The ExaNeSt Project: Interconnects, Storage, and Packaging for Exascale Systems

作者: M. Katevenis , N. Chrysos , M. Marazakis , I. Mavroidis , F. Chaix

DOI: 10.1109/DSD.2016.106

关键词: Big dataInterconnectionSystem deploymentEmbedded systemFootprintComputer scienceArchitecturePower consumptionARM architectureRack

摘要: ExaNest is one of three European projects that support a ground-breaking computing architecture for exascale-class systems built upon power-efficient 64-bit ARM processors. This group share an "everything-close" and "share-anything" paradigm, which trims down the power consumption -- by shortening distance signals most data transfers as well cost footprint area installation reducing number devices needed to meet performance targets. In ExaNeSt, we will design implement: (i) physical rack prototype its liquid-cooling subsystem providing ultra-dense compute packaging, (ii) storage with distributed (in-node) non-volatile memory (NVM) devices, (iii) unified, low-latency interconnect, designed efficiently uphold desired Quality-of-Service guarantees mix inter-processor flows, (iv) efficient rack-level sharing, where each page cacheable at only single node. Our target test alternative interconnect options on actual hardware, using real-world HPC applications. The ExaNeSt consortium brings together technology, skills, knowledge across entire value chain, from IP, system deployment, all way up operating systems, storage, HPC, big frameworks, cutting-edge

参考文章(19)
Rafael Vidal Aroca, Luiz Marcos Garcia Gonçalves, Towards green data centers: A comparison of x86 and ARM architectures power efficiency Journal of Parallel and Distributed Computing. ,vol. 72, pp. 1770- 1780 ,(2012) , 10.1016/J.JPDC.2012.08.005
Yves Durand, Paul M. Carpenter, Stefano Adami, Angelos Bilas, Denis Dutoit, Alexis Farcy, Georgi Gaydadjiev, John Goodacre, Manolis Katevenis, Manolis Marazakis, Emil Matus, Iakovos Mavroidis, John Thomson, EUROSERVER: Energy Efficient Node for European Micro-Servers digital systems design. pp. 206- 213 ,(2014) , 10.1109/DSD.2014.15
Baba Arimilli, Ravi Arimilli, Vicente Chung, Scott Clark, Wolfgan Denzel, Ben Drerup, Torsten Hoefler, Jody Joyner, Jerry Lewis, Jian Li, Nan Ni, Ram Rajamony, The PERCS High-Performance Interconnect 2010 18th IEEE Symposium on High Performance Interconnects. pp. 75- 82 ,(2010) , 10.1109/HOTI.2010.16
Roberto Ammendola, Andrea Biagioni, Ottorino Frezza, Francesca Lo Cicero, Alessandro Lonardo, Pier Stanislao Paolucci, Davide Rossetti, Francesco Simula, Laura Tosoratto, Piero Vicini, A hierarchical watchdog mechanism for systemic fault awareness on distributed systems Future Generation Computer Systems. ,vol. 53, pp. 90- 99 ,(2015) , 10.1016/J.FUTURE.2014.12.015
Min Xie, Yutong Lu, Kefei Wang, Lu Liu, Hongjia Cao, Xuejun Yang, Tianhe-1A Interconnect and Message-Passing Services IEEE Micro. ,vol. 32, pp. 8- 20 ,(2012) , 10.1109/MM.2011.97
Torsten Hoefler, Software and Hardware Techniques for Power-Efficient HPC Networking computational science and engineering. ,vol. 12, pp. 30- 37 ,(2010) , 10.1109/MCSE.2010.96
Avinash Karanth Kodi, Brian Neel, William C. Brantley, Photonic Interconnects for Exascale and Datacenter Architectures IEEE Micro. ,vol. 34, pp. 18- 30 ,(2014) , 10.1109/MM.2014.62
R Ammendola, A Biagioni, O Frezza, F Lo Cicero, A Lonardo, P S Paolucci, D Rossetti, F Simula, L Tosoratto, P Vicini, APEnet+: a 3D Torus network optimized for GPU-based HPC Systems Journal of Physics: Conference Series. ,vol. 396, pp. 042059- ,(2012) , 10.1088/1742-6596/396/4/042059
Wolfgang E. Denzel, Jian Li, Peter Walker, Yuho Jin, A Framework for End-to-End Simulation of High-performance Computing Systems international conference on advances in system simulation. ,vol. 86, pp. 331- 350 ,(2010) , 10.1177/0037549709340840
Nikolaos Chrysos, Lydia Chen, Christoforos Kachris, Manolis Katevenis, Discharging the network from its flow control headaches: packet drops and HOL blocking IEEE ACM Transactions on Networking. ,vol. 24, pp. 15- 28 ,(2016) , 10.1109/TNET.2014.2378012