SEARCH · Search NASA
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
In this contribution, we report on recent progress in coupled-cluster simulations of open-shell atomic nuclei using interactions consistently derived from chiral effective field theory. In particular, we compare different coupled-cluster approaches by computing binding energies and electric dipole polarizabilities in medium-mass calcium isotopes.
As compute clusters continue to grow in scale and complexity, the frequency of detected anomalies in their operation significantly increases. Timely detection of anomalous events is vital to maintain system efficiency and availability. This study presents an attentionbased graph neural network (GNN) for detecting anomalies in clusters at the compute node level and for providing detailed root cause analysis. We show the effectiveness of attention-based GNNs to accurately detect and localize anomalies on real-world datasets.
Monitoring the status of a high throughput computing cluster running computationally intensive production jobs is a crucial yet challenging system administration task due to the complexity of such systems. To this end, we train autoencoders using the Linux kernel CPU metrics of the cluster. Additionally, we explore assisting these models with graph neural networks to share information across threads within a compute node. The models are compared in terms of their ability to: 1) Produce a compressed latent representation that captures the salient features of the input, 2) Detect anomalous activity, and 3) Make distinction between different kinds of jobs run at Jefferson Lab. The goal is to have a robust encoder whose compressed embeddings are used for several downstream tasks. We extend this study further by deploying these models in a human-in-the-loop production-based setting for the anomaly detection task and discuss the associated implementation aspects such as continual learning and the criterion to generate alarms. This study represents a first step in the endeavor towards building self-supervised large-scale foundation models for computing centers.
Abstract Background The Basic Local Alignment Search Tool (BLAST) is a suite of commonly used algorithms for identifying matches between biological sequences. The user supplies a database file and query file of sequences for BLAST to find identical sequences between the two. The typical millions of database and query sequences make BLAST computationally challenging but also well suited for parallelization on high-performance computing clusters. The efficacy of parallelization depends on the data partitioning, where the optimal data partitioning relies on an accurate performance model. In previous studies, a BLAST job was sped up by 27 times by partitioning the database and query among thousands of processor nodes. However, the optimality of the partitioning method was not studied. Unlike BLAST performance models proposed in the literature that usually have problem size and hardware configuration as the only variables, the execution time of a BLAST job is a function of database size, query size, and hardware capability. In this work, the nucleotide BLAST application BLASTN was profiled using three methods: shell-level profiling with the Unix “time” command, code-level profiling with the built-in “profiler” module, and system-level profiling with the Unix “gprof” program. The runtimes were measured for six node types, using six different database files and 15 query files, on a heterogeneous HPC cluster with 500+ nodes. The empirical measurement data were fitted with quadratic functions to develop performance models that were used to guide the data parallelization for BLASTN jobs. Results Profiling results showed that BLASTN contains more than 34,500 different functions, but a single function, RunMTBySplitDB, takes 99.12% of the total runtime. Among its 53 child functions, five core functions were identified to make up 92.12% of the overall BLASTN runtime. Based on the performance models, static load balancing algorithms can be applied to the BLASTN input data to minimize the runtime of the longest job on an HPC cluster. Four test cases being run on homogeneous and heterogeneous clusters were tested. Experiment results showed that the runtime can be reduced by 81% on a homogeneous cluster and by 20% on a heterogeneous cluster by re-distributing the workload. Discussion Optimal data partitioning can improve BLASTN’s overall runtime 5.4-fold in comparison with dividing the database and query into the same number of fragments. The proposed methodology can be used in the other applications in the BLAST+ suite or any other application as long as source code is available.
Coupled-cluster methods provide highly accurate models of molecular structure by explicit numerical calculation of tensors representing the correlation between electrons. These calculations are dominated by a sequence of tensor contractions, motivating the development of numerical libraries for such operations. While based on matrix-matrix multiplication, these libraries are specialized to exploit symmetries in the molecular structure and in electronic interactions, and thus reduce the size of the tensor representation and the complexity of contractions. The resulting algorithms are irregular and their parallelization has been previously achieved via the use of dynamic scheduling or specialized data decompositions. We introduce our efforts to extend the Libtensor framework to work in the distributed memory environment in a scalable and energy efficient manner. We achieve up to 240 speedup compared with the best optimized shared memory implementation. We attain scalability to hundreds of thousands of compute cores on three distributed-memory architectures, (Cray XC30&XC40, BlueGene/Q), and on a heterogeneous GPU-CPU system (Cray XK7). As the bottlenecks shift from being compute-bound DGEMM's to communication-bound collectives as the size of the molecular system scales, we adopt two radically different parallelization approaches for handling load-imbalance. Nevertheless, we preserve a uni ed interface to both programming models to maintain the productivity of computational quantum chemists.
Coupled-cluster methods provide highly accurate models of molecular structure through explicit numerical calculation of tensors representing the correlation between electrons. These calculations are dominated by a sequence of tensor contractions, motivating the development of numerical libraries for such operations. While based on matrix–matrix multiplication, these libraries are specialized to exploit symmetries in the molecular structure and in electronic interactions, and thus reduce the size of the tensor representation and the complexity of contractions. The resulting algorithms are irregular and their parallelization has been previously achieved via the use of dynamic scheduling or specialized data decompositions. We introduce our efforts to extend the Libtensor framework to work in the distributed memory environment in a scalable and energy-efficient manner. We achieve up to 240× speedup compared with the optimized shared memory implementation of Libtensor. We attain scalability to hundreds of thousands of compute cores on three distributed-memory architectures (Cray XC30 and XC40, and IBM Blue Gene/Q), and on a heterogeneous GPU-CPU system (Cray XK7). As the bottlenecks shift from being compute-bound DGEMM's to communication-bound collectives as the size of the molecular system scales, we adopt two radically different parallelization approaches for handling load-imbalance, tasking and bulk synchronous models. Nevertheless, we preserve a unified interface to both programming models to maintain the productivity of computational quantum chemists.
The stability and distributions of small water clusters generated in a supersonic beam expansion are interrogated by tunable vacuum ultraviolet (VUV) radiation generated at a synchrotron. Time-of-flight mass spectrometry reveals enhanced population of various protonated water clusters (H + (H 2 O) $n$ ) based upon ionization energy and photoionization distance from source, suggesting there are "magic" numbers below the traditional $n$ = 21 that predominates in the literature. These intensity distributions suggest that VUV threshold photoionization (11.0-11.5 eV) of neutral water clusters close to the nozzle exit leads to a different nonequilibrium state compared to a skimmed molecular beam. This results in the appearance of a new magic number at 14. Metadynamics conformer searches coupled with modern density functional calculations are used to identify the global minimum energy structures of protonated water clusters between $n$ = 2 and 21, as well as the manifold of low-lying metastable minima. New lowest energy structures are reported for the cases of $n$ = 5, 6, 11, 12, 16, and 18, and special stability is identified by several measures. These theoretical results are in agreement with the experiments performed in this work in that $n$ = 14 is shown to exhibit additional stability, based on the computed second-order stabilization energy relative to most cluster sizes, though not to the extent of the well-known $n$ = 21 cluster. Other cluster sizes that show some additional energetic stability are $n$ = 7, 9, 12, 17, and 19. To gain insight into the balance between ion-water and water-water interactions as a function of the cluster size, an analysis of the effective two-body interactions (which sum exactly to the total interaction energy) was performed. In conclusion, this analysis reveals a crossover as a function of cluster size between a water-hydronium-dominated regime for small clusters and a water-water-dominated regime for larger clusters around $n$ = 17.
The recently introduced idea of identifying the most important higher-than-doubly excited determinants in the ground-state coupled-cluster (CC) calculations through stochastic configuration interaction Quantum Monte Carlo propagations [J.E. Deustua, J. Shen, and P. Piecuch, Phys. Rev. Lett. 119, 223003 (2017)] is extended to excited electronic states via the equation-of-motion (EOM) CC methodology. The advantages of the new method are illustrated by calculations aimed at recovering the ground- and excited-state energies of the CH + molecule at the equilibrium and stretched geometries resulting from the EOMCC calculations with a full treatment of singles, doubles, and triples.
A data protection system is provided that allows applications to access protected data in a way that restricts applications from outputting to unauthorized targets any unprotected data derived from the protected data and that ensures that the applications do not have access to a key that allows access to the unprotected data. The data protection system provides a policy server that may execute on a service node of a high performance computing system and a data encryption process that may execute on each compute node that is allocated to an application or batch job. The policy server maintains policies of entities specifying access control for protected data. The data encryption process generates a secure execution environment for an application process and interfaces with the policy server to retrieve keys for decrypting protected data in accordance with a policy, and it decrypts and provides the decrypted data to the application process.
Not Available
We have developed a set of reduced, proxy applications ("MiniApps") based on large-scale application codes supported at the Oak Ridge Leadership Computing Facility (OLCF). The MiniApps are designed to encapsulate the details of the most important (i.e. the most time-consuming and/or unique) facets of the applications that run in production mode on the OLCF. In each case, we have produced or plan to produce individual versions of the MiniApps using different specific programming models (e.g., OpenACC, CUDA, OpenMP). We describe some of our initial observations regarding these different implementations along with estimates of how closely the MiniApps track the actual performance characteristics (in particular, the overall scalability) of the large-scale applications from which they are derived.
Improving read performance is one of the major challenges with speeding up scientific data analytic applications. Utilizing the memory hierarchy is one major line of researches to address the read performance bottleneck. Related methods usually combine solide-state-drives(SSDs) with dynamic random-access memory(DRAM) and/or parallel file system(PFS) to mitigate the speed and space gap between DRAM and PFS. However, these methods are unable to handle key performance issues plaguing SSDs, namely read contention that may cause up to 50% performance reduction. In this paper, we propose a framework that exploits the memory hierarchy resource to address the read contention issues involved with SSDs. The framework employs a general purpose online read algorithm that able to detect and utilize memory hierarchy resource to relieve the problem. To maintain a near optimal operating environment for SSDs, the framework is able to orchastrate data chunks across different memory layers to facilitate the read algorithm. Compared to existing tools, our framework achieves up to 50% read performance improvement when tested on datasets from real-world scientific simulations.
The collective communication operations, which are widely used in parallel applications for global communication and synchronization are critical for application's performance and scalability. However, how faulty collective communications impact the application and how errors propagate between the application processes is largely unexplored. One of the critical reasons for this situation is the lack of fast evaluation method to investigate the impacts of faulty collective operations. The traditional random fault injection methods relying on a large amount of fault injection tests to ensure statistical significance require a significant amount of resources and time. These methods result in prohibitive evaluation cost when applied to the collectives. In this paper, we introduce a novel tool named Fast Fault Injection and Sensitivity Analysis Tool (FastFIT) to conduct fast fault injection and characterize the application sensitivity to faulty collectives. The tool achieves fast exploration by reducing the exploration space and predicting the application sensitivity using Machine Learning (ML) techniques. A basis for these techniques are implicit correlations between MPI semantics, application context, critical application features, and application responses to faulty collective communications. The experimental results show that our approach reduces the fault injection points and tests by 97% for representative benchmarks (NAS Parallel Benchmarks (NPB)) and a realistic application (Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS)) on a production supercomputer. Further, we statistically generalize the application sensitivity to faulty collective communications for these workloads, and present correlation between application features and the sensitivity.
The growing computing power on leadership HPC systems is often accompanied by ever-escalating failure rates. Checkpointing is a common defensive mechanism used by scientific applications for failure recovery. However, directly writing the large and bursty checkpointing dataset to parallel file systems can incur significant I/O contention on storage servers. Such contention in turn degrades bandwidth utilization of storage servers and prolongs the average job I/O time of concurrent applications. Recently burst buffers have been proposed as an intermediate layer to absorb the bursty I/O traffic from compute nodes to storage backend. But an I/O orchestration mechanism is still desirable to efficiently move checkpointing data from burst buffers to storage backend. In this paper, we propose a burst buffer based I/O orchestration framework, named TRIO, to intercept and reshape the bursty writes for better sequential write traffic to storage servers. Meanwhile, TRIO coordinates the flushing orders among concurrent burst buffers to alleviate the contention on storage server. Our experimental results demonstrated that TRIO could efficiently utilize storage bandwidth and reduce the average job I/O time by 37% on average for data-intensive applications in typical checkpointing scenarios.
Hybrid MPI + OpenMP is a popular means of programming modern machines that feature substantial parallelism both off-node and on-node. Determining the right mix of the two programming models to use, however, is not as straightforward as simply using exclusively OpenMP on-node and limiting MPI to only inter-node communication. We present a step-by-step methodology to help make the decision of which mix of the two programming models to use. It starts with an estimate of the performance of a generic hybrid application on a given machine and incorporates additional available information about the specific application and the machine to provide guidance for selecting effective mixes of MPI processes and OpenMP threads to use when running that application on the machine in question. We validate our approach on four different applications on an IBM Blue Gene/Q, a Cray XK7, and a Cray XC30.
Results from work on feedback control of a queuing system for HPC jobs with shared storage resources.
Collection of tools and code for evaluation testing of HPC orchestration solutions by HPCCEA summer students. These scripts and configuration files were written by HPCCEA summer students as part of their summer internship where they explore alternative solutions to our current HPC cluster management software. These scripts/configuration files range from using configuration management tools (Puppet, Chef, Ansible, Fabric, etc.) to scripting languages (Phython, BASH, Perl, etc.). Each year, new students will be added to the project and expand on previous years work.