EPiGRAM-HS along with EPEEC are organizing a Eurolab4HPC Thematic Session titled "Programming models for upcoming supercomputers" that is collocated with HiPEAC Computing Systems Week (CSW) Autumn 2019. The event is on October 28, 2019 in Bilbao, Spain.

The goal of the event is to merge efforts of the two European Exascale initiatives and exchange ideas about ongoing research on Exascale programming models. Participants not related to these projects are more than welcome and highly encouraged to join, including all the members within the Eurolab4HPC community.

More information about HiPEAC CSW 2019: https://www.hipeac.net/csw/2019/bilbao/#/

Meeting on a Structured Approach for Programming Extreme Scale and Heterogeneous Systems

EPiGRAM-HS has organized a focus meeting at ETH on the 8th-10th of September 2019 in Zurich and gathered scientists within the HPC field to discuss recent advances and current challenges that the community is facing.

The goal of the meeting was to bring together leading experts in programming models and systems for networks, heterogeneous memories and compute units to define a structured approach for programming extreme scale and heterogeneous systems. The meeting included short informal presentations about perceived strength and limitations of different approaches together with potential solutions. There was also dedicated sessions for discussion in small groups.


Hartwig Anzt
Valeria Bartsch
Michela Becchi
Siegfried Benkner
Tiziano De Matteis
Tom Deakin
Chen Ding
Tim Dykes
Maya Gokhale
William Gropp

Utz-Uwe Haus
Torsten Hoefler
Dan Holmes
Niclas Jansson
Erwin Laure
Dong Li
Glenn Lockwood
Pekka Manninen
Stefano Markidis
Shirley Moore
Antonio J. Peña

Harvey Richardson
Rob Ross
Martin Ruefenacht
Mitsuhisa Sato
Timo Schneider
Martin Schulz
Anthony Skjellum
Joost VandeVondele
Jeffrey Vetter
Michele Weiland
Felix Wolf

The meeting's agenda can be found here:


16 - 20 June 2019, Frankfurt

EPiGRAM-HS actively participated in ISC'19 by supporting WHPC (women in High Performance Computing) which took place in Frankfurt, Germany.

EPiGRAM-HS has provided 120 t-shirts to WHPC, thus supporting the cause of gender equality in HPC.

More information on ISC'19 can be found here:

EuroMPI 2019

10 - 13 September 2019, ETH Zurich

EPiGRAM-HS will actively participate in the organisation of EuroMPI 2019 which will take place in Zurich, Switzerland.

The EuroMPI conference is the preeminent meeting for users, developers and researchers to interact and discuss new developments and applications of message-passing parallel computing, in particular in and related to the Message Passing Interface (MPI). This includes parallel programming interfaces, libraries and languages, architectures, networks, algorithms, tools, applications, and High Performance Computing with particular focus on quality, portability, performance and scalability. The annual meeting has a long, rich tradition, and has been held in European countries since 1994.

More information can be found here:

Bernardino Romera Peredes – Semantic segmentation for medical images [Video]

On Friday 5th of October 2018 EPiGRAM-HS had the honour of sponsoring Bernardino Romera Peredes's talk on “Semantic segmentation for medical images“. 

Bio: Bernardino Romera-Paredes is a research scientist at DeepMind. He was a postdoctoral research fellow in the Torr Vision Group at the University of Oxford. Previously, he received his Ph.D. degree from University College London in 2014, supervised by Prof. Massimiliano Pontil and Prof. Nadia Berthouze, and also did an internship at Microsoft Research. He has published in top-tier machine-learning conferences such as in Conference on Neural Information Processing Systems (NIPS), International Conference on Machine Learning (ICML), and International Conference on Computer Vision (ICCV), as well as in journals, such as the Journal of Machine Learning Research (JMLR). His research focuses on structure prediction in computer vision, such as semantic and instance segmentation, and on multitask and transfer learning methods applied to vision tasks.

Abstract: Deep learning approaches such as the U-net, have shown a high performance in a wide range of medical image segmentation tasks. However, a deep learning architecture is only one part of building clinically applicable tools. In this talk I will present three projects from DeepMind Health Research addressing these challenges. The first project, a collaboration with University College London Hospital, deals with the challenging task of the precise segmentation of radiosensitive head and neck anatomy in CT scans, an essential input for radiotherapy planning. The second project, together with Moorfields Eye Hospital, developed a system that analyses 3D OCT (optical coherence tomography) eye scans to provide referral decisions for patients. The performance was on par with world experts with over 20 years experience. Finally, I will focus on the third project, which deals with the segmentation of ambiguous images. This is of particular relevance in medical imaging where ambiguities can often not be resolved from the image context alone. We propose a combination of a U-net with a conditional variational autoencoder that is capable of efficiently producing an unlimited number of plausible segmentation map hypotheses for a given ambiguous image. We show that each hypothesis provides a globally consistent segmentation, and that the probabilities of these hypotheses are well calibrated.