TEXTAROSSA: Towards EXtreme scale Technologies and Accelerators for euROhpc hw/Sw Supercomputing Applications for exascale

Agosta, Giovanni; Cattaneo, Daniele; Fornaciari, William; Galimberti, Andrea; Massari, Giuseppe; Reghenzani, Federico; Terraneo, Federico; Zoni, Davide; Brandolese, Carlo; Celino, Massimo; Iannone, F.; Palazzari, P.; Zummo, Giovanni; Bernaschi, Massimo; D'Ambra, Pasqua; Saponara, Sergio; Danelutto, Marco; Torquati, Massimo; Aldinucci, Marco; Arfat, Yasir; Cantalupo, Barbara; Colonnelli, Iacopo; Esposito, Roberto; Martinelli, Alberto Riccardo; Mittone, Gianluca; Beaumont, Olivier; Bramas, Bérenger; Eyraud-Dubois, Lionel; Goglin, Brice; Guermouche, Abdou; Namyst, Raymond; Thibault, Samuel; Filgueras, Antonio; Vidal, Miquel; Álvarez, Carlos; Martorell, Xavier; Oleksiak, Ariel; Kulczewski, Michal; Lonardo, A.; Vicini, P.; Cicero, Francesca Lo; Simula, Francesco; Biagioni, A.; Cretaro, Paolo; Frezza, Ottorino; Paolucci, Pier Stanislao; Turisini, Matteo; Giacomini, Francesco; Montangero, Simone; Ammendola, Roberto

doi:10.1109/dsd53832.2021.00051

Cited by 4 publications

(4 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…taffo is supported by the Italian National Interuniversity Consortium for Informatics (CINI) Workgroup on High Performance Computing as one of its key technologies [19]. As such, taffo will be used as the precision tuning tool by the EuroHPC TEXTAROSSA consortium [20], which aims at providing mixed precision support in extreme-scale computing systems. The consortium comprises, beyond CINI, three European HPC centers, as well as the key industrial providers of HPC in Europe.…”

Section: Impactmentioning

confidence: 99%

TAFFO: The compiler-based precision tuner

Cattaneo¹,

Chiari²,

Agosta³

et al. 2022

SoftwareX

View full text Add to dashboard Cite

Section: Impactmentioning

confidence: 99%

TAFFO: The compiler-based precision tuner

Cattaneo¹,

Chiari²,

Agosta³

et al. 2022

SoftwareX

View full text Add to dashboard Cite

“…The aspects related to the development of new IPs, including mixed-precision computing, data compression, security, scheduling and power monitoring go beyond the scope of this work, as they address a wide range of different contributions at software and hardware level, including the integration of electronic design automation tools. We note here that these technologies will be key to opening new usage domains, including High Performance Data Analytics (HPDA) and High Performance Artificial Intelligence (HPC-AI) [3].…”

Section: Strategic Goalsmentioning

confidence: 99%

The TEXTAROSSA Approach to Thermal Control of Future HPC Systems

Fornaciari

Terraneo

Agosta

et al. 2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…Recently, application malleability has become a hot research topic in several large-scale projects (ADMIRE [10], DEEP-SEA [11], RED-SEA [12], and TEX-TAROSSA [13], among others). In the context of application scheduling, the malleability comes as a rescue by providing support to dynamically increase or decrease the number of processors of the running applications [14] (we denote this action as application reconfiguration) and providing an extra dimension to the scheduling problem; the resources can now be dynamically re-allocated from some applications to others.…”

Section: Introductionmentioning

confidence: 99%

Performance-driven scheduling for malleable workloads

ALMAAITAH,

Singh,

Özden

et al. 2023

Preprint

View full text Add to dashboard Cite

The development of adaptive scheduling algorithms that take advantage of malleability has become a crucial area of research in many large-scale projects. Malleable workloads can improve the system's performance but at the same time provide an extra dimension to the scheduling problem. This paper proposes an adaptive, performance-based job scheduling method that emphasizes the backfilling concept with malleability. The proposed method performs the malleability operations only when the estimated execution time of the involved applications is better than or equal to the execution time on the allocated resources without reconfiguration. The reconfiguration feasibility is determined by performance models considering the application scalability and reconfiguration overheads. Different policies for implementing malleability are presented, each targeting a specific workload in terms of job size and scalability. The comprehensive evaluation shows an improvement in the slowdown up to 49\% compared to the non-adaptive baseline scheduling algorithm.

show abstract

TEXTAROSSA: Towards EXtreme scale Technologies and Accelerators for euROhpc hw/Sw Supercomputing Applications for exascale

Cited by 4 publications

References 49 publications

TAFFO: The compiler-based precision tuner

TAFFO: The compiler-based precision tuner

The TEXTAROSSA Approach to Thermal Control of Future HPC Systems

Performance-driven scheduling for malleable workloads

Contact Info

Product

Resources

About