Robert Kim Yates

Large-Scale First-Principles Molecular Dynamics simulations on the BlueGene/L Platform using the Qbox code

We demonstrate that the Qbox code supports unprecedented large-scale First-Principles Molecular Dynamics (FPMD) applications on the BlueGene/L supercomputer. Qbox is an FPMD implementation specifically designed for large-scale parallel platforms such as BlueGene/L. Strong scaling tests for a Materials Science application show an 86% scaling efficiency between 1024 and 32,768 CPUs. Measurements of performance by means of hardware counters show that 36% of the peak FPU performance can be attained.

Networks of Real-Time Processes

The input-output function computed by a network of asynchronous real-time processes is proved to be identical to the unique fixed point of a network functional even though the components of the network may compute nonmonotonic functions. The techniques used are those of contractive functions on metric spaces rather than the usual Scott continuity on partial orders. Thus a well-known principle of Kahn is extended to an important model of parallel systems that has been resistant to the traditional approach using Scott continuity.

Simulating solidification in metals at high pressure: The drive to petascale computing

We investigate solidification in metal systems ranging in size from 64,000 to 524,288,000 atoms on the IBM BlueGene/L computer at LLNL. Using the newly developed ddcMD code, we achieve performance rates as high as 103 TFlops, with a performance of 101.7 TFlop sustained over a 7 hour run on 131,072 cpus. We demonstrate superb strong and weak scaling. Our calculations are significant as they represent the first atomic-scale model of metal solidification to proceed, without finite size effects, from spontaneous nucleation and growth of solid out of the liquid, through the coalescence phase, and into the onset of coarsening. Thus, our simulations represent the first step towards an atomistic model of nucleation and growth that can directly link atomistic to mesoscopic length scales.

Performance of the IBM general parallel file system

We measure the performance and scalability of IBMs General Parallel File System (GPFS) under a variety of conditions. The measurements are based on benchmark programs that allow us to vary block sizes, access patterns, etc., and to measure aggregate throughput rates. We use the data to give performance recommendations for application development and as a guide to the improvement of parallel file systems.

Tera-Scalable Algorithms for Variable-Density Elliptic Hydrodynamics with Spectral Accuracy

We describe Miranda, a massively parallel spectral/compact solver for variabledensity incompressible flow, including viscosity and species diffusivity effects. Miranda utilizes FFTs and band-diagonal matrix solvers to compute spatial derivatives to at least 10th-order accuracy. We have successfully ported this communicationintensive application to BlueGene/L and have explored both direct block parallel and transpose-based parallelization strategies for its implicit solvers. We have discovered a mapping strategy which results in virtually perfect scaling of the transpose method up to 65,536 processors of the BlueGene/L machine. Sustained global communication rates in Miranda typically run at 85% of the theoretical peak speed of the BlueGene/L torus network, while sustained communication plus computation speeds reach 2.76 TeraFLOPS. This effort represents the first time that a high-order variable-density incompressible flow solver with species diffusion has demonstrated sustained performance in the TeraFLOPS range.

DI: an interactive debugging interpreter for applicative languages

The DI interpreter is both a debugger and interpreter of SISAL programs. Its use as a program interpreter is only a small part of its role; it is designed to be a tool for studying compilation techniques for applicative languages. DI interprets dataflow graphs expressed in the IF1 and IF2 languages, and is heavily instrumented to report the activity of dynamic storage activity, reference counting, copying and updating of structured data values. It also aids the SISAL language evaluation by providing an interim execution vehicle for SISAL programs. DI provides determinate, sequential interpretation of graph nodes for sequential and parallel operations in a canonical order. As a debugging aid, DI allows tracing, breakpointing, and interactive display of program data values. DI handles creation of SISAL and IF1 error values for each data type and propagates them according to a well-defined algebra. We have begun to implement IF1 optimizers and have measured the improvements with DI.

BlueGene/L applications: Parallelism On a Massive Scale

BlueGene/L (BG/L), developed through a partnership between IBM and Lawrence Livermore National Laboratory (LLNL), is currently the worlds largest system both in terms of scale, with 131,072 processors, and absolute performance, with a peak rate of 367 Tflop/s. BG/L has led the last four Top500 lists with a Linpack rate of 280.6 Tflop/s for the full machine installed at LLNL and is expected to remain the fastest computer in the next few editions. However, the real value of a machine such as BG/L derives from the scientific breakthroughs that real applications can produce by successfully using its unprecedented scale and computational power. In this paper, we describe our experiences with eight large scale applications on BG/ L from several application domains, ranging from molecular dynamics to dislocation dynamics and turbulence simulations to searches in semantic graphs. We also discuss the challenges we faced when scaling these codes and present several successful optimization techniques. All applications show excellent scaling behavior, even at very large processor counts, with one code even achieving a sustained performance of more than 100 Tflop/s, clearly demonstrating the real success of the BG/L design.

Scaling physics and material science applications on a massively parallel Blue Gene/L system

Blue Gene/L represents a new way to build supercomputers, using a large number of low power processors, together with multiple integrated interconnection networks. Whether real applications can scale to tens of thousands of processors (on a machine like Blue Gene/L) has been an open question. In this paper, we describe early experience with several physics and material science applications on a 32,768 node Blue Gene/L system, which was installed recently at the Lawrence Livermore National Laboratory. Our study shows some problems in the applications and in the current software implementation, but overall, excellent scaling of these applications to 32K nodes on the current Blue Gene/L system. While there is clearly room for improvement, these results represent the first proof point that MPI applications can effectively scale to over ten thousand processors. They also validate the scalability of the hardware and software architecture of Blue Gene/L.

A Kahn Principle for Networks of Nonmonotonic Real-time Processes

We show that the input-output function computed by a network of asynchronous real-time processes is denoted by the unique fixed point of a Scott continuous functional even though the network or its components may compute a discontinuous function. This extends a well known principle of Kahn to an important class of parallel systems that has resisted the traditional fixed point approach.

Ada syntax diagrams for top-down analysis

In AdaPS ( Ada Programming System a Siemens research project to investigate the efficient implementation of Ada ), the Ada compiler performs top down syntax analysis. The syntax diagrams presented in this paper have been produced in the early steps towards an Ada grammar in extended BNF to be used as input for a syntax analyser generator. The syntax diagrams have been designed in such a way that an LL grammar can be easily derived from them.