France-Paris: Super computer
Prior information notice
This notice is for prior information only
Section I: Contracting authority
Main address: http://www.genci.fr
Main address: https://www.bsc.es/
Main address: https://www.cineca.it/
Main address: http://www.fz-juelich.de/ias/jsc/EN/Home/home_node.html
French procurement law.
Section II: Object
Joint procurement for high performance IT equipment for European supercomputing centres.
This procurement takes place within the PPI4HPC project funded by the European Commission under the call EINFRA-21-2017. It involves four public procurers (BSC, CINECA, FZJ/JSC and GENCI) located in four different countries (Spain, Italy, Germany and France) working together in a joint procurement. The purpose of the procurement is, for each public procurer, to buy an innovative high performance supercomputer and/or an innovative high performance storage system that will be integrated in their computing centre. These systems are expected to be deployed in the 2019-2021 time frame. In combination these systems should allow to cover a large range of applications including traditional HPC applications, HPDA and AI. These systems are furthermore supposed to address common topics of innovation.
Compute and storage infrastructure for high performance data analytics
BSC will acquire in this lot a compute and storage infrastructure for high performance data analytics, to be installed during the first half of 2019. The BSC HPDA infrastructure will combine compute nodes with innovative storage technologies like NVRAM in conjunction storage technologies in place at BSC, like normal hard drives and tape infrastructure, configured in a tiered storage solution to store hundreds of petabytes of scientific data.
For the storage part, an innovative Tiered storage solution will be expected scalable to hundreds of petabytes, being able to feed and provide data to the different BSC HPC and HPDA resources for the actual and further years inside the pre-exascale era.
For the compute part, a High performance for data analytics infrastructure will be acquired to complement actual BSC HPC productions clusters, which will pre or post process simulations' data with new Big Data or Analytics paradigms and algorithms.
Innovative solutions for massive data storage and tiering will be expected for this lot.
Supercomputer with innovative support for data-intensive applications
The goal of the lot is to procure a high-performance computing resource that is planned to be installed at Jülich Supercomputing Centre's (JSC) supercomputing facility by the end of 2020. The system will replace the currently operated JURECA cluster and must be able to execute the diverse workloads of this system with significantly higher performance and energy efficiency. In addition, the system shall have architectural features specifically targeted at data intense science. In particular, JSC expects the system to integrate dense memory technologies into its memory hierarchy and expose this memory to applications through novel APIs. The system is expected to be connected to the central storage subsystem in the JSC facility with high bandwidth.
Besides the dense memory integration, JSC expects innovative hardware and software solutions offering high energy efficiency as well as improved facility integration (e.g., improved power monitoring and control) and application performance monitoring capabilities.
Evolution of the Tier-0 system towards Exascale architectures
The goal of the lot is to procure a new supercomputing system that will replace the throughput partition of MARCONI installed in the CINECA's data centre. With the deploy of the new system, targeted in the second part of 2019, CINECA aims to reach a computing power of the order of tens PFlops featuring node architecture that maximize the performance throughput per energy unit. The new system will support the CINECA's commitment to provide leading edge innovative computing resources to European research as a PRACE Tier-0 hosting member.
The procured resource integrated in the MARCONI infrastructure will represent a new step towards the architectural convergence between high performance computing and high performance data analytics (data-centric architecture) and will be able to run computing and data intensive workloads, with particular relevance towards high energy efficiency solutions. In this regard, the system will benefit from innovative hardware and software solutions that aim to reduce power consumption, improve power management, enhance I/O performance, to help making supercomputing sustainable toward exascale for a broad range of scientific application.
High performance and energy efficient computing systems towards the convergence between HPC, HPDA and AI
CEA DAM Ile-De-France, TGCC (Très Grand Centre de calcul du CEA), Chemin Grand Rué, 91680 Bruyères Le Châtel France.
The goal of the lot is to increase, during the second half of 2019, the computing power installed in the TGCC (Très Grand Centre de Calcul du CEA, located near Paris) and made available to French and European researchers. In the range of several tens of Petaflop/s new HPC and data resources will be seamlessly integrated with the system installed in early 2018 in the same location. The new system will provide a balanced architecture for scientific applications, including very high IO bandwidth connection to the storage system of the TGCC.
In order to fulfil such requirements and to achieve high energy efficiency together with excellent TCO performance ratio towards exascale systems, GENCI expects innovative solutions both in terms of hardware (processors, interconnection network…) and in terms of software (novel programming models….) as well as a strong support from the provider especially during the early operation of the new system including contribution to application porting and optimization.
Section IV: Procedure
Section VI: Complementary information
This joint procurement aims at fostering:
— Science and engineering applications in Europe within PRACE, the pan-European HPC infrastructure by providing more computing resources.
— R&I on HPC architectures and technologies in Europe, strengthened by this joint procurement and the coordinated deployment of supercomputer incorporating innovative solutions which will require strong relationship and possibly collaboration between the procurers and the suppliers for large scale testing, tuning and maturation.
— A greater weight and more impact on common topics of innovation and on the design of the solutions according to the need of scientists and engineers in Europe by a coordinated approach.
The procedure expected to be used is the competitive dialogue (‘Dialogue compétitif’ as defined in the article 42 of the French Ordinance n.2015-899 dated 23.7.2015) organised as a joint procurement (‘Groupement de commandes’ as defined in the article 28 of the French Ordinance n.2015-899 dated 23.7.2015). The procurement will be conducted by GENCI, acting as the lead procurer in the name and on behalf of the four contracting authorities, until the invitation by the lead procurer to participate in the dialogue. The dialogue itself will be conducted independently by each public procurer for its own lot. French law will govern all the procurement procedure. The performance of each lot is subject to the national law of the respective public procurer.
The public procurers involved in the procurement are willing to inform the market in advance about their expectations and plans and to gather feedback from the market on them. Therefore, an Open Dialogue Event (ODE) will take place on 6.9.2017 in Brussels. All interested possible suppliers are invited to participate to this event. Specific details (contact point, agenda, registration information…) will be provided by 3.7.2017 on the web site of the project (http://www.ppi4hpc.eu or https://www.ppi4hpc.eu). In addition, one to one meetings will be organized with each supplier wishing to do so during 28.9.2017-29.9.2017 and 4.10.2017-6.10.2017. Details on the organization of these meetings will be provided on the web site of the project.
This procurement receives funding under the European Union's Horizon 2020 research and innovation programme under the grant agreement No 754271. The EU is however not participating as a contracting authority in this procurement.