Search NASA⌕ Search

SEARCH · Search NASA

Results for “Workload”

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.

At least 1,477 records · Page 82

Profiling the BLAST bioinformatics application for load balancing on high-performance computing clusters

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.

59 BASIC BIOLOGICAL SCIENCES↗

SMC 2021 : Analyzing Resource Utilization and User Behavior on Titan Supercomputer

Resource utilization statistics of submitted jobs on a supercomputer can help us understand how users from various scientific domains use HPC platforms and better design a job scheduler. We explore to generate insight regarding workload distribution and usage pattern domains from job scheduler trace, GPU failure information, and project-specific information collected from Titan supercomputer. Furthermore, we want to know how the scheduler performance varies over time and how the users' scheduling behavior changes following a system failure. These observations have the potential to provide valuable insight, which is helpful to prepare for system failures. These practices will help us develop and apply novel machine learning algorithms in understanding system behavior, requirement, and better scheduling of HPC systems. There are two datasets, RUR and GPU. RUR: This dataset is the job scheduler traces collected from the Titan supercomputerfrom 01/01/2015 to 07/31/2019 (2015.csv - 2019.csv). These were collected usingResource Utilization Report (RUR), a Cray-developed resource-usage data collectionand reporting system. It contains the usage information of its critical resources (CPU,Memory, GPU, and I/O) of each running job on Titan during that period [2]. ProjectAreas: Every job is associated with a project ID. TheProjectAreas.csvdatasetprovides a mapping of the project ID to its domain science. GPU: There have been some hardware-related issues in the GPUs in Titan that caused some GPUs to fail, sometimes irrecoverably during some job runs. This dataset provides information regarding these failures during the execution of the submitted jobs. GPUs on Titan are uniquely identified by a serial number (SN), and they are installed in a location. A GPU can be installed in a location, then removed from that location following a failure, and then re-installed in a different location after fixing the problem. If the failure can't be recovered, the GPU might be removed entirely from Titan. There are two prominent types of failures that resulted in the removal of GPUs from Titan: Double Bit Error (DBE) and Out of the Bus (OTB). The dataset (gc_full.csv) has the following fields: 1. SN : Serial number of a GPU 2. location : The location where it is installed 3. insert : The time when it was inserted into that location 4. remove : The time when it was removed from that location 5. duration : Amount of time the GPU spent in this location 6. out : If the device was taken out entirely w/o a re-installment into a new location. 7. event : If the GPU was taken out entirely, the reason for its removal. To learn more about this dataset, please refer to the git repositoryhttps://github.com/olcf/TitanGPULifeand the related publication [1]. References [1] George Ostrouchov, Don Maxwell, Rizwan A Ashraf, Christian Engelmann, MallikarjunShankar, and James H Rogers. Gpu lifetimes on titan supercomputer: Survival analysisand reliability. InSC20: International Conference for High Performance Computing,Networking, Storage and Analysis, pages 1-14. IEEE, 2020. [2] Feiyi Wang, Sarp Oral, Satyabrata Sen, and Neena Imam. Learning from five-yearresource-utilization data of titan system. In2019 IEEE International Conference onCluster Computing (CLUSTER), pages 1-6. IEEE, 2019.

42 ENGINEERING↗

SMC 2021 Data Challenge: Analyzing Resource Utilization and User Behavior on Titan Supercomputer

Resource utilization statistics of submitted jobs on a supercomputer can help us understand how users from various scientific domains use HPC platforms and better design a job scheduler. We explore to generate insight regarding workload distribution and usage pattern domains from job scheduler trace, GPU failure information, and project-specific information collected from Titan supercomputer. Furthermore, we want to know how the scheduler performance varies over time and how the users' scheduling behavior changes following a system failure. These observations have the potential to provide valuable insight, which is helpful to prepare for system failures. These practices will help us develop and apply novel machine learning algorithms in understanding system behavior, requirement, and better scheduling of HPC systems. There are two datasets, RUR and GPU: RUR dataset is the job scheduler traces collected from the Titan supercomputer from 01/01/2015 to 07/31/2019 (2015.csv - 2019.csv). These were collected using resource Utilization Report (RUR), a Cray-developed resource-usage data collection and reporting system. It contains the usage information of its critical resources (CPU, Memory, GPU, and I/O) of each running job on Titan during that period (https://ieeexplore.ieee.org/abstract/document/8891001). It includes ProjectAreas as additional information, every job is associated with a project ID. TheProjectAreas.csv dataset provides a mapping of the project ID to its domain science. GPU dataset has information regarding GPU failure on Titan. There have been some hardware-related issues in the GPUs in Titan that caused some GPUs to fail, sometimes irrecoverably during some job runs. This dataset provides information regarding these failures during the execution of the submitted jobs. GPUs on Titan are uniquely identified by a serial number (SN), and they are installed in a location. A GPU can be installed in a location, then removed from that location following a failure, and then re-installed in a different location after fixing the problem. If the failure can't be recovered, the GPU might be removed entirely from Titan. There are two prominent types of failures that resulted in the removal of GPUs from Titan: Double Bit Error (DBE) and Out of the Bus (OTB). The dataset (gc_full.csv) has seven attributes, we provided a short description of these attributes in the ReadMe file. To learn more about this dataset, please refer to the git repository https://github.com/olcf/TitanGPULife and the related publication (https://ieeexplore.ieee.org/abstract/document/9355319).

42 ENGINEERING↗

Content-aware foveated camera for multi-target tracking

Modern image sensors deliver substantial space-time bandwidth, yet indiscriminate acquisition often overwhelms memory, computation, and downstream perception. We present a content-aware, multi-foveated camera that dynamically reallocates sensing and magnification to multiple regions of interest (ROIs). A phase-only spatial light modulator (SLM) serves as a solid-state, inertia-free beam-steering and lens element, enabling per-frame field-of-view (FOV) reconfiguration and content-aware target tracking. By interleaving frames across foveae, our system preserves a wide-FOV situational context while refreshing each ROI at high rates, thereby reducing data volume without degrading task performance. We constructed a prototype employing a single-SLM, single-sensor architecture and demonstrated its application in real-time multi-object tracking with dynamic ROI maintenance across multiple viewpoints. The approach offers a general pathway to integrate detection, tracking, and segmentation algorithms in the acquisition loop, shifting workload from post hoc processing to intelligent capture.

Zang, Zihan [Univ. of California, Los Angeles, CA ↗

Traversing large graphs on GPUs with unified memory

Due to the limited capacity of GPU memory, the majority of prior work on graph applications on GPUs has been restricted to graphs of modest sizes that fit in memory. Recent hardware and software advances make it possible to address much larger host memory transparently as a part of a feature known as unified virtual memory. While accessing host memory over an interconnect is understandably slower, the problem space has not been sufficiently explored in the context of a challenging workload with low computational intensity and an irregular data access pattern such as graph traversal. We analyse the performance of breadth first search (BFS) for several large graphs in the context of unified memory and identify the key factors that contribute to slowdowns. Next, we propose a lightweight offline graph reordering algorithm, HALO (Harmonic Locality Ordering), that can be used as a pre-processing step for static graphs. HALO yields speedups of 1.5x-1.9x over baseline in subsequent traversals. Our method specifically aims to cover large directed real world graphs in addition to undirected graphs whereas prior methods only account for the latter. Additionally, we demonstrate ties between the locality ordering problem and graph compression and show that prior methods from graph compression such as recursive graph bisection can be suitably adapted to this problem.

Gera, Prasun↗

An Autonomous System for Grouping Events in a Developing Aftershock Sequence

We describe a prototype detection framework that automatically clusters events in real time from a rapidly unfolding aftershock sequence. We use the fact that many aftershocks are repetitive, producing similar waveforms. By clustering events based on correlation measures of waveform similarity, the number of independent event instances that must be examined in detail by analysts may be reduced. Our system processes array data and acquires waveform templates with a short-term average (STA)/long-term average (LTA) detector operating on a beam directed at the P phases of the aftershock sequence. The templates are used to create correlation-type (subspace) detectors that sweep the subsequent data stream for occurrences of the same waveform pattern. Events are clustered by association with a particular detector. Hundreds of subspace detectors can run in this framework a hundred times faster than in real time. Nonetheless, to check the growth in the number of detectors, the framework pauses periodically and reclusters detections to reduce the number of event groups. These groups define new subspace detectors that replace the older generation of detectors. Because low-magnitude occurrences of a particular signal template may be missed by the STA/LTA detector, we advocate restarting the framework from the beginning of the sequence periodically to reprocess the entire data stream with the existing detectors. We tested the framework on 10 days of data from the Nevada Seismic Array (NVAR) covering the 2003 San Simeon earthquake. One hundred eighty-four automatically generated detectors produced 676 detections resulting in a potential reduction in analyst workload of up to 73%.

58 GEOSCIENCES↗

The Oklahoma Geological Survey Statewide Seismic Network

The Oklahoma Geological Survey (OGS) monitors seismicity throughout the state of Oklahoma utilizing permanent and temporary seismometers installed by OGS and other agencies, while producing a real–time earthquake catalog. The OGS seismic network was recently added to the Advanced National Seismic System (ANSS) as a self–supporting regional seismic network, and earthquake locations and magnitudes are automatically reported through U.S. Geological Survey and are part of the ANSS Comprehensive Earthquake Catalog. In Oklahoma, before 2009, background seismicity rates were about 2 M 3.0+ earthquakes per year, which increased to 579 and 903 M 3.0+ earthquakes in 2014 and 2015, respectively. After seismicity peaked, the rate fell to 624, 304, and 194 M 3.0+ earthquakes in 2016, 2017, and 2018, respectively. The catalog is complete down to M 2.2 from mid–2014 to present, despite the significant workload for a primarily state–funded regional network. That astonishing uptick in seismicity has been largely attributed to wastewater injection practices. The OGS provides the Oklahoma Corporation Commission, the agency responsible for regulating oil and gas activities within the state, with technical guidance and earthquake products that inform their “traffic–light” mitigation protocol and other mitigating actions. We have initiated a citizen–scientist–driven, educational seismometer program by installing Raspberry Shake geophones throughout the state at local schools, museums, libraries, and state parks. The seismic hazard of the state portends a continued need for expansion and densification of seismic monitoring throughout Oklahoma.

58 GEOSCIENCES↗

Rapid Simulation of Unconventional Reservoirs by Multidomain Multiresolution Modeling Based on the Diffusive Time of Flight

Summary The fast marching method (FMM)-based rapid flow simulation has been shown to accelerate simulation efficiency by orders of magnitude by transforming 3D simulation to equivalent 1D simulation using the concept of the “diffusive time of flight” (DTOF). However, the 1D transformation does not directly apply to multiwell problems. In this paper, we propose a novel DTOF-based multidomain multiresolution discretization scheme to accelerate multiwell simulation of unconventional reservoirs. Our method formulates multiwell simulation problems based on the DTOF which displays the pressure front propagation in unconventional reservoirs. The DTOF contours are used to partition the reservoir into local and shared domains. A local domain is where the flow is dominated by a single well, and the shared domain is where the fluid flow is influenced by multiple wells. The DTOF contours expand independently in local domains and interfere in the shared domain. After the partitioning, each domain is discretized using a multiresolution scheme whereby the original 3D fine mesh is preserved near the wells to account for detailed physics including gravity, and the rest of the domain is discretized into 1D mesh based on the DTOF contours to alleviate the simulation workload. The power and efficacy of our approach are demonstrated using synthetic and field-scale simulation models with different degrees of geologic and well-completion complexity. The simulation results, number of active cells, and computation time for the proposed discretization scheme are compared with the original high-fidelity 3D model for each case. The results show that the proposed method is suitable for multiwell simulation problems in unconventional reservoirs and can accelerate flow simulations by orders of magnitude with minimal loss of accuracy. The novelty of this work is the creation of DTOF-derived multiresolution discretization with local and shared domains to simplify and accelerate the calculation of subsurface flow problems, especially in unconventional reservoirs. Our workflow can be easily interfaced with commercial simulators, making it suitable for large-scale field applications.

Engineering↗

Scientific data analysis on data-parallel platforms.

As scientific computing users migrate to petaflop platforms that promise to generate multi-terabyte datasets, there is a growing need in the community to be able to embed sophisticated analysis algorithms in the computing platforms' storage systems. Data Warehouse Appliances (DWAs) are attractive for this work, due to their ability to store and process massive datasets efficiently. While DWAs have been utilized effectively in data-mining and informatics applications, they remain largely unproven in scientific workloads. In this paper we present our experiences in adapting two mesh analysis algorithms to function on five different DWA architectures: two Netezza database appliances, an XtremeData dbX database, a LexisNexis DAS, and multiple Hadoop MapReduce clusters. The main contribution of this work is insight into the differences between these DWAs from a user's perspective. In addition, we present performance measurements for ten DWA systems to help understand the impact of different architectural trade-offs in these systems.

99 GENERAL AND MISCELLANEOUS↗

Institutional impediments to using alternative water sources in thermoelectric power plants.

This report was funded by the U.S. Department of Energy's (DOE's) National Energy Technology Laboratory (NETL) Existing Plants Research Program, which has an energy-water research effort that focuses on water use at power plants. This study complements the Existing Plants Research Program's overall research effort by evaluating water issues that could impact power plants. Obtaining adequate water supplies for cooling and other operations at a reasonable cost is a key factor in siting new and maintaining existing thermoelectric power plant operations. One way to reduce freshwater consumption is to use alternative water sources such as reclaimed (or recycled) water, mine pool water, and other nontraditional sources. The use of these alternative sources can pose institutional challenges that can cause schedule delays, increase costs, or even require plants to abandon their plans to use alternative sources. This report identifies and describes a variety of institutional challenges experienced by power plant owners and operators across the country, and for many of these challenges it identifies potential mitigating approaches. The information comes from publically available sources and from conversations with power plant owners/operators familiar with using alternative sources. Institutional challenges identified in this investigation include, but are not limited to, the following: (1) Institutional actions and decisions that are beyond the control of the power plant. Such actions can include changes in local administrative policies that can affect the use of reclaimed water, inaccurate growth projections regarding the amount of water that will be available when needed, and agency workloads and other priorities that can cause delays in the permitting and approval processes. (2) Developing, cultivating, and maintaining institutional relationships with the purveyor(s) of the alternative water source, typically a municipal wastewater treatment plant (WWTP), and with the local political organizations that can influence decisions regarding the use of the alternative source. Often a plan to use reclaimed water will work only if local politics and power plant goals converge. Even then, lengthy negotiations are often needed for the plans to come to fruition. (3) Regulatory requirements for planning and developing associated infrastructure such as pipelines, storage facilities, and back-up supplies that can require numerous approvals, permits, and public participation, all of which can create delays and increased costs. (4) Permitting requirements that may be difficult to meet, such as load-based discharge limits for wastewater or air emissions limitations for particulate matter (which will be in the mist of cooling towers that use reclaimed water high in dissolved solids). (5) Finding discharge options for cooling tower blowdown of reclaimed water that are acceptable to permitting authorities. Constituents in this wastewater can limit options for discharge. For example, discharge to rivers requires National Pollutant Discharge Elimination System (NPDES) permits whose limits may be difficult to meet, and underground injection can be limited because many potential injection sites have already been claimed for disposal of produced waters from oil and gas wells or waters associated with gas shale extraction. (6) Potential liabilities associated with using alternative sources. A power plant can be liable for damages associated with leaks from reclaimed water conveyance systems or storage areas, or with mine water that has been contaminated by unscrupulous drillers that is subsequently discharged by the power plant. (7) Community concerns that include, but are not limited to, increased saltwater drift on farmers fields; the possibility that the reclaimed water will contaminate local drinking water aquifers; determining the 'best' use of WWTP effluent; and potential health concerns associated with emissions from the cooling towers that use recycled water. (8) Interveners that raise public concerns about the potential for emissions of emerging pollutants of concern to cause health or environmental problems. Mitigating solutions range from proactive communications with the local communities (which can be implemented by the power plants) to technical solutions, such as developing means to reduce the concentrations of total dissolved solids (TDS) and other contaminants in cooling water to maintain plant efficiency and while meeting discharge limits. These kinds of solutions may be appropriate for DOE research and development (R&D) funding.

03 NATURAL GAS↗

Keeping checkpoint/restart viable for exascale systems.

Next-generation exascale systems, those capable of performing a quintillion (10{sup 18}) operations per second, are expected to be delivered in the next 8-10 years. These systems, which will be 1,000 times faster than current systems, will be of unprecedented scale. As these systems continue to grow in size, faults will become increasingly common, even over the course of small calculations. Therefore, issues such as fault tolerance and reliability will limit application scalability. Current techniques to ensure progress across faults like checkpoint/restart, the dominant fault tolerance mechanism for the last 25 years, are increasingly problematic at the scales of future systems due to their excessive overheads. In this work, we evaluate a number of techniques to decrease the overhead of checkpoint/restart and keep this method viable for future exascale systems. More specifically, this work evaluates state-machine replication to dramatically increase the checkpoint interval (the time between successive checkpoint) and hash-based, probabilistic incremental checkpointing using graphics processing units to decrease the checkpoint commit time (the time to save one checkpoint). Using a combination of empirical analysis, modeling, and simulation, we study the costs and benefits of these approaches on a wide range of parameters. These results, which cover of number of high-performance computing capability workloads, different failure distributions, hardware mean time to failures, and I/O bandwidths, show the potential benefits of these techniques for meeting the reliability demands of future exascale platforms.

99 GENERAL AND MISCELLANEOUS↗

Using the Sirocco File System for high-bandwidth checkpoints.

The Sirocco File System, a file system for exascale under active development, is designed to allow the storage software to maximize quality of service through increased flexibility and local decision-making. By allowing the storage system to manage a range of storage targets that have varying speeds and capacities, the system can increase the speed and surety of storage to the application. We instrument CTH to use a group of RAM-based Sirocco storage servers allocated within the job as a high-performance storage tier to accept checkpoints, allowing computation to potentially continue asynchronously of checkpoint migration to slower, more permanent storage. The result is a 10-60x speedup in constructing and moving checkpoint data from the compute nodes. This demonstration of early Sirocco functionality shows a significant benefit for a real I/O workload, checkpointing, in a real application, CTH. By running Sirocco storage servers within a job as RAM-only stores, CTH was able to store checkpoints 10-60x faster than storing to PanFS, allowing the job to continue computing sooner. While this prototype did not include automatic data migration, the checkpoint was available to be pushed or pulled to disk-based storage as needed after the compute nodes continued computing. Future developments include the ability to dynamically spawn Sirocco nodes to absorb checkpoints, expanding this mechanism to other fast tiers of storage like flash memory, and sharing of dynamic Sirocco nodes between multiple jobs as needed.

99 GENERAL AND MISCELLANEOUS↗

Recovery Act: A Measurement -- Management Technology for Improving Energy Efficiency in Data Centers and Telecommunication Facilities (Final Technical Report)

Data center (DC) electricity use is increasing at an annual rate of over 20% and presents a concern for the Information Technology (IT) industry, governments, and the society. A large fraction of the energy use is consumed by the compressor cooling to maintain the recommended operating conditions for IT equipment. The most common way to improve the DC efficiency is achieved by optimally provisioning the cooling power to match the global heat dissipation in the DC. However, at a more granular level, the large range of heat densities of today's IT equipment makes the task of provisioning cooling power optimized to the level of individual computer room air conditioning (CRAC) units much more challenging. Distributed sensing within a DC enables the development of new strategies to improve energy efficiency, such as hot spot elimination through targeted cooling, matching power consumption at rack level with workload schedule, and minimizing power losses. The scope of Measurement and Management Technologies (MMT) is to develop a software tool and the underlying sensing technology to provide critical decision support and control for DC and telecommunication facilities (TF) operations. A key aspect of MMT technology is integration of modeling tools to understand how changes in one operational parameter affect the overall DC response. It is demonstrated that reduced ordered models for DC can generate, in less than 2 seconds computational time, a three dimensional thermal model in a 50 kft 2 DC. This rapid modeling enables real time visualization of the DC conditions and enables "0what if" scenarios simulations to characterize response to "disturbances". One such example is thermal zone modeling that matches the cooling power to the heat generated at a local level by identifying DC zones cooled by a specific CRAC. Turning off a CRAC unit can be simulated to understand how the other CRAC utilization changes and how server temperature responds. Several new sensing technologies were added to the existing MMT platform: (1) air contamination (corrosion) sensors, (2) power monitoring, and (3) a wireless environmental sensing network. All three technologies are built on cost effective sensing solutions that increase the density of sensing points and enable high resolution mapping of DCs. The wireless sensing solution enables Air Conditioning Unit (ACU) control while the corrosion sensor enables air side economization and can quantify the risk of IT equipment failure due to air contamination. Validation data for six test sites demonstrate that leveraging MMT energy efficiency solutions combined with industry best practices results in an average of 20% reduction in cooling energy, without major infrastructure upgrades. As an illustration of the unique MMT capabilities, a data center infrastructure efficiency (DCIE) of 87% (industry best operation) was achieved. The technology is commercialized through IBM System and Technology Lab Services that offers MMT as a solution to improve DC energy efficiency. Estimation indicates that deploying MMT in existing DCs can results in an 8 billion kWh savings and projection indicates that constant adoption of MMT can results in obtainable savings of 44 billion kWh in 2035. Negotiations are under way with business partners to commercialize/license the ACU control technology and the new sensor solutions (corrosion and power sensing) to enable third party vendors and developers to leverage the energy efficiency solutions.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Design of FastQuery: How to Generalize Indexing and Querying System for Scientific Data

Modern scientific datasets present numerous data management and analysis challenges. State-of-the-art index and query technologies such as FastBit are critical for facilitating interactive exploration of large datasets. These technologies rely on adding auxiliary information to existing datasets to accelerate query processing. To use these indices, we need to match the relational data model used by the indexing systems with the array data model used by most scientific data, and to provide an efficient input and output layer for reading and writing the indices. In this work, we present a flexible design that can be easily applied to most scientific data formats. We demonstrate this flexibility by applying it to two of the most commonly used scientific data formats, HDF5 and NetCDF. We present two case studies using simulation data from the particle accelerator and climate simulation communities. To demonstrate the effectiveness of the new design, we also present a detailed performance study using both synthetic and real scientific workloads.

97 MATHEMATICS AND COMPUTING↗

Performance analysis of Darshan 2.2.3 on the Cray XE6 platform

Darshan is a production-quality I/O characteriza tion tool that captures and summarizes the I/O behavior of parallel applications. It records a variety of information with minimal overhead, including access patterns, number of files accessed, and the amount of time consumed by I/O routines. Darshan’s lightweight design makes it suitable for full-time deployment for workload characterization of large HPC systems.

Carns, Philip↗

How to Plug In a Supercomputer: Quantifying Power Consumption at Scale [Slides]

Power is becoming an increasingly important concern for large super computercenters. However,todate,therehavebeenadearthofstudiesof power usage “in the wild”—on production supercomputers running pro duction workloads. In this talk, we present the initial results of an effort to characterize the power usage of three Top100 supercomputers at Los Alamos National Laboratory: Cielo, Roadrunner, and Luna (#18, #22, and #64, respectively, on the November 2012 Top500 list). Power measure ments taken both at the switchboard level and within the compute racks are presented and discussed.

97 MATHEMATICS AND COMPUTING↗

Canopy 2.1 User Guide

Canopy is a visual analytic software suite, developed by researchers at the Pacific Northwest National Laboratory (PNNL), to support intelligence analysts’ ability to efficiently analyze massive amounts of multimedia data. Canopy incorporates state-of-the-art computer vision algorithms and interactive visualizations so that analysts can comprehend and articulate the big picture, thereby reducing their workload and ultimately the effort of identifying the critical intelligence for decision makers.

Canopy↗

Toward Highly Secure and Autonomic Computing Systems: A Hierarchical Approach

The overall objective of this research project is to develop novel architectural techniques as well as system software to achieve a highly secure and intrusion-tolerant computing system. Such system will be autonomous, self-adapting, introspective, with self-healing capability under the circumstances of improper operations, abnormal workloads, and malicious attacks. The scope of this research includes: (1) System-wide, unified introspection techniques for autonomic systems, (2) Secure information-flow microarchitecture, (3) Memory-centric security architecture, (4) Authentication control and its implication to security, (5) Digital right management, (5) Microarchitectural denial-of-service attacks on shared resources. During the period of the project, we developed several architectural techniques and system software for achieving a robust, secure, and reliable computing system toward our goal.

97 MATHEMATICS AND COMPUTING↗