ABOUT US

MHPC

Industry and academia throughout the world are suffering from a severe shortage of skilled personnel able to bridge the gap between the dynamic environments in IT and research/science. From 2014, MHPC aims to create a group of experts who have the hands-on training and skills needed to fill this gap, by providing students with a solid background in scientific computing approaches, algorithms, and modeling.

MHPC coursework is driven by challenging scientific and technical problems that require an HPC approach. Lectures are provided by SISSA and ICTP staff and highly recognized international experts: both institutes have long histories and experience in developing and applying scientific and research computation models.

The skills acquired make it possible to respond to the growing demand for access to large-scale scientific computing resources, which is now essential in many research areas, and to bridge the gap between IT and academia, providing the opportunity to successfully pursue different types of careers. 

Our alumni testify that this specialization course opens doors in academia as well as in industry and consulting companies. Data from the first eight editions confirm this trend, as almost all alumni found a job in the field of HPC within 12 months of completing their training: 66% in academia and 34% in companies.

read more
success stories

MHPC Alumni

A section dedicated to the stories of our alumni: where did MHPC take them? 

read more
News

Latest News & Events

NEWS December 15, 2025

HPC Course Showcased at USRSE25 to Train Future Supercomputing Experts

A paper presenting a hands-on course that trains students in the deployment and management of high-performance computing clusters was introduced at the annual User and System Administrator Research Software Event (USRSE) earlier this fall. The paper highlights the growing need for practical skills to support increasing supercomputing demands.

The course, titled High-Performance Computing Technologies (HPCT), was presented by Fernando Posada of the U.S. Department of Energy’s Oak Ridge National Laboratory and co-authored by Richard Berger of Los Alamos National Laboratory. It outlines the course design, methodology, and outcomes, focusing on real-world HPC cluster administration.

HPCT is part of a master’s program in high-performance computing taught annually at the International Center of Theoretical Physics in Trieste, Italy, and provides students with hands-on experience in building, configuring, and maintaining computing clusters used in scientific research.

 

🔗 More about https://shorturl.at/ID5AI  

NEWS June 11, 2025

MHPC alumna Keshvi Tuteja wins "Women in HPC Poster Award" at ISC 2025

We are proud to announce that Keshvi Tuteja, a graduate of the Master in High Performance Computing (MHPC), has received the prestigious "Women in HPC Poster Award" at the ISC High Performance 2025 conference, currently taking place in Hamburg, Germany.

Keshvi completed the MHPC programme in April 2024 and is currently pursuing a PhD in Karlsruhe, Germany. Her winning poster stood out for its clarity, scientific value, and contribution to the advancement of high performance computing — a testament to her commitment and talent.

The Women in HPC initiative aims to promote diversity and inclusion in the HPC community by highlighting the outstanding work of women researchers and professionals in the field.

We warmly congratulate Keshvi on this important recognition and wish her all the best in her future research and academic career.

🔗 More about ISC High Performance: https://isc-hpc.com

EVENT April 3, 2025

𝗔 𝗖𝗮𝘀𝗲 𝗦𝘁𝘂𝗱𝘆 𝗼𝗳 𝟲𝟰-𝗯𝗶𝘁 𝗙𝗹𝗼𝗮𝘁𝗶𝗻𝗴 𝗣𝗼𝗶𝗻𝘁 𝗘𝗺𝘂𝗹𝗮𝘁𝗶𝗼𝗻 𝗶𝗻 𝗤𝘂𝗮𝗻𝘁𝘂𝗺 𝗘𝘀𝗽𝗿𝗲𝘀𝘀𝗼

The evolution of computer architecture, driven by AI applications, has significantly enhanced the reduced- and mixed-precision computing capabilities of GPUs, offering various low-precision representations such as FP16, BF16, and FP8. There is growing interest in utilizing these capabilities through mixed-precision algorithms and emulation techniques to boost scientific computing performance without compromising accuracy.

This presentation explores the potential of leveraging lower precision Tensor Cores for scientific applications by emulating FP64 and FP32 matrix multiplications without sacrificing accuracy, thereby achieving improved performance per watt.

We will discuss a novel library developed at NVIDIA, which transparently accelerates DGEMM and ZGEMM calls using INT8 Tensor Cores. Results from several use cases in Quantum Espresso will be presented, highlighting the effectiveness of this approach.

Would you like to attend the seminar? Send an e-mail to our Director Ivan Girotto: igirotto@ictp.it