IPDPS 2027 Call for Tutorials

41st IEEE International Parallel and Distributed Processing Symposium
June 1–5, 2027 | Seattle, Washington, USA
Tutorials: Saturday, June 5, 2027

TUTORIALS CHAIR

• Rajesh Sankaran, Argonne National Lab, USA

TUTORIALS CO-CHAIRS

• Aalap Tripathy - Hewlett Packard Enterprise
• Marco Domenico Santambrogio - Politecnico di Milano
• Márcio Castro, Federal University of Santa Catarina (UFSC)

Overview

The 41st IEEE International Parallel and Distributed Processing Symposium (IPDPS 2027) invites proposals for tutorials to be presented in person on Saturday, June 5, 2027, at the Hyatt Regency Bellevue in the Seattle metropolitan area.

The IPDPS Tutorials Program provides an opportunity for researchers, educators, practitioners, and students to engage with emerging technologies, methodologies, and practical techniques in parallel and distributed processing.

For 2027, the symposium anticipates selecting a focused program of approximately four to five  tutorials, with particular emphasis on topics of interest to students and early-career researchers. Proposals that combine technical depth with practical instruction, interactive exercises, and transferable research skills are especially encouraged.

The Tutorials Program will be coordinated with the IPDPS 2027 Student Program to promote participation, encourage professional development, and strengthen connections between emerging researchers and the broader parallel and distributed computing community.

The Challenges we want Tutorials to Address

What students need to learn. Parallel and distributed computing is changing faster than most curricula. Students now work where AI, HPC and distributed infrastructure meet, on heterogeneous hardware, with Python-first software stacks and data arriving from the edge. Papers rarely teach the practical skills this demands: moving code from a laptop to a GPU cluster, measuring and reasoning about performance, building workflows that span instruments, cloud and HPC, and doing it all reproducibly. Having clear understanding of not just what happened, but also why, particularly as agentic workflows and AI generated code is incorporated is crucial for trusting the generated artifacts. This is currently a hard to impossible problem. We want tutorials that close this gap.

Keeping a Saturday audience. Tutorials run on Saturday, after the main program ends, and we expect attendance to be lower than on other days. Many attendees travel on; HPDC (June 5-11, Atlanta) and ISC (June 7-11, Hamburg). We state this up front for two reasons:

• We want proposers to help solve it. Every proposal must explain why a student would stay the extra day for this tutorial.

• If your tutorial has attendance targets, as project- or vendor-backed tutorials often do, please weigh this before you submit.

Scope and Topics of Interest

Tutorial proposals are welcome across the broad technical scope of IPDPS, consistent with the IPDPS 2027 Call for Papers.

Topics of particular interest include, but are not limited to:

• Artificial Intelligence and Large Language Model Systems: Distributed training, efficient inference, model serving, and agentic workflows across high-performance computing and cloud environments.

• Python-Based Parallel and GPU Programming: Practical approaches to parallel and accelerated computing using frameworks such as mpi4py, Dask, Numba, CuPy, JAX, and Warp, including scaling applications from interactive notebooks to multi-node systems.

• The Edge–Cloud–HPC Continuum: Distributed data acquisition, streaming analytics, computational workflows, orchestration, and integration of scientific instruments with cloud and high-performance computing resources.

• Performance Engineering for Heterogeneous Systems: Performance measurement, profiling, benchmarking, roofline analysis, optimization, and performance portability across CPUs, GPUs, and emerging accelerators.

• AI-Assisted Parallel Programming: Applications of large language models and AI-assisted development tools to code generation, porting, debugging, testing, and optimization of parallel applications, including their limitations and validation requirements.

• Reproducible and Open Computational Research: Research artifact preparation, software environments, reproducible benchmarking, experimental methodology, and practices aligned with the IPDPS Reproducibility Initiative.

• Access to Research Computing Infrastructure: Practical guidance on accessing and using shared research testbeds, national computing facilities, cloud resources, and allocation programs.

• Education in Parallel and Distributed Computing: Innovative teaching methodologies, instructional tools, curriculum development, and practical approaches to introducing parallel and distributed computing concepts.

• Emerging Computing Platforms and Paradigms: Quantum–HPC integration, energy-aware computing, and other emerging approaches to large-scale computational systems.

• Trustable Scientific Computing: Having proper provenance, particularly for automated decision making and automatic code generation sufficient for someone to audit and understand how a result/discovery was created is crucial to the scientific method.

Tutorial Format and Expectations

Tutorials should provide a structured learning experience with clearly defined objectives, an appropriate level of technical depth, and tangible outcomes for participants.

The following guidelines will apply:

1. Interactive and Hands-On Learning
   
   Preference will be given to tutorials that incorporate substantial hands-on components, including guided exercises, programming activities, demonstrations, or practical problem-solving sessions.
   
   Lecture-oriented tutorials will also be considered when the proposed format is well justified by the subject matter and learning objectives.
2. Tutorial Duration
   Proposals may be submitted for either of the following formats:
   • Half-day tutorials: 3 hours (preferred)
   • Full-day tutorials: 6 hou

   The half-day format is preferred to allow participants to attend multiple tutorials. Full-day proposals should clearly justify the additional time required to achieve their educational objectives.

3. Student and Early-Career Researcher Engagement
   
   Tutorials are expected to provide meaningful educational value to students and early-career researchers. Proposals must describe how the tutorial will engage this audience through practical activities, relevant research applications, and clearly identifiable learning outcomes.
   
   As tutorials will take place on the final day of the symposium, proposers should also articulate the distinctive value of attending their tutorial and describe approaches to encouraging sustained participation.
   
   Accepted tutorials are encouraged to recruit attendees with a short teaser earlier in the week, such as a poster at the cross-workshop poster session on Tuesday evening, June 1, or a lightning slot coordinated with the Student Program.
4. In-Person Participation
   
   All tutorials will be conducted in person. Presenters must commit to attending IPDPS 2027 and delivering their sessions on site. There will be no online attendance by participants as well, simplifying tutorial delivery.
5. Tutorial Materials
   
   Accepted tutorial organizers will be required to make instructional materials, including slides, handouts, and practical exercises, available to registered participants in advance of the conference.
   
   Organizers are encouraged to make materials openly accessible whenever appropriate and to adopt licenses that support their reuse for educational and research purposes.

Attendee Requirements and Computing Infrastructure

To maximize accessibility and participation, tutorials should be designed around a minimal attendee computing requirement: ideally a laptop equipped with a web browser and wireless internet connectivity.

Tutorial organizers should take into account the diversity of computing environments available to participants, including institutional devices with restricted software installation privileges.

In particular:

• Tutorials should not require participants to purchase software licenses, maintain paid cloud accounts, or independently procure computational resources.

• Proposals should not assume that participants can install specialized toolchains, execute containers, or obtain administrative privileges on their devices.

• Tutorials requiring substantial computational resources should provide access to suitable environments, such as hosted notebooks, temporary training accounts, or appropriately provisioned cloud or high-performance computing resources.

• Any requirements beyond the recommended baseline must be clearly identified, together with reasonable alternatives for participants unable to meet those requirements.

• Proposals must specify anticipated infrastructure needs, including network bandwidth, concurrent connections, electrical power, and room configuration.

Accepted organizers will be responsible for providing setup instructions sufficiently in advance and validating the tutorial computing environment before the scheduled session.

Proposal Submission Requirements

Tutorial proposals must be submitted as a single PDF document of no more than three pages, single-spaced, double-column pages in 10-point font on 8.5x11 inch pages (IEEE conference style), including figures, tables, presenter biographies and references. The submission link for proposals is https://ssl.linklings.net/conferences/ipdps/ (select "Tutorial Proposal" under the "Make a New Submission" tab). 

Each proposal must include the following information:

1. Tutorial Title and Duration: Proposed title and requested half-day or full-day format.

2. Abstract: A concise description of the tutorial, not exceeding 200 words, suitable for publication on the conference website.

3. Objectives and Intended Audience: Learning objectives, technical scope, expected outcomes, target audience, prerequisite knowledge, and intended proficiency level (beginner, intermediate, or advanced).

4. Tutorial Structure and Schedule: A detailed outline of the proposed content, including the time allocated to each section and the identification of lectures, demonstrations, and hands-on activities.

5. Student Engagement Plan: A description of the tutorial's relevance to students and early-career researchers, its expected educational benefits, approaches to attracting and retaining participants on the final conference day, and any proposed coordination with the IPDPS Student Program.

6. Attendee Prerequisites and Infrastructure: Required software, hardware, computational resources, participant preparation, access arrangements, contingency plans, and venue requirements.

7. Expected Attendance: Estimated participation, minimum viable enrollment, and any attendance expectations or commitments associated with supporting projects, institutions, or vendors.

8. Previous Tutorial Offerings: Information about prior presentations of the tutorial, if applicable, including conference or venue, dates, attendance, links to publicly available materials, and a description of substantive updates or adaptations for IPDPS 2027.

9. Presenter Information: Names, affiliations, and email addresses of all presenters; identification of the primary contact; brief biographies of no more than 200 words per presenter; and confirmation of in-person availability. Previous tutorial teaching experience should be highlighted.

Proposers are also requested to indicate whether any presenters anticipate requiring visa invitation letters. This information will be used exclusively to facilitate administrative planning and will not influence proposal evaluation.

Evaluation and Selection Criteria

Tutorial proposals will be reviewed by  the IPDPS 2027 Tutorials Committee.

The committee will consider both the individual merits of each proposal and its contribution to a balanced, complementary tutorial program.

Evaluation criteria will include:

• Technical relevance: Significance of the proposed topic to the IPDPS community and its alignment with emerging research directions.

• Educational value: Clarity of learning objectives, appropriateness for the intended audience, and potential benefit to students and early-career researchers.

• Instructional quality: Organization, technical depth, pedagogical approach, and effective integration of practical or hands-on activities.

• Accessibility and feasibility: Suitability of attendee requirements, availability of computational resources, and practicality of the proposed delivery format.

• Student engagement: Strength of the proposed approach to encourage student participation and sustained attendance.

• Presenter qualifications: Demonstrated expertise in the topic and relevant experience in instruction, training, or community engagement.

• Program complementarity: Contribution to the overall breadth and balance of the tutorial program, including avoidance of substantial overlap with other selected tutorials.

• Presenter representation: Breadth of institutional, geographic, and career-stage representation within the tutorial program.

Tutorials supported by research projects, institutions, or commercial organizations are welcome. Such proposals should emphasize broadly applicable concepts, methodologies, and transferable skills rather than instruction focused exclusively on a particular product or proprietary platform.

The committee anticipates selecting approximately four to five tutorials, subject to the quality of submissions and program constraints.

Important Dates

All dates are tentative and end at 11:59 PM Anywhere on Earth (AOE). Notification in mid-January leaves presenters more than four months for travel and visa planning.

You are welcome to contact tutorials@ipdps.org submitting to discuss your idea.