Background

MAMHYP-22 is the sixth workshop after the first five held in Torun, Warsaw, Krakow, Lublin and Bialystok in 2011, 2013, 2015, 2017, and 2019 respectively, jointly to the PPAM conferences. Despite the progress done in the field, the development of efficient algorithms for HPC systems with multiple forms of parallelism is still a challenging problem.

More precisely, from an architectural point of view, a High-Performance Computing system can be described using a hybrid multi-level structure: at the highest level, there are several systems connected among them by geographic networks (System level); an intermediate level is composed by the nodes in a single system communicating among them through dedicated fast networks or high-performance switches (Node level); at the lowest level, finally, there are several computing elements, computing cores as well as graphic accelerators, sharing resources in a single CPU (Core level).

These architecture levels have very different features, requiring different algorithmic development methodologies. For such a reason, the development of algorithms and scientific software for these systems implies a suitable combination of several methodologies to deal with the different kinds of parallelism corresponding to each architectural level. The general aim is to develop hybrid and hierarchical algorithms, able to be aware of the underlying platform. The main problems in this field are the management of large parallelism degrees due to several computing units, the heterogeneity of these devices, and the combination of several kinds of parallelism in a single algorithm. These topics are mainly investigated to gain the so-called exascale performance and, from another side, to face new forms of distributed computing as cloud computing, edge computing, and their interaction with HPC.

This workshop focuses specifically on Models, Methodologies, Algorithms, and Environments to exploit all forms of parallelism and their combination at all levels in the emerging HPC systems to gather the current state of knowledge in the field.

 

The topics include, but are not limited, to:

• Hybrid and hierarchical based parallel algorithms

• Multi-core and many-core parallel computing, GPU computing

• Architecture-aware parallelization on HPC platforms

• Auto tuning techniques for heterogeneous and parallel environments

• Performance analysis

• Techniques for multi-/many-core platforms, NUMA architectures, or accelerator components

• Task scheduling and load balancing among computing elements

• Synchronization and access to shared resources

• Multilevel cache management

• Computational kernels for scientific computing and applications

• Performance and scalability models

• Tools and programming environments supporting efficient usage of multilevel parallelism.

• Resources virtualization

• High Performance cloud/edge/fog computing

• Intra/Infra Virtual machines high performance communications algorithms

• Fault tolerant implementation

 

 

Program

 

      abstract of the talks

 

Monday, Sept 12. 15:40 - 16:55

• Malleability Techniques for HPC Systems - Jesus Carrettero
• Benchmarking A High Performance Computing Heterogeneous Cluster - Luisa Carracciuolo, Davide Bottalico, Davide Michelino, Gianluca Sabella and Bernardino Spisso
• Algorithm and software overhead: a theoretical approach to performance portability - Valeria Mele and Giuliano Laccetti

Tuesday, Sept 13. 10:50 - 12:30

• A GPU accelerated Hyperspectral 3D Convolutional Neural Network Classification at the Edge with Principal Component Analysis preprocessing - Gianluca De Lucia, Marco Lapegna and Diego Romano
• A Generative Adversarial Network approach for noise and artifacts reduction in MRI head and neck imaging - Salvatore Cuomo, Francesco Fato, Lorenzo Ugga, Gaia Spadarella, Renato Cuocolo, Edoardo Prezioso, Fabio Giampaolo and Francesco Piccialli (online)
• Parallel EUD models for accelerated IMRT planning on modern HPC platforms - Juan Jose' Moreno Riado, Janusz Miroforidis, Ignacy Kaliszewski and Gracia Ester Martin Garzon
• Environmental data tiling: store in Cloud, process at the Edge - Gennaro Mellone, Ciro Giuseppe De Vita, Diana Di Luccio, Sokol Kosta and Raffaele Montella

 

Paper Submission and Publication

All rules of paper submission of the PPAM conference apply. In particular:

• Papers will be refereed and accepted on the basis of their scientific merit and relevance to the
  Workshop topics.

• Papers presented at the Workshop will be included into the conference proceedings and
  published after the conference by Springer in the LNCS series.

• Before the Workshop, abstracts of accepted papers will be posted on this site.

• Authors should submit papers (PDF files)  using the online submission tool

• Papers are not to exceed 12 pages (LNCS style)

• Final camera-ready versions of accepted papers will be required by November 2, 2022.
   

 

Dates

Submission of Papers: May 20, 2022  (extended)
Notification of Acceptance: June 25, 2022
Conference: September 11-14, 2022
Camera-Ready Papers: November 2, 2022

Workshop Organizers

• Giuliano Laccetti (Univ. of Naples Federico II and INFN)    giuliano.laccetti@unina.it

• Marco Lapegna (Univ. of Naples Federico II and INFN)    marco.lapegna@unina.it

• Raffaele Montella (Univ. of Naples Parthenope)    raffaele.montella@uniparthenope.it

• Valeria Mele (Univ. of Naples Federico II, Italy)  valeria.mele@unina.it
• Diego Romano (Italian National Research Council) diego.romano@cnr.it

 

 

Program committee (tentative)

 

-  Alfredo Buttari, CNRS (National Center for Scientific Research) , France

-  Diana Di Luccio University of Naples Parthenope, Italy

-  Javier Garcia Blas, University Carlos III, Madrid Spain

-  Sokol Kosta, Aalborg University Copenhagen, Denmark
-  Giuliano Laccetti, University of Naples Federico II, Italy, and INFN (Nuclear Physics National Institute), Italy

-  Marco Lapegna, University of Naples Federico II, Italy, and INFN (Nuclear Physics National Institute), Italy
-  Valeria Mele, University of Naples Federico II, Italy

-  Raffaele Montella, University of Naples Parthenope, Italy

-  Gloria Ortega, University of Almeria, Spain
-  Guido Russo, University of Naples Federico II and INFN (Nuclear Physics National Institute), Italy

-  Lucas Szustak, Polish Academy of Sciences, Poland

 

 

Special Event 

 

Special Issue on Scientific Journal 

Authors of selected papers from MAMHYP22 will be invited to submit an extended version of their papers to a Special Issue of the journal "Concurrency and Computation: Practice and Experiences." Other high-quality papers that fit the Special Issue topics are also welcome for submission. More info available here