Reconfigurable computing for Monte Carlo simulations: results and prospects of the Janus project

© EDP Sciences, Springer-Verlag 2012. Artículo firmado por 26 autores. We wish to thank several past members of the Janus Collaboration, F. Belletti, M. Cotallo, D. Sciretti and J.L. Velasco, for their important contributions to the project. Over the years, the Janus project has been supported by the EU (FEDER funds, No. UNZA05-33-003, MEC-DGA, Spain), by the MICINN (Spain) (contracts FIS2006- 08533, FIS2009-12648, FIS2007-60977, FIS2010-16587, FPA2004-02602, TEC2010- 19207), by CAM(Spain), by the Junta de Extremadura (GR10158), by UCM-Banco Santander (GR32/10-A/910383), by the Universidad de Extremadura (ACCVII-08) and by the Microsoft Prize 2007. We thank ETHlab for their technical help. E.M. was supported by the DREAM SEED project and by the Computational Platform of IIT (Italy). M.B.-J. and B.S. were supported by the FPU program (Ministerio de Educación, Spain); R.A.B. and J.M.-G. were supported by the FPI program (Diputación de Aragón, Spain); finally J.M.G.-N. was supported by the FPI program (Ministerio de Ciencia e Innovación, Spain).
We describe Janus, a massively parallel FPGA-based computer optimized for the simulation of spin glasses, theoretical models for the behavior of glassy materials. FPGAs (as compared to GPUs or many-core processors) provide a complementary approach to massively parallel computing. In particular, our model problem is formulated in terms of binary variables, and floating-point operations can be (almost) completely avoided. The FPGA architecture allows us to run many independent threads with almost no latencies in memory access, thus updating up to 1024 spins per cycle. We describe Janus in detail and we summarize the physics results obtained in four years of operation of this machine; we discuss two types of physics applications: long simulations on very large systems (which try to mimic and provide understanding about the experimental non equilibrium dynamics), and low-temperature equilibrium simulations using an artificial parallel tempering dynamics. The time scale of our non-equilibrium simulations spans eleven orders of magnitude (from picoseconds to a tenth of a second). On the other hand, our equilibrium simulations are unprecedented both because of the low temperatures reached and for the large systems that we have brought to equilibrium. A finite-time scaling ansatz emerges from the detailed comparison of the two sets of simulations. Janus has made it possible to perform spin glass simulations that would take several decades on more conventional architectures. The paper ends with an assessment of the potential of possible future versions of the Janus architecture, based on state-of-the-art technology.
EU (FEDER funds)
Ministerio de Economía y Competitividad (MINECO)
Ministerio de Ciencia e Innovación (MICINN)
Comunidad de Madrid
Junta de Extremadura (Spain)
UCM-Banco Santander (Spain)
Universidad de Extremadura (Spain)
Microsoft Prize 2007
Computational Platform of IIT (Italy)
FPU program (Ministerio de Educación, Spain)
FPI program (Diputación de Aragón, Spain)
FPI program (Ministerio de Ciencia e Innovación, Spain)
Depto. de Física Teórica
Fac. de Ciencias Físicas
TRUE
pub