|
The 26th International Symposium on Parallel and Distributed Computing
(ISPDC 2027)
July 7-9, 2027, 5* St. Raphael Resort and Marina, Limassol, Cyprus
Distributed systems have seen an increasing convergence between traditional workflows
(such as HPC simulations) and agentic AI workflows. Simulation remains the workhorse of
high-performance computing, but supercomputers are increasingly asked to host a
second class of workload alongside it. Agentic AI systems (e.g., foundation models acting
autonomously through iterative loops of reasoning and tool use) now issue tool calls that
are themselves HPC tasks: simulations, solvers, and data-analysis pipelines. Serving them
adds interactive, asynchronous and bursty demands to machines designed for bulk-
synchronous batch jobs. The question this raises for our community is not one of
replacement but of extension: how schedulers, runtimes, communication layers, I/O
stacks and memory hierarchies should be extended so that agentic and other AI
workloads can coexist with, and feed back into, traditional large-scale computation. ISPDC
welcomes work on these new capabilities as much as continued advances in the classical
foundations of parallel and distributed computing.
We invite original research contributions to ISPDC 2027 in the following categories: full
papers, short papers, posters, and talks.
Topics of Interest
Contributions are welcome in, but not limited to, the following areas:
● Algorithms and models for parallel and distributed computing
● Parallel and distributed workflows for large-scale applications, including AI-driven and
agentic orchestration
● Applications of parallel and distributed systems (HPC, Big Data, AI, Quantum Computing)
● Performance modeling, analysis, and evaluation
● Parallel and distributed I/O, storage and data management
● Architectures and environments for parallel and distributed computing
● Resource management and scheduling, including co-scheduling of interactive AI
services with batch workloads
● Reproducibility and explainability in parallel and distributed computing
● Cloud computing, edge computing, federated learning
● Machine learning in the context of parallel and distributed systems
● Distributed training and inference at scale: parallel strategies, memory offloading,
checkpointing, model serving
● Extending HPC systems for agentic AI: tool calls into simulations, agent memory and
context data management
Contribution Types and Page Limits
Page limits reflect the Springer formatting standard.
● Full papers: Max. 20 pages (excluding references) and 25 pages (including references).
● Short papers: Max. 12 pages (excluding references) and 15 pages (including references).
● Poster papers: Max. 5 pages (including references). Final printed poster recommended
size is A0.
● Talks: Extended abstract of max. 2-3 pages (including references).
Submission and Formatting Guidelines
● Publisher & Format: Accepted papers will be published by Springer in the Lecture Notes
in Computer Science (LNCS) series.
● Templates: Authors must prepare their manuscripts using the standard Springer LNCS
templates (LaTeX or Word). Detailed author instructions, templates, and formatting
guidelines can be found at: Springer Information for Authors and Editors.
● Submission Details: All contributions must be submitted electronically in PDF format
via the official conference submission portal:
● Review Process: Submissions must include author names and affiliations (single blind).
Publication Requirements
At least one author of each accepted paper or poster paper to be published in the
conference proceedings must register at the full rate and present the work at the
conference.
Important Dates (AoE)
● Paper Submission: February 26, 2027
● Author Notification: April 19, 2027
● Camera-Ready & Author Registration: April 30, 2027
● Conference Dates: July 7-9, 2027
Organisation
General Chair
• George A. Papadopoulos, University of Cyprus, Cyprus
Program Committee Chair
• Bogdan Nicolae, Argonne Național Laboratory
|
October 13th, 2026 
Daniela Lopez de Luise 