Summary The Cray HPE Slingshot 11 network is used on the new exascale systems arriving at the U.S. Department of Energy (DoE) laboratories (e.g., Frontier, Aurora, Perlmutter). As such, the support of this network is an important capability to meet the needs of exascale applications. This article highlights recent work to develop supporting infrastructure to enable Open MPI to efficiently support these new platforms. A key component of this effort involves development of a new Open Fabrics Interface (OFI) provider, LinkX. We discuss the design and development of enhancements that take advantage of the new Slingshot 11 network and AMD GPUs. We include performance data from tests on the Frontier supercomputer using synthetic communication benchmarks, and the vendor provided MPI as a baseline for comparison. The tests demonstrate full functionality of Open MPI on the system and initial results show favorable performance when compared to the highly tuned vendor implementation.

This paper describes an unplugged computational thinking (CT) resource for primary and secondary schools developed from Bebras tasks. In Ireland, CT is not part of the primary school curriculum or mandatory in secondary schools. However, the National Council for Curriculum and Assessment is in the process of revising the primary school curriculum to include aspects of CT. Our aim for creating this CT Obstacle Course is to introduce teachers (and pupils) without formal computer science training to the subject of CT. This is done in a manner that informs and motivates, and gives them the confidence to deliver CT materials in the classroom. We also want to find out from teachers how useful and important this type of resource is for developing problem-solving skills, and if our unplugged activity can support learning at various skill levels. Our CT Obstacle Course includes 14 Bebras tasks for primary schools and an additional 6 Bebras tasks for secondary schools. The activity is suitable for indoors and outdoors and is completed in groups, promoting teamwork and communication. We have delivered it to 146 primary school classes during 38 school visits between May 2021 and June 2022. It has been undertaken by 3,445 pupils and 195 teachers and other school staff. This paper describes our CT resource in detail, and reports teacher feedback from primary schools.

In this poster, we highlight computational thinking resources for schools from the PACT team at Maynooth University, Ireland. The resources are derived from tasks from the Bebras international computational thinking initiative. The different modalities work together throughout the school year to provide initial exposure to computational thinking, and include an obstacle course, seasonal tasks, and a workbook.

This paper describes our process of creating computational thinking (CT) resources for primary school teachers in Ireland. The National Council for Curriculum and Assessment has proposed a revised primary mathematics curriculum with an emphasis on CT skills and problem solving, and some teachers would like to introduce it already on an informal basis. However, CT is not yet part of teacher training. Our motivating question has been: how can teachers without a computer science background deliver CT at primary level in Ireland? Our process involves third-level computer science academics co-creating resources with in-service and pre-service teachers during workshops. The resources comprise a workbook and lesson plans. Our resources are based on tasks from the International Bebras Challenge, a well-known large-scale international CT contest with a reasonably gender-neutral profile of school-age participants. The workbook consists of ten Bebras tasks, each followed by a page of original activities on the theme of the task. A set of ten lesson plans accompanies the workbook. Each lesson plan has information about how to use the corresponding workbook activities in the classroom, where the activity might fit into the existing curriculum, categorisation of the task in terms of eight CT topics, differentiation, and extension activities. This paper explains our process of workshop planning, workbook creation, and lesson plan co-creation. Preliminary evaluation of our process uses teacher feedback.

Softwarization of networked infrastructures combined with containerization of codes promises unprecedented computing capabilities distributed across the federations of computing systems and physical instruments. The development and testing of a software stack that implements these capabilities over an expensive physical production infrastructure is not cost-effective, and in the early stages, may potentially cause service disruptions. To address these aspects, we develop the Virtual Federated Science Instrument Environment (VFSIE), a digital twin of the physical infrastructure that emulates a multi-site federation. Each federated site is emulated using containers and virtual hosts that are connected over local-area networks, and the sites, in turn, are connected over an emulated wide-area network. We describe the framework design and implementation details. We also illustrate its application by emulating a federation of four laboratories that use Jupyter Notebook for computations and the EPICS software system for instrument control.

supercomputers, and beyond. Applications oblivious to and incapable of handling transient soft and hard errors could waste supercomputing resources or, worse, yield misleading scientific insights. We introduce a novel application-driven silent error detection and recovery strategy based on application health monitoring. Our methodology uses application output that follows known patterns as indicators of an application's health, and knowledge that violation of these patterns could be indication of faults. Information from system monitors that report hardware and software health status is used to corroborate faults. Collectively, this information is used by a fault coordinator agent to take preventive and corrective measures by applying computational steering to an application between checkpoints. This cooperative fault management system uses the Fault Tolerance Backplane as a communication channel. The benefits of this framework are demonstrated with two real application case studies, molecular dynamics and quantum chemistry simulations, on scalable clusters with simulated memory and I/O corruptions. The developed approach is general and can be easily applied to other applications.

Summary The Exascale Computing Project (ECP) is currently the primary effort in the United States focused on developing “exascale” levels of computing capabilities, including hardware, software, and applications. In order to obtain a more thorough understanding of how the software projects under the ECP are using, and planning to use the Message Passing Interface (MPI), and help guide the work of our own project within the ECP, we created a survey. Of the 97 ECP projects active at the time the survey was distributed, we received 77 responses, 56 of which reported that their projects were using MPI. This paper reports the results of that survey for the benefit of the broader community of MPI developers.

Summary Investigating the performance of parallel applications at scale on future high‐performance computing (HPC) architectures and the performance impact of different HPC architecture choices is an important component of HPC hardware/software co‐design. The Extreme‐scale Simulator (xSim) is a simulation toolkit for investigating the performance of parallel applications at scale. xSim scales to millions of simulated Message Passing Interface (MPI) processes. The xSim toolkit strives to limit simulation overheads in order to maintain performance and productivity criteria. This paper documents two improvements to xSim: (1) a new deadlock resolution protocol to reduce the parallel discrete event simulation overhead and (2) a new simulated MPI message matching algorithm to reduce the oversubscription management cost. These enhancements resulted in significant performance improvements. The simulation overhead for running the NASA Advanced Supercomputing Parallel Benchmark suite dropped from 1,020% to 238% for the conjugate gradient benchmark and 102% to 0% for the embarrassingly parallel benchmark. Additionally, the improvements were beneficial for reducing overheads in the highly accurate simulation mode of xSim, which is useful for resilience investigation studies for tracking intentional MPI process failures. In the highly accurate mode, the simulation overhead was reduced from 37,511% to 13,808% for conjugate gradient and from 3,332% to 204% for embarrassingly parallel. Copyright © 2016 John Wiley & Sons, Ltd.

Future extreme-scale high-performance computing systems will be required to work under frequent component failures. The MPI Forum's User Level Failure Mitigation proposal has introduced an operation, MPI_Comm_shrink, to synchronize the alive processes on the list of failed processes, so that applications can continue to execute even in the presence of failures by adopting algorithm-based fault tolerance techniques. This MPI_Comm_shrink operation requires a fault tolerant failure detection and consensus algorithm. This paper presents and compares two novel failure detection and consensus algorithms. The proposed algorithms are based on Gossip protocols and are inherently fault-tolerant and scalable. The proposed algorithms were implemented and tested using the Extreme-scale Simulator. The results show that in both algorithms the number of Gossip cycles to achieve global consensus scales logarithmically with system size. The second algorithm also shows better scalability in terms of memory and network bandwidth usage and a perfect synchronization in achieving global consensus.

No abstract available.

Cloud Computing environments rely heavily on system-level virtualization. This is due to the inherent benefits of virtualization including fault tolerance through checkpoint/restart (C/R) mechanisms. Because clouds are the abstraction of large data enters and large data enters have a higher potential for failure, it is imperative that a C/R mechanism for such an environment provide minimal latency as well as a small checkpoint file size. Recently, there has been much research into C/R with respect to virtual machines (VM) providing excellent solutions to reduce either checkpoint latency or checkpoint file size. However, these approaches do not provide both. This paper presents a method of check pointing VMs by utilizing virtual machine introspection (VMI). Through the usage of VMI, we are able to determine which pages of memory within the guest are used or free and are better able to reduce the amount of pages written to disk during a checkpoint. We have validated this work by using various benchmarks to measure the latency along with the checkpoint size. With respect to checkpoint file size, our approach results in file sizes within 24% or less of the actual used memory within the guest. Additionally, the checkpoint latency of our approach is up to 52% faster than KVM's default method.

Investigating the performance of parallel applications at scale on future high-performance computing (HPC) architectures and the performance impact of different architecture choices is an important component of HPC hardware/software co-design. The Extreme-scale Simulator (xSim) is a simulation-based toolkit for investigating the performance of parallel applications at scale. xSim scales to millions of simulated Message Passing Interface (MPI) processes. The overhead introduced by a simulation tool is an important performance and productivity aspect. This paper documents two improvements to xSim: (1) a new deadlock resolution protocol to reduce the parallel discrete event simulation management overhead and (2) a new simulated MPI message matching algorithm to reduce the oversubscription management overhead. The results clearly show a significant performance improvement, such as by reducing the simulation overhead for running the NAS Parallel Benchmark suite inside the simulator from 1,020% to 238% for the conjugate gradient (CG) benchmark and from 102% to 0% for the embarrassingly parallel (EP) and benchmark, as well as, from 37,511% to 13,808% for CG and from 3,332% to 204% for EP with accurate process failure simulation.

An emerging aspect of high-performance computing (HPC) hardware/software co-design is investigating performance under failure. The work in this paper extends the Extreme-scale Simulator (xSim), which was designed for evaluating the performance of message passing interface (MPI) applications on future HPC architectures, with fault-tolerant MPI extensions proposed by the MPI Fault Tolerance Working Group. xSim permits running MPI applications with millions of concurrent MPI ranks, while observing application performance in a simulated extreme-scale system using a lightweight parallel discrete event simulation. The newly added features offer user-level failure mitigation (ULFM) extensions at the simulated MPI layer to support algorithm-based fault tolerance (ABFT). The presented solution permits investigating performance under failure and failure handling of ABFT solutions. The newly enhanced xSim is the very first performance tool that supports ULFM and ABFT.

xSim is a simulation-based performance investigation toolkit that permits running high-performance computing (HPC) applications in a controlled environment with millions of concurrent execution threads, while observing application performance in a simulated extreme-scale system for hardware/software co-design. The presented work details newly developed features for xSim that permit the injection of MPI process failures, the propagation/detection/notification of such failures within the simulation, and their handling using application-level checkpoint/restart. These new capabilities enable the observation of application behavior and performance under failure within a simulated future-generation HPC system using the most common fault handling technique.

We present a multi-heuristic evolutionary task allocation algorithm to dynamically map tasks to processors in a heterogeneous distributed system. It utilizes a genetic algorithm, combined with eight common heuristics, in an effort to minimize the total execution time. It operates on batches of unmapped tasks and can preemptively remap tasks to processors. The algorithm has been implemented on a Java distributed system and evaluated with a set of six problems from the areas of bioinformatics, biomedical engineering, computer science and cryptography. Experiments using up to 150 heterogeneous processors show that the algorithm achieves better efficiency than other state-of-the-art heuristic algorithms.

Proactive fault tolerance (FT) in high-performance computing is a concept that prevents compute node failures from impacting running parallel applications by preemptively migrating application parts away from nodes that are about to fail. This paper provides a foundation for proactive FT by defining its architecture and classifying implementation options. This paper further relates prior work to the presented architecture and classification, and discusses the challenges ahead for needed supporting technologies.

In order to address anticipated high failure rates, resiliency characteristics have become an urgent priority for next-generation extreme-scale high-performance computing (HPC) systems. This poster describes our past and ongoing efforts in novel fault resilience technologies for HPC. Presented work includes proactive fault resilience techniques, system and application reliability models and analyses, failure prediction, transparent process- and virtual-machine-level migration, and trade-off models for combining preemptive migration with checkpoint/restart. This poster summarizes our work and puts all individual technologies into context with a proposed holistic fault resilience framework.

Fault tolerance is a major concern to guarantee availability of critical services as well as application execution. Traditional approaches for fault tolerance include checkpoint/restart or duplication. However it is also possible to anticipate failures and proactively take action before failures occur in order to minimize failure impact on the system and application execution. This document presents a proactive fault tolerance framework. This framework can use different proactive fault tolerance mechanisms, i.e., migration and pause/un-pause. The framework also allows the implementation of new proactive fault tolerance policies thanks to a modular architecture. A first proactive fault tolerance policy has been implemented and preliminary experimentations have been done based on system-level virtualization and compared with results obtained by simulation.

System-level virtualization has been a research topic since the 70's but regained popularity during the past few years because of the availability of efficient solution such as Xen and the implementation of hardware support in commodity processors (e.g. Intel-VT, AMD-V). However, a majority of system-level virtualization projects is guided by the server consolidation market. As a result, current virtualization solutions appear to not be suitable for high performance computing (HPC) which is typically based on large-scale systems. On another hand there is significant interest in exploiting virtual machines (VMs) within HPC for a number of other reasons. By visualizing the machine, one is able to run a variety of operating systems and environments as needed by the applications. Virtualization allows users to isolate workloads, improving security and reliability. It is also possible to support non-native environments and/or legacy operating environments through virtualization. In addition, it is possible to balance work loads, use migration techniques to relocate applications from failing machines, and isolate fault systems for repair. This document presents the challenges for the implementation of a system-level virtualization solution for HPC. It also presents a brief survey of the different approaches and techniques to address these challenges.

Various mechanisms for fault-tolerance (FT) are used today in order to reduce the impact of failures on application execution. In the case of system failure, standard FT mechanisms are checkpoint/restart (for reactive FT) and migration (for pro-active FT). However, each of these mechanisms create an overhead on application execution, overhead that for instance becomes critical on large-scale systems where previous studies have shown that applications may spend more time checkpointing state than performing useful work. In order to decrease this overhead, researchers try to both optimize existing FT mechanisms and implement new FT policies. For instance, combining reactive and pro-active approaches in order to decrease the number of checkpoints that must be performed during the application's execution. However, currently no solutions exist which enable the evaluation of these FT approaches through simulation, instead experimentations must be done using real platforms. This increases complexity and limits experimentation into alternate solutions. This paper presents a simulation framework that evaluates different FT mechanisms and policies. The framework uses system failure logs for the simulation with a default behavior based on logs taken from the ASCI White at Lawrence Livermore National Laboratory. We evaluate the accuracy of our simulator comparing simulated results with those taken from experiments done on a 32-node compute cluster. Therefore such a simulator can be used to develop new FT policies and/or to tune existing policies.

We present a novel nonuniform quantization compression technique-histogram quantization-for digital holograms of 3-D real-world objects. We exploit a priori knowledge of the distribution of the values in our data. We compare this technique to another histogram based approach: a modified version of Max's algorithm that has been adapted in a straight-forward manner to complex-valued 2-D signals. We conclude the compression procedure by applying lossless techniques to our quantized data. We demonstrate improvements over previous results obtained by applying uniform and nonuniform quantization techniques to the hologram data.

A distributed Monte Carlo simulation which models the propagation of light through tissue has been developed. It allows for improved calibration of medical imaging devices for investigating tissue oxygenation in the white matter of the cerebral cortex. The application can distribute the simulation over an unbounded number of processors in parallel. We have found that this application is highly parallelisable resulting in up to 91% efficiency at 60 processors running on a homogeneous Java distributed system. A distributed system with 150 heterogeneous processors was used to simulate the paths of photons in a brain tissue model. We found that the source illumination footprint has an effect on the distribution of photons in the head and that lasers do produce a small beam in a highly scattering medium. This application will help researchers to improve the accuracy of their experiments

Since the first public release in 2001, there have been well over 160,000 downloads of the Open Source Cluster Application Resources (OSCAR) software stack. OSCAR is a self-extracting cluster configuration, installation, maintenance, and operation suite consisting of "best known practices" for cluster computing. This BoF will be a focal point for the OSCAR community at SC06 where both developers and users may gather to discuss the "current state" as well as future directions for the OSCAR software stack.

Digital holography is a technique that permits digital capture of holograms and subsequent processing on a digital computer. This paper reviews various applications of this technique. The presented applications cover three-dimensional (3-D) imaging as well as several associated problems. For the case of 3-D imaging, optical and digital methods to reconstruct and visualize the recorded objects are described. In addition, techniques to compress and encrypt 3-D information in the form of digital holograms are presented. Lastly, 3-D pattern recognition applications of digital holography are discussed. The described techniques constitute a comprehensive approach to 3-D imaging and processing.

In the past number of years the demand for high performance computing has greatly increased in the area of bioinformatics. The huge increase in size of many genomic databases has meant that many common tasks in bioinformatics are not possible to complete in a reasonable amount of time on a single processor. Recently distributed computing has emerged as an inexpensive alternative to dedicated parallel computing. We have developed a general-purpose distributed computing platform that is capable of using semi-idle computing resources to simulate a dedicated computing cluster. We have identified the suitability of a number of bioinformatics tasks to distributed computing. We briefly outline and evaluate two distributed bioinformatics programs, DSEARCH and DPRml, which have been developed for our system.

UNLABELLED: We present a distributed and fully cross-platform database search program that allows the user to utilize the idle clock cycles of machines to perform large searches using the most sensitive algorithms. For those in an academic or corporate environment with hundreds of idle desktop machines, DSEARCH can deliver a 'free' database search supercomputer. AVAILABILITY: The software is publicly available under the GNU general public licence from http://www.cs.may.ie/distributed CONTACT: [email protected] SUPPLEMENTARY INFORMATION: Full documentation and a user manual is available from http://www.cs.may.ie/distributed.

MOTIVATION: In recent years there has been increased interest in producing large and accurate phylogenetic trees using statistical approaches. However for a large number of taxa, it is not feasible to construct large and accurate trees using only a single processor. A number of specialized parallel programs have been produced in an attempt to address the huge computational requirements of maximum likelihood. We express a number of concerns about the current set of parallel phylogenetic programs which are currently severely limiting the widespread availability and use of parallel computing in maximum likelihood-based phylogenetic analysis. RESULTS: We have identified the suitability of phylogenetic analysis to large-scale heterogeneous distributed computing. We have completed a distributed and fully cross-platform phylogenetic tree building program called distributed phylogeny reconstruction by maximum likelihood. It uses an already proven maximum likelihood-based tree building algorithm and a popular phylogenetic analysis library for all its likelihood calculations. It offers one of the most extensive sets of DNA substitution models currently available. We are the first, to our knowledge, to report the completion of a distributed phylogenetic tree building program that can achieve near-linear speedup while only using the idle clock cycles of machines. For those in an academic or corporate environment with hundreds of idle desktop machines, we have shown how distributed computing can deliver a 'free' ML supercomputer.