. The 35th HPDC will take place in Cleveland, Ohio, United States, from July 13 to 16, 2026.

Scope and Topics

We welcome submissions on high-performance parallel and distributed computing (HPDC) topics, including but not limited to: clouds, clusters, grids, big data, massively multicore, and extreme-scale computing systems. Experience reports of operational deployments that provide significantly novel insights for future research on HPDC applications and systems are also welcome.

In the context of high-performance parallel and distributed computing, the topics of interest include, but are not limited to:

• Infrastructure, including datacenter, HPC, cloud, serverless, and edge/IoT computing platforms
• Accelerators and heterogeneous computing
• Memory systems
• File and storage systems, I/O, and data management
• Operating systems and networks
• System software and middleware for parallel and distributed systems
• Programming languages and runtime systems
• Big data stacks and significant data ecosystems
• Parallel and distributed computing systems and software for AI
• Scientific applications, algorithms, and workflows
• Resource management and scheduling
• Performance modeling and benchmarking
• Fault tolerance, reliability, and availability
• Deployment topics, including operational guarantees, risk assessment, and management
• Energy efficiency and sustainability
• Software engineering for parallel and distributed systems
• Novel post-Moore computing technologies, including but not limited to neuromorphic, brain-inspired computing, and quantum computing

HPDC welcomes submissions that utilize Artificial Intelligence to enhance the topics of interest mentioned above or that employ parallel and distributed systems to improve AI workloads. However, we emphasize that an HPDC submission must clearly address and explain its connection to research in parallel and distributed computing.

Paper Submission Categories

Authors can submit papers to HPDC in one of the following two categories: (1) regular papers, or (2) open-source tools and data papers. The primary focus of regular papers should be to describe new research ideas supported by experimental implementation and evaluation of the proposed research ideas. The primary focus of open-source tools and data should be to describe the design, development, and evaluation of new open-source tools or novel data sources. Submissions in the regular papers category are also strongly encouraged to open-source their software or hardware artifacts.

The authors are required to indicate the category of the paper as part of the submitted manuscript's title. The last line of the title should indicate the paper type by using one of the two phrases (1) Paper Type: Regular, or (2) Paper Type: Open-source tools and data paper.

Submissions in both categories will be evaluated to the same standards in terms of novelty, scientific value, demonstrated usefulness, and potential impact on the field. The chosen category at the time of submission cannot be changed after the submission deadline.

ACM Publication

Starting January 1, 2026, ACM will fully transition to an open-access publication model. All ACM publications, including those from ACM-sponsored conferences, will be 100% Open Access. Authors will have two primary options for publishing Open Access articles with ACM: the ACM Open institutional model or by paying Article Processing Charges (APCs). With over 1,800 institutions already part of ACM Open, the majority of ACM-sponsored conference papers will not require APCs from authors or conferences (currently, around 70-75%).

Authors from institutions not involved in ACM Open will need to pay an APC to publish their papers unless they qualify for a financial or discretionary waiver. To determine whether an APC applies to your article, consult the list of participating institutions in ACM Open and review the APC Waivers and Discounts Policy. Please note that waivers are rare and are granted only based on specific criteria established by ACM.

Recognizing that this change may present financial challenges, ACM has approved a temporary subsidy for 2026 to facilitate the transition and provide additional time for institutions to join ACM Open. The subsidy will offer:

• $250 APC for ACM/SIG members
• $350 for non-members

The subsidy represents a 65% discount, funded directly by ACM. Authors are encouraged to help advocate for their institutions to join ACM Open during this transition period.

This temporary subsidized pricing will apply to all conferences scheduled for 2026.

Based on funding availability, HPDC will partially cover the APC charges for institutes that are not part of ACM Open.

Submission Guidelines

Authors are invited to submit technical papers of at most 11 pages in PDF format, excluding references. Accepted papers will have the flexibility to use an additional page in the camera-ready to incorporate feedback from the reviewers. Papers should be formatted in the ACM Proceedings Style (using sigconf from https://www.acm.org/publications/proceedings-template) and submitted via the conference submission website. The formatting requirement is strict, and submissions violating the ACM Proceedings Style formatting will be rejected without review. Submitted papers must be original work that has not appeared in and is not under consideration for another conference or a journal.

Anonymizing Submissions

HPDC uses dual-anonymous (identities of reviewers and authors are not shared) reviewing. Avoid identifying yourself or your institution explicitly or by implication (e.g., through the references or acknowledgments). The first page should use the paper ID assigned during registration in place of the author names.

Use care in referring to your own related work. Do not omit references to your prior work, as this would make it difficult for reviewers to place your submission in its proper context. Instead, reference your past work in the third person, just as you would any other piece of related work. In some cases, it is not credible to refer to your related work in the third person. For example, your submission may extend a previous workshop paper, or it may relate to a submission currently under review at HPDC or another venue. In these cases, you must still explain the differences between your HPDC submission and the other work; however, you should cite the other work anonymously and email the deanonymized work to the PC chairs.

If your submission reports on experiences with a system at your institution, you should refer to the system anonymously but describe the properties of the system that are needed to evaluate the work (e.g., hardware specifications, size of the user base, volume of requests, etc.). We recognize that, in some cases, these properties may allow a reviewer to identify your institution. Tool/data papers should also adhere to the double-blind submission policy. If the described tool/dataset framework is already widely used by the research community, consider describing the framework using a different name and not sharing the open-source code repository in the paper.

Confidential Information

Papers containing information that is subject to a non-disclosure agreement (NDA) will not be considered for review.

Reproducibility

To promote scientific rigor, transparency, and reproducibility, authors may optionally include up to two pages of an Artifact Description (AD) and/or Artifact Evaluation (AE) appendix with their paper submission. The AD appendix can be used to describe any significant research products (e.g., software, datasets, models, workflows) that support the paper's scientific conclusions. If the work does not rely on such products, the appendix may simply indicate this.

The AE appendix allows authors to provide details on how artifacts can be accessed, built, and executed for the purpose of validating key results. Authors are encouraged to include: (i) clear instructions for obtaining artifacts, (ii) a description of the required software or hardware environment (e.g., packages, container images), and (iii) a simple way to reproduce results (e.g., script or Makefile). While public availability of artifacts is encouraged, submissions are welcome even if artifacts cannot be shared publicly.

The two pages of the AD/AE appendix should respect the double-blind policy.

arXiv Submission Policy

Please note that having an arXiv paper does not prohibit authors from submitting a paper to HPDC. arXiv papers are not peer-reviewed and not considered as formal publications, and hence do not count as prior work. Authors are not expected to compare against arXiv papers that have not formally appeared in previous conference or journal proceedings. If a submitted paper is already on arXiv, please continue to follow the double-blind submission guidelines. Authors are encouraged to take preventive measures to reduce the risk of accidental breach of anonymity (e.g., use a different title in the submission and refrain from uploading or revising the arXiv version during the review period after the submission deadline).

Author List After Acceptance

Please note that the author list cannot be changed after acceptance.

Conflict of Interest Declaration

At the time of submission, all authors must indicate their conflicts of interest with the PC members. A conflict of interest may be institutional, collaborative, or personal in nature. Please refer to the detailed guidelines on the submission website for accurate declaration of a conflict of interest.

Policy on the use of AI

HPDC follows the ACM Policy on Authorship for the use of AI in authoring papers. A comprehensive FAQ on the use of AI in publications is available from ACM:
https://www.acm.org/publications/policies/frequently-asked-questions.

Authors should note that the use of Generative AI tools is acceptable. However, authors should also note that they are fully responsible for the correctness and accuracy of AI-generated content in their papers. The use of Generative AI tools in the paper must be disclosed in accordance with ACM policy. We recommend that authors who choose to use Generative AI tools (such as ChatGPT, Claude, CoPilot, Grammarly, etc.) do so only for the purpose of improving the reading quality of their text and not for generating technical content.

ACM Policy on Authorship

Please refer to the ACM Policy on Authorship for all other guidelines.

By submitting your article to an ACM Publication, you are hereby acknowledging that you and your co-authors are subject to all ACM Publications Policies, including ACM's new Publications Policy on Research Involving Human Participants and Subjects. ACM will investigate alleged violations of this policy or any ACM Publications Policy, which may result in a full retraction of your paper, in addition to other potential penalties, as per the ACM Publications Policy.

Please ensure that you and your co-authors obtain an ORCID ID so that you can complete the publishing process for your accepted paper. ACM has been involved with ORCID from the beginning, and we recently committed to collecting ORCID IDs from all our published authors. The collection process is mandatory. We are dedicated to improving author discoverability, ensuring proper attribution, and supporting community efforts around name normalization. Your ORCID ID will help us achieve these goals.

Contact PC Chairs

Alfredo Goldman (gold@ime.usp.br)
Dimitrios S. Nikolopoulos (dsn@vt.edu)

Submissions

You can reach the submission site by clicking on the red button below.

HPDC Submission Site

Suggestions for Document Preparation for Authors

Formatting the Introduction Section

HPDC authors are encouraged to structure their introduction section of the paper in the following format (as subsections or headings).

1. Motivation. Clearly state the objective of the paper and provide (quantitative) support to motivate the specific problem your submission is solving.
2. Limitation of state-of-art approaches. Briefly review the most relevant and most recent prior works. Clearly articulate the limitations of previous works and how your approach breaks away from those limitations. A more detailed discussion should be reserved for the related work section. But this section should be sufficient to help readers recognize the novelty of your approach.
3. Key insights and contributions. Briefly articulate the major insights that enable your approach or make it effective. Clearly specify the novelty of these insights and how they advance state-of-the-art. Describe the key ideas of your approach and design. List the key contributions including flagship empirical results and improvement over the prior art as applicable.
4. Experimental methodology and artifact availability. Clearly specify the key experimental / simulation infrastructure and methodological details. Support the experimental methodology choices (e.g., cite that most relevant and most recent prior works have evaluated their ideas using similar methodology). Include a line to indicate whether the software/hardware artifact will be available upon acceptance
5. Limitations of the proposed approach. Almost all scientific contributions have limitations and scope for improvement. Clearly articulate all the major limitations of the proposed approach and identify conclusions that are sensitive to specific assumptions made in the paper.

Please note this suggested format is not a requirement for submission, and authors will not be penalized for using a different format.

Presenting your Experimental Setup and Results

Authors are encouraged to organize their experimental setup and results into subsections or headings. We also provide the following suggestions, which are not a requirement for submission.

1. Workloads and Datasets. List each benchmark/workload and why it is representative of HPDC use cases. If you are reporting subsets of benchmark suites, explain the reasons why you have chosen the specific subsets. Include version/commit IDs, data sizes, and any pre-/post-processing requirements. If the benchmark/workload is synthetic, justify the parameter choices and provide a generation script along with your artifact. Note any licensing or access constraints with your artifact as well.
2. Hardware and System Environment Describe CPU(s), accelerator(s), memory hierarchy, storage devices, network/interconnect (e.g., InfiniBand/NVLink), and power/cooling modes relevant to your experiments. Include node count, topology, memory layout (e.g., NUMA), and resource partitioning. Report firmware/driver/BIOS, DVFS settings, and any throttling controls.
3. Software Stack and Configuration. Specify OS, kernel, compilers, MPI/NCCL/RDMA libraries, CUDA/HIP/oneAPI versions, container runtime, scheduler (e.g., Slurm, Kubernetes), and key flags. State commit hashes for your code and baselines. Provide exact build commands and configuration files in the artifact (if available).
4. Baselines and Comparisons. Select recent, competitive baselines; explain inclusion/exclusion criteria. Document how baselines were tuned (search space, budgets) to ensure fairness. If you reimplemented a baseline, validate it (e.g., match a published number or show deltas/regression tests).
5. Methodology and Experimental Protocol. Define the primary metrics (e.g., execution time, throughput, time-to-solution, tail latency, energy, cost) and specify the warm-up policy, number of trials, random seeds, and processes to ensure statistical confidence in your data. If you present speedup, make sure you clarify what your baseline is. Clarify load models (open/closed-loop), input mixes, and failure/retry policies. State when results were averaged and how outliers were handled. Explain the choice of averaging metrics (e.g., mean, median) according to the data distribution and give clear indications of deviations from the average.
6. Reproducibility Kit (AD/AE). Indicate whether artifacts are or will be available (code, configs, datasets, or pointers, container image, Makefile/script to reproduce key tables/figures). Provide a one-line path to the artifact or “available upon acceptance.”

Please note this suggested format is not a requirement for submission, and authors will not be penalized for using a different format.

Ethical Considerations

If your research describes a new security-related attack, please consider adding information about the responsible disclosure to the relevant entity. Overall, as appropriate and relevant, the paper should follow the ethical principles and not alter the security/privacy/equality expectations of the associated human users.

Inclusive Description of Research Contributions

Please consider making your research contribution description inclusive in nature. For example, consider using gender-neutral pronouns, consider using examples that are ethnicity/culture-rich, consider engaging users from diverse backgrounds if your research involves a survey, etc. Best efforts should be made to make the paper accessible to visually impaired or color-blind readers.

Go back to Calls