Spin Glass Simulations on the Janus Architecture: A Desperate Quest for Strong Scaling
نویسندگان
چکیده
We describe Janus, an application-driven architecture for Monte Carlo simulations of spin glasses. Janus is a massively parallel architecture, based on reconfigurable FPGA nodes; it offers two orders of magnitude better performance than commodity systems for spin glass applications. The first generation Janus machine has been operational since early 2008; we are currently developing a new generation, that will be on line in early 2013. In this paper we present the Janus architecture, describe both implementations and compare their performances with those of commodity systems.
منابع مشابه
Reconfigurable computing for Monte Carlo simulations: results and prospects of the Janus project
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 (a...
متن کاملThermodynamic glass transition in a spin glass without time-reversal symmetry.
Spin glasses are a longstanding model for the sluggish dynamics that appear at the glass transition. However, spin glasses differ from structural glasses in a crucial feature: they enjoy a time reversal symmetry. This symmetry can be broken by applying an external magnetic field, but embarrassingly little is known about the critical behavior of a spin glass in a field. In this context, the spac...
متن کاملEffect of Asymmetric Functionalized Graphene Oxide (Janus GO) on Young′s Modulus and Glass Transition Temperature of PSf Ultrafiltration Membrane
In this study, effect of asymmetric functionalized graphene oxide (Janus GO) on Young′s modulus and glass transition temperature of Polysulfone (PSf) ultrafiltration membranes was investigated. The membranes were prepared via phase inversion method and GO nanosheets were dispersed in casting solution by sonication. Results showed that the Normalized Young’s modulus (on the basis of neat ...
متن کاملO ct 1 99 4 Numerical simulations on the 4 d Heisenberg spin glass
We study the 4d Heisenberg spin glass model with Gaussian nearestneighbor interactions. We use finite size scaling to analyze the data. We find a behavior consistent with a finite temperature spin glass transition. Our estimates for the critical exponents agree with the results from ε-expansion. PACS Numbers 7510N-0270 Preprint Roma1 n 1054
متن کاملMonte Carlo Simulations of Spin Systems on Multi-core Processors
We implement Monte Carlo algorithms for the simulation of spin-glass systems and optimize our codes for recent multi-core CPU and GPU architectures. We consider both the Ising (binary) and Heisenberg (floating-point) spin-glass models. We provide performance figures for the Intel Nehalem and the IBM Cell/BE CPUs and the Nvidia Tesla C1060 GPU; we also draw a comparison with the performance of d...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2012