In recent years, the convergence of high-performance computing and eScience has opened new scientific discovery and innovation avenues. Researchers across various domains leverage advanced computational techniques to analyze vast data, simulate complex systems, and accelerate scientific workflows.
The 1^st Workshop on High-Performance eScience (HiPES) aims to bring together experts from academia, industry, and government organizations to discuss the latest developments in this rapidly evolving field and explore emerging trends, challenges, and opportunities within the framework of tools and applications.

The workshop focuses on managing distributed workflows in eScience, encompassing a broad spectrum of topics essential for interoperability and efficiency. This includes exploring design patterns, languages, and orchestration tools tailored to the specific needs of eScience high-performance applications. Additionally, researchers delve into scheduling algorithms, fault tolerance mechanisms, performance monitoring, and benchmarking procedures to ensure the smooth functioning of distributed workflows across diverse scientific domains.
Emphasizing the concept of FAIRness (Findable, Accessible, Interoperable, and Reusable), the workshop encourages contributors to explore this crucial aspect of eScience. Methodologies and tools are discussed to ensure data and computational resources are discoverable, accessible, and usable across various platforms and research communities. Privacy and security considerations, particularly concerning sensitive scientific data, are integral components open for discussion during the workshop.

The workshop will feature invited talks, contributed presentations, and panel discussions.

Scope

The HiPES 2024 workshop will be held on August 26^th with the EuroPar 2024 conference in Madrid, Spain. All papers accepted to HiPES 2024 will be published in a revised form in a special Euro-Par Workshop Volume in the Lecture Notes in Computer Science (LNCS) series after the Euro-Par conference.

The HiPES workshop provides a platform for researchers to present their latest work on high-performance eScience tools, algorithms, and applications. It fosters collaboration and knowledge exchange among participants from diverse backgrounds, including computer science, computational science, domain sciences, and engineering. The audience is encouraged to discuss practical challenges and best practices in developing and deploying high-performance
eScience solutions in real-world settings are identifying future research directions and potential areas for innovation in high-performance eScience.

For some time, eScience tools and applications have been first-class citizens in the parallel and distributed computing ecosystem, especially at extreme scales. With the advent of deep learning and foundational models, this trend has been fated to stand for a long time. Furthermore, when moving from the researcher’s desk to a production-grade pipeline, all parallel and distributed computing algorithms must integrate more or less directly with eScience tools and applications.

The convergence between the modular nature of modern applications and the eScience domain, coupled with diverse hardware configurations, necessitates adaptable workflow systems. These systems must support various execution environments, from traditional High-Performance Computing (HPC) infrastructures to cloud environments and emerging Edge computing platforms. Optimization policies balancing performance and energy efficiency are paramount, alongside the integration of various computational models, including classical and quantum computing paradigms. In line with the ethos of the Workflow Community Initiative, the HiPES workshop offers a collaborative platform for researchers to share expertise, exchange ideas, and gain insights from diverse perspectives within the context of eScience. The event facilitates direct engagement through open discussions, fostering the exchange of views and ideas among participants and encouraging collaborative endeavors to tackle the challenges and opportunities in distributed workflows within the eScience domain.

The HiPES workshop convenes researchers specializing in distributed workflows, a vital aspect of eScience. These workflows are a robust tool for developing and executing intricate applications across distributed architectures, driving forward eScience initiatives. The complexities inherent in modeling, orchestrating, and monitoring distributed workflows within the eScience landscape offer fertile ground for research and innovation.

Call For Papers

The 1^st Workshop on High-Performance eScience (HiPES) aims to bring together experts from academia, industry, and government organizations to discuss the latest developments in this rapidly evolving field and explore emerging trends, challenges, and opportunities within the framework of tools and applications.

The HiPES 2024 workshop will be held on August 26^th with the EuroPar 2024 conference in Madrid, Spain. All papers accepted to HiPES 2024 will be published in a revised form in a special Euro-Par Workshop Volume in the Lecture Notes in Computer Science (LNCS) series after the Euro-Par conference.

The workshop is designed for researchers, practitioners, and students interested in high-performance computing, science, and interdisciplinary research. Participants from academia, industry, government labs, and nonprofit organizations are encouraged to attend and contribute to the vibrant discussions and collaborations.

We invite submissions of original research contributions, case studies, and innovative applications in the following areas (but not limited to):

• High-performance computing (HPC) techniques and architectures for eScience
• Languages, paradigms, and tools for scientific workflows
• Scalable and efficient scientific workflows orchestration for eScience
• Optimization techniques for parallel and distributed computing
• Integration of machine learning and AI with eScience applications
• Big data analytics and visualization in scientific domains
• Cloud computing and edge computing for eScience
• Reproducibility and sustainability in computational research
• Performance evaluation and benchmarking of eScience tools and platforms
• Paradigms and techniques for FAIRness in eScience
• Ad-hoc file systems for I/O intensive eScience applications
• Prospective and retrospective provenance models for large-scale applications
• Security, privacy, and trust models and tools for eScience platforms
• Case studies and success stories of real-world eScience applications

We encourage submissions of full papers describing original research, work-in-progress, or experience reports related to the workshop topics.

Important dates:

Paper submission: May 20
Notification of acceptance: June 20
Camera-ready: July 1

Submission guidelines:

Authors are invited to submit papers electronically through EasyChair. The papers should be submitted in PDF format following the Springer LNCS format. Paper length must not exceed 12 pages (including references). All submitted manuscripts will be checked for originality by Springer iThenticate (papers that show insufficient originality might be rejected without a review).

Workshop proceedings:

Accepted papers presented at the workshop will be published in a revised form in a special Euro-Par Workshop Volume in the Lecture Notes in Computer Science (LNCS) series after the Euro-Par conference.

Schedule

The schedule will be available after papers acceptance.

Time	Activity	Speakers	Title
8:30-9:00	Registration
9:00-9:15	Opening and welcoming	Raffaele Montella, Iacopo Colonnelli, Paula Olaya
9:15-10:00	Keynote speech	Felix Wolf, Technical University of Darmstad	Tackling the imbalance between computation and I/O
10:00-10:30	Coffee break
10:30-10:50	Presentation 1	Juan José Moreno Riado, Savíns Puertas Martín, Juana López Redondo, Pilar Martínez Ortigosa and Ester Martín Garzón	Exploiting Multicore Servers to Optimize IMRT Radiotherapy Planning
10:50-11:10	Presentation 2	Alberto Mulone, Doriana Medic and Marco Aldinucci	A Fault Tolerance Mechanism for Hybrid Scientific Workflows
11:10-11:40	Presentation 3	Dante Sánchez-Gallegos, Diana Carrizales, Catherine Alessandra Torres Charles, Alejandro De La Rosa Zequeira, Jose Luis Gonzalez-Compean and Jesus Carretero	GeoNimbus: A serverless framework to build earth observation and environmental services
11:40-12:10	Presentation 4	Simone Perrotta, Ciro Giuseppe De Vita, Gennaro Mellone, Marco Edoardo Santimaria, Giuseppe Salvi, Marco Lapegna, Massimo Torquati and Angelo Ciaramella	Extending a scientific workflow engine with streaming I/O capabilities: DAGonStar and CAPIO
12:10-12:30	Presentation 5	Hanwen Dai, Changbo Chen and Yuxuan Song	Accelerating GCN Inference on Small Graphs
12:30-13:30	Launch break
13:30-13:40	Presentation 6	Melesio Crespo-Sanchez, Hugo German Reyes-Anastacio, Jose Luis Gonzalez-Compean, Jaqueline Calderon-Hernandez and Ignacio Castillo-Barrios	Towards the implementation of ONCA: A Generic, Scalable, and Massive Data Processing Platform for Information Discovery and Analytics
13:40-14:10	Presentation 7	Lorenzo Di Rocco, Riccardo Ceccaroni, Umberto Ferraro Petrillo and Pierpaolo Brutti	A Distributed Workflow for Long Reads Self-Correction
14:10-14:30	Presentation 8	Subhajit Sahu, Kishore Kothapalli and Dip Sankar Banerjee.	GVE-LPA: Fast Label Propagation Algorithm (LPA) for Community Detection in the Shared Memory Setting
14:30-15:30	Panel	Chair: Iacopo Colonnelli	Real-world high-performance eScience Applications
15:30-16:00	Closing remarks	Raffaele Montella

Invited Speakers

Felix Wolf, Technical University of Darmstadt, Tackling the imbalance between computation and I/O

Abstract. Data-intensive workflows spend a significant portion of their execution time on I/O operations such as reading input, writing output, or checkpointing intermediate results. This means that any speedups achieved by accelerating computation are limited by the fraction of time spent on I/O. Conversely, to achieve good performance, data-intensive applications depend to a higher degree than compute-intensive ones on the available I/O bandwidth, which is usually shared with other potentially data-intensive applications, leading to a situation where inter-application interference can significantly impact performance. In this talk, we will review this problem in more detail and propose a novel scheduling algorithm that does not require the explicit scheduling of I/O bandwidth. In the second part of the talk, we will introduce compositional performance modeling, a technique that can assist developers in deriving performance models of workflows from performance models of their components based on composition patterns. 

Biography. Felix Wolf is a full professor at the Department of Computer Science of the Technical University of Darmstadt in Germany, where he leads the Laboratory for Parallel Programming. He works on methods, tools, and algorithms that support developing and deploying parallel software systems in various life-cycle stages. Wolf received his Ph.D. degree from RWTH Aachen University in 2003. After working more than two years as a postdoc at the Innovative Computing Laboratory of the University of Tennessee, he was appointed research group leader at Juelich Supercomputing Centre. Between 2009 and 2015, he was head of the Laboratory for Parallel Programming at the German Research School for Simulation Sciences in Aachen and a full professor at RWTH Aachen University. Wolf has made major contributions to several open-source performance tools for parallel programs, including Scalasca, Score-P, and Extra-P. Moreover, he has initiated the Virtual Institute – High Productivity Supercomputing, an international initiative of HPC programming-tool builders to enhance, integrate, and deploy their products. He has published over 150 refereed articles on parallel computing, several of which have received awards.

Organizers

• Raffaele Montella, Associate Professor, Department of Science and Technologies, University of Naples “Parthenope”, Italy.
• Iacopo Colonnelli, Assistant Professor, Department of Computer Science Department, University of Turin, Italy
• Paula Olaya, Graduate Research Assistant, Department of Electrical Engineering and Computer Science, University of Tennessee, Knoxville, TN-USA

Program Committee

• Barbara Cantalupo (University of Turin, Italy)
• Taina Coleman (California State University Long Beach, CA-USA)
• Diana Di Luccio (University of Naples “Parthenope”, Italy)
• Antonella Galizia (IMATI-CNR, Italy)
• Sandra Gesing (San Diego Supercomputing Center, San Diego, CA-USA)
• José Luis Gonzales Compeán (CINVESTAV-IPN, Mexico City, Mexico)
• Antony Kougkas (Illinois Institute of Technology, IL-USA)
• Marco Lapegna (University of Naples “Federico II”, Italy)
• Jay Lofstead (Sandia National Laboratories, NM-USA)
• Anirban Mandal (RENCI, Chapel Hill, NC-USA)
• Jack Marquez (University fo Tennessee, Knoxville, TN-USA)
• Claudia Misale (IBM, NY-USA)
• Gabriele Morabito (University of Messina, Messina, Italy)
• Dante Sánchez-Gallegos (University Carlos III of Madrid, Spain)

Contact

• Raffaele Montella, raffaele.montella@uniparthenope.it
• Iacopo Colonnelli, iacopo.colonnelli@unito.it
• Paula Olaya, polaya@vols.utk.edu