User’s needs influencing HPC technologies

E. Athanasaki, N. Meyer, M. Cestari, A.Tuncer Durak, A. Tekin, P. Gschwandtner
Greek Research and Technology Network
Partnership for Advanced Computing in Europe, 2022


   title={User’s needs influencing HPC technologies},

   author={Athanasakia, E and Meyerb, N and Cestaric, M and Durak, A Tuncer and Tekin, A and Gschwandtner, P},



Download Download (PDF)   View View   Source Source   



The user requirements imposed by modern challenges are influencing future High Performance Computing (HPC) technologies and use cases. This report analyses a wide range of user requirements and new technologies and their impact on European and worldwide HPC trends, in particular in the PRACE and EuroHPC ecosystems, as well as HPC infrastructures provided by member countries. Applications that did not require the use of advanced computing or for which HPC solutions were not possible due to excessive costs and availability are now available in HPC and cloud computing. The way of using computing infrastructure and its services is changing, and due to its unification and decrease of costs, the popularity of HPC and cloud computing will continue to increase. The present variety of applications significantly exceeds the scientific community and also applies to industry, governmental use and security. An additional driving force of future HPC technology originates from the policies of the European Commission and member countries’ regional needs. Big data analytics and artificial intelligence are among the identified applications, Interactivity in computing systems can also be essential for certain workflows. FPGAs and other specialized hardware or software packages are ideal for selected classes of applications. The above examples in conjunction with other requirements, outlined within this report, affects the design of HPC systems and can be used to justify the diversification of hardware and software set ups according to their target users.
No votes yet.
Please wait...

* * *

* * *

HGPU group © 2010-2023 hgpu.org

All rights belong to the respective authors

Contact us: