Search NASA⌕ Search

Engineering topics

Ostrouchov, George

Publications and source records attributed to Ostrouchov, George.

At least 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Understanding GPU Memory Corruption at Extreme Scale: The Summit Case Study

GPU memory corruption and in particular double-bit errors (DBEs) remain one of the least understood aspects of HPC system reliability. Albeit rare, their occurrences always lead to job termination and can potentially cost thousands of node-hours, either from wasted computations or as the overhead from regular checkpointing needed to minimize the losses. As supercomputers and their components simultaneously grow in scale, density, failure rates, and environmental footprint, the efficiency of HPC operations becomes both an imperative and a challenge. We examine DBEs using system telemetry data and logs collected from the Summit supercomputer, equipped with 27,648 Tesla V100 GPUs with 2nd-generation high-bandwidth memory (HBM2). Using exploratory data analysis and statistical learning, we extract several insights about memory reliability in such GPUs. We find that GPUs with prior DBE occurrences are prone to experience them again due to otherwise harmless factors, correlate this phenomenon with GPU placement, and suggest manufacturing variability as a factor. On the general population of GPUs, we link DBEs to short- and long-term high power consumption modes while finding no significant correlation with higher temperatures. We also show that the workload type can be a factor in memory’s propensity to corruption.

Oles, Vlad↗
DOE OSTI2023

OLCF Summit Supercomputer GPU Snapshots During Double-Bit Errors and Normal Operations

As we move into the exascale era, the power and energy footprints of high-performance computing (HPC) systems have grown significantly larger. Due to the harsh power and thermal conditions the system, components are exposed to extreme operating conditions. Operation of such modern HPC systems requires deep insights into long term system behavior to maintain its efficiency as well as its longevity. To help the HPC community to gain such insights, we provide double-bit errors using system telemetry data and logs collected from the Summit supercomputer, equipped with 27,648 Tesla V100 GPUs with 2nd-generation high-bandwidth memory (HBM2). The dataset relies on Nvidia XID records internally collected by GPU firmware at the time of failure occurrence, on the reboot-time logs of each Summit node, on node-level job scheduler records collected after each job termination, and on a 1Hz data rate from the baseboard management controllers (BMCs) of each Summit compute node using the OpenBMC event subscription protocol. Technical details can be found in the paper Oles et. al “Understanding GPU Memory Corruption at Extreme Scale: The Summit Case Study” ICS’24 (https://doi.org/10.1145/3650200.3656615).

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Two methods to study inelastic neutron-scattering measurements based on ω n (q) versus S(q, ω) applied to the magnetic open honeycomb lattice Tb 2 Ir 3 Ga 9

This work describes two methods to fit the inelastic neutron-scattering spectrum S(q, ω) with wavevector q and frequency ω. The common and well-established method extracts the experimental spin-wave branches ω n (q) from the measured spectra S(q, ω) and then minimizes the difference between the observed and predicted frequencies. When n branches of frequencies are predicted but the measured frequencies overlap to produce only m < n branches, the weighted average of the predicted frequencies must be compared to the observed frequencies. A penalty is then exacted when the width of the predicted frequencies exceeds the width of the observed frequencies. The second method directly compares the measured and predicted intensities S(q, ω) over a grid {q i , ω j } in wavevector and frequency space. After subtracting background noise from the observed intensities, the theoretical intensities are scaled by a simple wavevector-dependent function that reflects the instrumental resolution. Furthermore, the advantages and disadvantages of each approach are demonstrated by studying the open honeycomb material Tb 2 Ir 3 Ga 9 .

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2021

Clinical knowledge extraction via sparse embedding regression (KESER) with multi-center large scale electronic health record data

The increasing availability of electronic health record (EHR) systems has created enormous potential for translational research. However, it is difficult to know all the relevant codes related to a phenotype due to the large number of codes available. Traditional data mining approaches often require the use of patient-level data, which hinders the ability to share data across institutions. In this project, we demonstrate that multi-center large-scale code embeddings can be used to efficiently identify relevant features related to a disease of interest. We constructed large-scale code embeddings for a wide range of codified concepts from EHRs from two large medical centers. We developed knowledge extraction via sparse embedding regression (KESER) for feature selection and integrative network analysis. We evaluated the quality of the code embeddings and assessed the performance of KESER in feature selection for eight diseases. Besides, we developed an integrated clinical knowledge map combining embedding data from both institutions. The features selected by KESER were comprehensive compared to lists of codified data generated by domain experts. Features identified via KESER resulted in comparable performance to those built upon features selected manually or with patient-level data. The knowledge map created using an integrative analysis identified disease-disease and disease-drug pairs more accurately compared to those identified using single institution data. Analysis of code embeddings via KESER can effectively reveal clinical knowledge and infer relatedness among codified concepts. KESER bypasses the need for patient-level data in individual analyses providing a significant advance in enabling multi-center studies using EHR data.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗
DOE OSTI2020

GPU Lifetimes on Titan Supercomputer: Survival Analysis and Reliability

The Cray XK7 Titan was the top supercomputer system in the world for a long time and remained critically important throughout its nearly seven year life. It was an interesting machine from a reliability viewpoint as most of its power came from 18,688 GPUs whose operation was forced to execute three rework cycles, two on the GPU mechanical assembly and one on the GPU circuitboards. We write about the last rework cycle and a reliability analysis of over 100,000 years of GPU lifetimes during Titan’s 6-year-long productive period. Using time between failures analysis and statistical survival analysis techniques, we find that GPU reliability is dependent on heat dissipation to an extent that strongly correlates with detailed nuances of the cooling architecture and job scheduling. We describe the history, data collection, cleaning, and analysis and give recommendations for future supercomputing systems. We make the data and our analysis codes publicly available.

Ostrouchov, George↗
DOE OSTI2020

GPU Lifetimes on Titan Supercomputer: Survival Analysis and Reliability

George Ostrouchov, Don Maxwell, Rizwan Ashraf, Mallikarjun Shankar, and James Rogers. 2020. GPU Lifetimes on Titan Supercomputer: Survival Analysis and Reliability. In Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis (SC '20). Association for Computing Machinery, New York, NY, USA. Data and code for SC20 paper about Titan GPU reliability analysis: https://github.com/olcf/TitanGPULife. Includes R code to generate graphics for paper and additional analyses. See code/README for instructions. Includes original Titan GPU reliability data on over 100,000 collective hours of operation: data/titan.gpu.history.txt - history data, data/titan.service.txt - service nodes for exclusion. Includes output data files produced by code/TitanGPUmodel.Rmd: data/gc_full.csv - cleaned up data (see paper and R code); data/gc_summary_loc.csv - one record per GPU (variables: SN, time, nlife, nloc, last, col, row, cage, slot, node, max_loc_events, time_max_loc, dbe, dbe_loc, otb, otb_loc, out, batch, days, years, dead, dead_otb, dead_dbe) (see paper and R code). Includes .Rmd analysis document as TitanGPUmode.html. Includes Python code to process data/gc_full.csv into graphics from time-between-failure analyses: See code/tbf-analyses/README for instructions.

42 ENGINEERING↗
DOE OSTI2020

JSONize: A Scalable Machine Learning Pipeline to Model Medical Notes as Semi-structured Documents

The Department of Veteran's Affairs (VA) archives the largest corpora of clinical notes in their corporate data warehouse (CDW) as unstructured text data. Unstructured text easily supports keyword searches and regular expressions. Often these simple searches do not adequately support the complex searches that need to be performed on notes. For example, a researcher may want all notes with a Duke Treadmill Score less than 5 or people that smoke more than 1 pack per day. Range queries like this and more can be supported by modelling text as semi-structured documents. In this paper, we implement a scalable machine learning pipeline that models plain medical text as useful semi-structured documents. We improve on existing models and achieve a F1-score of 0.912 and scale our methods to the entire VA corpus.

Rush III, Everett↗
DOE OSTI2020

Extending the Publish/Subscribe Abstraction for High-Performance I/O and Data Management at Extreme Scale

The Adaptable I/O System (ADIOS) represents the culmination of substantial investment in Scientific Data Management, and it has demonstrated success for several important extreme-scale science cases. However, looking towards the exascale and beyond, we see the development of yet more stringent data management requirements that require new abstractions. Therefore, there is an opportunity to attempt to connect the traditional realms of HPC I/O optimization with the Database / Data Management community. As such, in this paper we offer some specific examples from our ongoing work in managing data structures, services, and performance at the extreme scale for scientific computing. Using the publish/subscribe model afforded by ADIOS, we demonstrate a set of services that connect data format, metadata, queries, data reduction, and high-performance delivery. The resulting publish/subscribe framework facilitates connection to on-line workflow systems to enable the dynamic capabilities that will be required for exascale science.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗
DOE OSTI2019

Can I/O Variability Be Reduced on QoS-Less HPC Storage Systems?

For a production high-performance computing (HPC) system, where storage devices are shared between multiple applications and managed in a best effort manner, I/O contention is often a major problem. In this paper, we propose a balanced messaging-based re-routing in conjunction with throttling at the middleware level. This work tackles two key challenges that have not been fully resolved in the past: whether I/O variability can be reduced on a QoS-less HPC storage system, and how to design a runtime scheduling system that can scale up to a large amount of cores. The proposed scheme uses a two-level messaging system to re-route I/O requests to a less congested storage location so that write performance is improved, while limiting the impact on read by throttling re-routing. An analytical model is derived to guide the setup of optimal throttling factor. We thoroughly analyze the virtual messaging layer overhead and explore whether the in-transit buffering is effective in managing I/O variability. Contrary to the intuition, in-transit buffer cannot completely solve the problem. It can reduce the absolute variability but not the relative variability. The proposed scheme is verified against a synthetic benchmark as well as being used by production applications.

Huang, Dan↗
DOE OSTI2018

A View from ORNL: Scientific Data Research Opportunities in the Big Data Age

One of the core issues across computer and computational science today is adapting to, managing, and learning from the influx of "Big Data". In the commercial space, this problem has led to a huge investment in new technologies and capabilities that are well adapted to dealing with the sorts of human-generated logs, videos, texts, and other large-data artifacts that are processed and resulted in an explosion of useful platforms and languages (Hadoop, Spark, Pandas, etc.). However, translating this work from the enterprise space to the computational science and HPC community has proven somewhat difficult, in part because of some of the fundamental differences in type and scale of data and timescales surrounding its generation and use. We describe a forward-looking research and development plan which centers around the concept of making Input/Output (I/O) intelligent for users in the scientific community, whether they are accessing scalable storage or performing in situ workflow tasks. Much of our work is based on our experience with the Adaptable I/O System (ADIOS 1.X), and our next generation version of the software ADIOS 2.X [1].

Klasky, Scott↗
DOE OSTI2017

Analysis and Modeling of the End-to-End I/O Performance on OLCF's Titan Supercomputer

With the increase of scale and complexity seen in a variety of leadership-class scientific computation and simulation applications, it has become more important to understand their I/O performance characteristics. The user-observed performance is a combination of properties of how the application is using the HPC facility, as well as how others' use of the facility causes variability in the static machine capabilities. Our work leverages statistical analysis of I/O performance data gathered with fine time resolution over a full week from Titan supercomputer. Based on observed properties of the distribution of I/O latencies, we build a three-state hidden Markov model (HMM) to characterize the end-to-end I/O performance on Titan. We parameterize our model using part of the field-gathered I/O performance data and validate it against the rest. The validation results demonstrate that our model can capture the dynamics of end-to-end I/O performance on Titan accurately.

Wan, Lipeng↗
DOE OSTI2017

An evaluation of the state of time synchronization on leadership class supercomputers

We present a detailed examination of time agreement characteristics for nodes within extreme-scale parallel computers. Using a software tool we introduce in this paper, we quantify attributes of clock skew among nodes in three representative high-performance computers sited at three national laboratories. Our measurements detail the statistical properties of time agreement among nodes and how time agreement drifts over typical application execution durations. We discuss the implications of our measurements, why the current state of the field is inadequate, and propose strategies to address observed shortcomings.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

Extending Skel to Support the Development and Optimization of Next Generation I/O Systems

As the memory and storage hierarchy get deeper and more complex, it is important to have new benchmarks and evaluation tools that allow us to explore the emerging middleware solutions to use this hierarchy. Skel is a tool aimed at automating and refining this process of studying HPC I/O performance. It works by generating application I/O kernel/benchmarks as determined by a domain-specific model. This paper provides some techniques for extending Skel to address new situations and to answer new research questions. For example, we document use cases as diverse as using Skel to troubleshoot I/O performance issues for remote users, refining an I/O system model, and facilitating the development and testing of a mechanism for runtime monitoring and performance analytics. We also discuss data oriented extensions to Skel to support the study of compression techniques for Exascale scientific data management.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

TGE: Machine Learning Based Task Graph Embedding for Large-Scale Topology Mapping

Task mapping is an important problem in parallel and distributed computing. The goal in task mapping is to find an optimal layout of the processes of an application (or a task) onto a given network topology. We target this problem in the context of staging applications. A staging application consists of two or more parallel applications (also referred to as staging tasks) which run concurrently and exchange data over the course of computation. Task mapping becomes a more challenging problem in staging applications, because not only data is exchanged between the staging tasks, but also the processes of a staging task may exchange data with each other. We propose a novel method, called Task Graph Embedding (TGE), that harnesses the observable graph structures of parallel applications and network topologies. TGE employs a machine learning based algorithm to find the best representation of a graph, called an embedding, onto a space in which the task-to-processor mapping problem can be solved. We evaluate and demonstrate the effectiveness of TGE experimentally with the communication patterns extracted from runs of XGC, a large-scale fusion simulation code, on Titan.

Choi, Jong Youl↗
DOE OSTI2017

Variable-Speed Screw Chiller, Sidney Yates Building, Washington, DC

This report captures the findings from an evaluation ORNL performed on a new chiller technology as part of GSA's Proving Ground Program. Note: Appendices B&C were removed from this report while the author looks for a way to insert them without consuming over 200MB of file size.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2017

Exacution: Enhancing Scientific Data Management for Exascale

As we continue toward exascale, scientific data volume is continuing to scale and becoming more burdensome to manage. In this paper, we lay out opportunities to enhance state of the art data management techniques. We emphasize well-principled data compression, and using it to achieve progressive refinement. This can both accelerate I/O and afford the user increased flexibility when she interacts with the data. The formulation naturally maps onto enabling partitioning of the progressively improving-quality representations of a data quantity into different media-type destinations, to keep the highest priority information as close as possible to the computation, and take advantage of deepening memory/storage hierarchies in ways not previously possible. Careful monitoring is requisite to our vision, not only to verify that compression has not eliminated salient features in the data, but also to better understand the performance of massively parallel scientific applications. Increased mathematical rigor would be ideal,to help bring compression on a better-understood theoretical footing, closer to the relevant scientific theory, more aware of constraints imposed by the science, and more tightly error-controlled. Throughout, we highlight pathfinding research we have begun exploring related these topics, and comment toward future work that will be needed.

97 MATHEMATICS AND COMPUTING↗