_version_ 1866910128930816000
author Mantovani, Filippo
Banchelli, Fabio
Vizcaino, Pablo
Ferrer, Roger
Palomar, Oscar
Minervini, Francesco
Labarta, Jesus
Olivieri, Mauro
Pomata, Sebastiano
Marcuello, Pedro
Cortina, Jordi
Moreno, Alberto
Sans, Josep
Espasa, Roger
Papaefstathiou, Vassilis
Dimou, Nikolaos
Ieronymakis, Georgios
Psathakis, Antonis
Giaourtas, Michalis
Mastorakis, Iasonas
Marazakis, Manolis
Guthmuller, Eric
Bocco, Andrea
Fereyre, Jérôme
Fuguet, César
Kovač, Mate
Kovač, Mario
Mrković, Luka
Ramljak, Josip
Bertaccini, Luca
Fischer, Tim
Gurkaynak, Frank K.
Scheffler, Paul
Benini, Luca
Goel, Bhavishya
Manivannan, Madhavan
Rocha, Tiago
Neves, Nuno
Krüger, Jens
author_facet Mantovani, Filippo
Banchelli, Fabio
Vizcaino, Pablo
Ferrer, Roger
Palomar, Oscar
Minervini, Francesco
Labarta, Jesus
Olivieri, Mauro
Pomata, Sebastiano
Marcuello, Pedro
Cortina, Jordi
Moreno, Alberto
Sans, Josep
Espasa, Roger
Papaefstathiou, Vassilis
Dimou, Nikolaos
Ieronymakis, Georgios
Psathakis, Antonis
Giaourtas, Michalis
Mastorakis, Iasonas
Marazakis, Manolis
Guthmuller, Eric
Bocco, Andrea
Fereyre, Jérôme
Fuguet, César
Kovač, Mate
Kovač, Mario
Mrković, Luka
Ramljak, Josip
Bertaccini, Luca
Fischer, Tim
Gurkaynak, Frank K.
Scheffler, Paul
Benini, Luca
Goel, Bhavishya
Manivannan, Madhavan
Rocha, Tiago
Neves, Nuno
Krüger, Jens
contents This paper presents EPAC, a RISC-V-based accelerator chip developed within the European Processor Initiative (EPI) as part of a multi-year, multi-partner effort to build a European HPC processor ecosystem. EPAC is implemented in GlobalFoundries 22FDX (GF22FDX) technology, covers an area of 27 sq mm with approximately 0.3 billion transistors, and integrates three distinct RISC-V compute tiles targeting different workload classes: VEC, a vector processing tile for double-precision HPC workloads; STX, a many-core tile optimized for stencil and machine learning computations; and VRP, a variable-precision tile for iterative numerical solvers requiring extended floating-point formats. All tiles are connected through a Coherent Hub Interface (CHI) based network-on-chip with a distributed L2 cache system and communicate with external memory via a SerDes link. The chip was taped out in GF22FDX technology and successfully brought up, with all major IP blocks validated. This paper describes the architecture of each tile and the uncore infrastructure, the integration and physical implementation process, and the board-level bring-up activities. It also reflects on the engineering and coordination lessons learned from a full chip design effort distributed across academic and industrial partners in Europe.
format Preprint
id arxiv_https___arxiv_org_abs_2604_12715
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle EPAC: The Last Dance
Mantovani, Filippo
Banchelli, Fabio
Vizcaino, Pablo
Ferrer, Roger
Palomar, Oscar
Minervini, Francesco
Labarta, Jesus
Olivieri, Mauro
Pomata, Sebastiano
Marcuello, Pedro
Cortina, Jordi
Moreno, Alberto
Sans, Josep
Espasa, Roger
Papaefstathiou, Vassilis
Dimou, Nikolaos
Ieronymakis, Georgios
Psathakis, Antonis
Giaourtas, Michalis
Mastorakis, Iasonas
Marazakis, Manolis
Guthmuller, Eric
Bocco, Andrea
Fereyre, Jérôme
Fuguet, César
Kovač, Mate
Kovač, Mario
Mrković, Luka
Ramljak, Josip
Bertaccini, Luca
Fischer, Tim
Gurkaynak, Frank K.
Scheffler, Paul
Benini, Luca
Goel, Bhavishya
Manivannan, Madhavan
Rocha, Tiago
Neves, Nuno
Krüger, Jens
Hardware Architecture
Distributed, Parallel, and Cluster Computing
This paper presents EPAC, a RISC-V-based accelerator chip developed within the European Processor Initiative (EPI) as part of a multi-year, multi-partner effort to build a European HPC processor ecosystem. EPAC is implemented in GlobalFoundries 22FDX (GF22FDX) technology, covers an area of 27 sq mm with approximately 0.3 billion transistors, and integrates three distinct RISC-V compute tiles targeting different workload classes: VEC, a vector processing tile for double-precision HPC workloads; STX, a many-core tile optimized for stencil and machine learning computations; and VRP, a variable-precision tile for iterative numerical solvers requiring extended floating-point formats. All tiles are connected through a Coherent Hub Interface (CHI) based network-on-chip with a distributed L2 cache system and communicate with external memory via a SerDes link. The chip was taped out in GF22FDX technology and successfully brought up, with all major IP blocks validated. This paper describes the architecture of each tile and the uncore infrastructure, the integration and physical implementation process, and the board-level bring-up activities. It also reflects on the engineering and coordination lessons learned from a full chip design effort distributed across academic and industrial partners in Europe.
title EPAC: The Last Dance
topic Hardware Architecture
Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2604.12715