Modern High Performance Computing (HPC) sys-tems are built with innovative system architectures and novel programming models to further push the speed limit of computing. The increased complexity poses challenges for performance portability and performance evaluation. The Standard Perfor-mance Evaluation Corporation (SPEC) has a long history of producing industry-standard benchmarks for modern computer systems. SPEC's newly released SPEChpc 2021 benchmark suites, developed by the High Performance Group, are a bold attempt to provide a fair and objective benchmarking tool designed for state-of-the-art HPC systems. With the support of multiple host and accelerator programming models, the suites are portable across both homogeneous and heterogeneous architectures. Different workloads are developed to fit system sizes ranging from a few compute nodes to a few hundred compute nodes. In this work we present our first experiences in performance benchmarking the new SPEChpc2021 suites and evaluate their portability and basic performance characteristics on various popular and emerging HPC architectures, including x86 CPU, NVIDIA GPU, and AMD GPU. This study provides a first-hand experience of executing the SPEChpc 2021 suites at scale on production HPC systems, discusses real-world use cases, and serves as an initial guideline for using the benchmark suites.

Data analytics is undergoing a revolution in many scientific domains, and demands cost-effective parallel data analysis techniques. Traditional Java-based Big Data processing tools like Hadoop MapReduce are designed for commodity CPUs. In contrast, emerging manycore processors like the Xeon Phi have an order of magnitude greater computation power and memory bandwidth. To harness their computing capabilities, we propose the Harp-DAAL framework. We show that enhanced versions of MapReduce can be replaced by Harp, a Hadoop plug-in, that offers useful data abstractions for both high-performance iterative computation and MPI-quality communication, as well as drive Intel's native DAAL library. We select a subset of three machine learning algorithms and implement them within Harp-DAAL. Our scalability benchmarks ran on Knights Landing (KNL) clusters and achieved up to 2.5 times speedup of performance over the HPC solution in NOMAD and 15 to 40 times speedup over Java-based solutions in Spark. We further quantify the workloads on single node KNL with a performance breakdown at the micro-architecture level.

Latent Dirichlet Allocation (LDA) is a widely used machine learning technique in topic modeling and data analysis. Training large LDA models on big datasets involves dynamic and irregular computation patterns and is a major challenge to both algorithm optimization and system design. In this paper, we present a comprehensive benchmarking of our novel synchronized LDA training system HarpLDA+ based on Hadoop and Java. It demonstrates impressive performance when compared to three other MPI/C++ based state-of-the-art systems, which are LightLDA, F+NomadLDA, and WarpLDA. HarpLDA+ uses optimized collective communication with a timer control for load balance, leading to stable scalability in both shared-memory and distributed systems. We demonstrate in the experiments that HarpLDA+ is effective in reducing synchronization and communication overhead and outperforms the other three LDA training systems.

SUMMARY The computational demands of an ever increasing number of scholars at universities and research institutions throughout the country are outgrowing the capacity of desktop workstations. Researchers are turning to high performance computing facilities, both on their campuses and at regional and national centers, to run simulations and analyze data. The Extreme Science and Engineering Discovery Environment (XSEDE) is one of the first places researchers turn to when they outgrow their campus resources. XSEDE machines are far larger (by at least an order of magnitude) than what most universities offer. Transitioning from a campus resource to an XSEDE resource is seldom a trivial task. XSEDE has taken many steps to make this transition easier, including the campus bridging initiative, the Campus Champions program, and the Extended Collaborative Support Service program. In this paper, we present a new facet to the campus bridging initiative in the form of the campus bridging expert, an information technology professional dedicated to aid researchers in transitioning from desktop, to campus, to regional, and to national resources. We outline the current state of affairs and explore how campus bridging experts could provide maximal impact for minimal investment on the part of the organizing body. Copyright © 2014 John Wiley & Sons, Ltd.

Digital instruments and simulations are creating an ever-increasing amount of data. The need for institutions to acquire these data and transfer them for analysis, visualization, and archiving is growing as well. In parallel, networking technology is evolving, but at a much slower rate than our ability to create and store data. Single fiber 100 Gbps networking solutions have recently been deployed as national infrastructure. This article describes our experiences with data movement and video conferencing across a networking testbed, using the first commercially available single fiber 100 Gbps technology. The testbed is unique in its ability to be configured for a total length of 60, 200, or 400 km, allowing for tests with varying network latency. We performed low-level TCP tests and were able to use more than 99.9% of the theoretical available bandwidth with minimal tuning efforts. We used the Lustre file system to simulate how end users would interact with a remote file system over such a high performance link. We were able to use 94.4% of the theoretical available bandwidth with a standard file system benchmark, essentially saturating the wide area network. Finally, we performed tests with H.323 video conferencing hardware and quality of service (QoS) settings, showing that the link can reliably carry a full high-definition stream. Overall, we demonstrated the practicality of 100 Gbps networking and Lustre as excellent tools for data management.

As part of the SCinet Research Sandbox at the IEEE/ACM International Conference for High Performance Computing, Networking, Storage and Analysis (SC11), Indiana University utilized a dedicated 100 Gbps wide area network (WAN) link spanning more than 3,500 km (2,175 mi) to demonstrate the capabilities of the Lustre high performance parallel file system in a high bandwidth, high latency WAN environment. This demonstration functioned as a proof of concept and provided an opportunity to study Lustre's performance over a 100 Gbps WAN. To characterize the performance of the network and file system a series of benchmarks and tests were undertaken. These included low level iperf network tests, Lustre networking tests, file system tests with the IOR benchmark, and a suite of real-world applications reading and writing to the file system. All of the tests and benchmarks were run over a the WAN link with a latency of 50.5 ms. In this article we describe the configuration and constraints of the demonstration and focus on the key findings regarding the networking layer for this extremely high bandwidth and high latency connection. Of particular interest are the challenges presented by link aggregation for a relatively small number of high bandwidth connections, and the specifics of virtual local area network routing for 100 Gbps routing elements.

As part of the SCinet Research Sandbox at the Supercomputing 2011 conference, Indiana University (IU) demonstrated use of the Lustre high performance parallel file system over a dedicated 100 Gbps wide area network (WAN) spanning more than 3,500 km (2,175 mi). This demonstration functioned as a proof of concept and provided an opportunity to study Lustre's performance over a 100 Gbps WAN. To characterize the performance of the network and file system, low level iperf network tests, file system tests with the IOR benchmark, and a suite of real-world applications reading and writing to the file system were run over a latency of 50.5 ms. In this article we describe the configuration and constraints of the demonstration and outline key findings.

As Cloud computing emerges as a dominant paradigm in distributed systems, it is important to fully understand the underlying technologies that make Clouds possible. One technology, and perhaps the most important, is virtualization. Recently virtualization, through the use of hyper visors, has become widely used and well understood by many. However, there are a large spread of different hyper visors, each with their own advantages and disadvantages. This paper provides an in-depth analysis of some of today's commonly accepted virtualization technologies from feature comparison to performance analysis, focusing on the applicability to High Performance Computing environments using Future Grid resources. The results indicate virtualization sometimes introduces slight performance impacts depending on the hyper visor type, however the benefits of such technologies are profound and not all virtualization technologies are equal. From our experience, the KVM hyper visor is the optimal choice for supporting HPC applications within a Cloud infrastructure.

In this article we propose a revolutionary new paradigm for the processing and analysis of astronomical image data. We describe a blueprint for a centralized data repository and processing system, which leverages national cyber infrastructure. Included is a brief discussion of the current paradigm in astronomical image processing. The upcoming One Degree Imager (ODI) instrument, to be installed at the Wisconsin, Indiana, Yale and NOAO (WIYN) observatory in 2011, is examined as an ideal use case. Details on the major components and a detailed workflow in the case of data processing for the ODI instrument are highlighted.

Astrophysical simulations of protoplanetary disks and gas giant planet formation are being performed with a variety of numerical methods. Some of the codes in use today have been producing scientifically significant results for several years, or even decades. Each must simulate millions of resolution elements for millions of time steps, capture and store output data, and rapidly and efficiently analyze this data. To do this effectively, a parallel code is needed that scales to tens or hundreds of processors. Furthermore, an efficient workflow for the transport, analysis, and interpretation of the output data is needed to achieve scientifically meaningful results.

The growing size of sequence, protein and other biological databases results in an increased computational complexity of the analysis process. Often parallelization is the only solution to limit the turnaround time within reasonable limits. Most scalability studies focus on the parallel algorithm and the resulting communication and synchronization patterns of the implementations. In this paper we examine to what extend I/O bottlenecks limit the scalability on current and future architectures. We study the behavior of two different bioinformatics applications (THREADER, HMMER) and show that these applications are representatives of two different classes with distinct I/O profiles and demands.