Summary The newest NERSC supercomputer Cori is a Cray XC40 system consisting of 2,388 Intel Xeon Haswell nodes and 9,688 Intel Xeon‐Phi “Knights Landing” (KNL) nodes. Compared to the Xeon‐based clusters NERSC users are familiar with, optimal performance on Cori requires consideration of KNL mode settings; process, thread, and memory affinity; fine‐grain parallelization; vectorization; and use of the high‐bandwidth MCDRAM memory. This paper describes our efforts preparing NERSC users for KNL through the NERSC Exascale Science Application Program, Web documentation, and user training. We discuss how we configured the Cori system for usability and productivity, addressing programming concerns, batch system configurations, and default KNL cluster and memory modes. System usage data, job completion analysis, programming and running jobs issues, and a few successful user stories on KNL are presented.

In this poster, we describe how MPI is used at the National Energy Research Scientific Computing Center (NERSC) NERSC is the production high-performance computing center for the US Department of Energy, with more than 5000 users and 800 distinct projects. Through a variety of tools (e.g., User Survey, application team collaborations, etc.), we determine how MPI is used on our latest systems, with a particular focus on advanced features and how early applications intend to use MPI on NERSC's upcoming Intel Knights Landing (KNL) many-core system1 - one of the first to be deployed. In the poster, we also compare the usage of MPI to exascale developmental programming models such as UPC++ and HPX, with an eye on what features and extensions to MPI are plausible and useful for NERSC users. We also discuss perceived shortcomings of MPI, and why certain groups use other parallel programming models on the systems. In addition to a broad survey of the NERSC HPC population, we follow the evolution of a few key application codes2 that are being highly optimized for the KNL architecture using advanced OpenMP techniques. We study how these highly optimized on-node proxy apps and full applications start to make the transition to using full hybrid MPI+OpenMP implementations on the self-hosted KNL system.

We present novel parallel language constructs for the communication intensive part of a magnetic fusion simulation code. The focus of this work is the shift phase of charged particles of a tokamak simulation code in toroidal geometry. We introduce new hybrid PGAS/OpenMP implementations of highly optimized hybrid MPI/OpenMP based communication kernels. The hybrid PGAS implementations use an extension of standard hybrid programming techniques, enabling the distribution of high communication work loads of the underlying kernel among OpenMP threads. Building upon lightweight one-sided CAF (Fortran 2008) communication techniques, we also show the benefits of spreading out the communication over a longer period of time, resulting in a reduction of bandwidth requirements and a more sustained communication and computation overlap. Experiments on up to 130560 processors are conducted on the NERSC Hopper system, which is currently the largest HPC platform with hardware support for one-sided communication and show performance improvements of 52 % at highest concurrency.

The HPC Challenge (HPCC) benchmark suite and the Intel MPI Benchmark (IMB) are used to compare and evaluate the combined performance of processor, memory subsystem and interconnect fabric of five leading supercomputers - SGI Altix BX2, Cray XI, Cray Opteron Cluster, Dell Xeon cluster, and NEC SX-8. These five systems use five different networks (SGI NUMALINK4, Cray network, Myrinet, InfiniBand, and NEC IXS). The complete set of HPCC benchmarks are run on each of these systems. Additionally, we present Intel MPI Benchmarks (IMB) results to study the performance of 11 MPI communication functions on these systems

Teraflop performance is no longer something of the future as complex integrated and multiscale 3D simulations drive supercomputer development. This tutorial addresses computation at the highest end. An overview of architectures is given (BlueGene/L, Columbia, NEC SX-8, Cray and IBM lines, high-performing clusters) along with programming tools necessary for application development. Parallel programming concepts (MPI, OpenMP, HPF, UPC, CAF) are reviewed and compared. What are the major issues facing application code developers today? How do the challenges vary from cluster computing to the complex hybrid architectures with superscalar and vector processors? What are the barriers we must overcome to achieve true sustained petascale performance? We address these questions and give tips, tricks, and tools of the trade for large mulitscale application development. We discuss some advanced MPI including dynamic process management and optimization. We draw from a series of terascale and multiscale applications and discuss specific challenges and performance issues.

We describe the design and MPI implementation of two benchmarks created to characterize the balanced system performance of high-performance clusters and supercomputers. We start with a communication-specific benchmark, called b_eff that characterizes the message passing performance of a system. Following the same line of development, we extend this work to the design and implementation of the effective I/O bandwidth benchmark (b_eff_io). Both of these benchmarks have two goals: a) to obtain a single bandwidth number that characterizes the average performance of the system namely processor communication for b_eff, and the I/O subsystem for b_eff_io, and b) to get a detailed insight into the performance strengths and weaknesses of different parallel communication and I/O patterns. Both benchmarks use a time-driven approach and loop over a variety of communication and access patterns to characterize a system in a fairly automated fashion. Results of the two benchmarks are given for several systems including IBM SPs, Cray T3E, NEC SX-5, and Hitachi SR 8000.

MPI-IO/GPFS is an optimized prototype implementation of the I/O chapter of the Message Passing Interface (MPI) 2 standard. It uses the IBM General Parallel File System (GPFS) Release 3 as the underlying file system. This paper describes optimization features of the prototype that take advantage of new GPFS programming interfaces. It also details how collective data access operations have been optimized by minimizing the number of messages exchanged in sparse accesses and by increasing the overlap of communication with file access. Experimental results show a performance gain. A study of the impact of varying the number of tasks running on the same node is also presented.

We measure the performance and scalability of IBM's 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.