Figure 1: An example of particle data from the MC 3 dark matter simulation. The images show the comparison between full resolution data and statistically-based level-of-detail data samples generated via in-situ sampling. AbstractWe describe a simulation-time random sampling of a large-scale particle simulation, the RoadRunner Universe MC 3 cosmological simulation, for interactive post-analysis and visualization. Simulation data generation rates will continue to be far greater than storage bandwidth rates by many orders of magnitude. This implies that only a very small fraction of data generated by a simulation can ever be stored and subsequently post-analyzed. The limiting factors in this situation are similar to the problem in many population surveys: there aren't enough human resources to query a large population. To cope with the lack of resources, statistical sampling techniques are used to create a representative data set of a large population. Following this analogy, we propose to store a simulationtime random sampling of the particle data for post-analysis, with level-of-detail organization, to cope with the bottlenecks. A sample is stored directly from the simulation in a level-of-detail format for post-visualization and analysis, which amortizes the cost of post-processing and reduces workflow time. Additionally by sampling during the simulation, we are able to analyze the entire particle population to record full population statistics and quantify sample error.
Repeated measures degradation studies are used to assess product or component reliability when there are few or even no failures expected during a study. Such studies are used to assess the shelf life of materials and products. We show how to evaluate the properties of proposed test plans needed to identify statistically efficient tests. We consider test plans for applications where parameters related to the degradation distribution or the lifetime distribution are to be estimated. We use the approximate largesample variance-covariance matrix of the parameters of a mixed effects linear regression model for repeated measures degradation data to assess the effect of sample size (number of units and number of measurements within the units) on estimation precision of both degradation and failure-time distribution quantiles. We also illustrate the complementary use of simulation-based methods for evaluating and comparing test plans. These test-planning methods are illustrated with examples.
Power is becoming an increasingly important concern for large supercomputer centers. However, to date, there have been a dearth of studies of power usage 'in the wild'-on production supercomputers running production workloads. In this paper, we present the initial results of a project to characterize the power usage of the three Top500 supercomputers at Los Alamos National Laboratory: Cielo, Roadrunner, and Luna (#15, #19, and #47, respectively, on the June 2012 Top500 list). Power measurements taken both at the switchboard level and within the compute racks are presented and discussed. Some noteworthy results of this study are that (1) variability in power consumption differs across architectures, even when running a similar workload and (2) Los Alamos National Laboratory's scientific workload draws, on average, only 70-75% of LINPACK power and only 40-55% of nameplate power, implying that power capping may enable a substantial reduction in power and cooling infrastructure while impacting comparatively few applications. 275 system's power usage and examine the discrepancies between measuring power at the switchboard level and at the sub-rack level.Our findings are that (1) power variability differs substantially across architectures; (2) from a power perspective, real scientific workloads bear little in common with the LINPACK benchmark (not too surprisingly); (3) the difference between worst-case and average-case power draws indicates that supercomputing data centers may contain a fair amount of 'trapped capacity' in their power systems, more if power capping can be implemented on a full-system basis; (4) job schedulers theoretically have the potential to increase trapped capacity even further by pairing jobs of different power envelopes; and (4) energy savings are unlikely to be achieved merely by frequency and voltage scaling, given how supercomputers are currently run.We anticipate that this paper will assist future power studies that require knowledge of real-world supercomputer power data to drive their approach and solutions.The remainder of the paper is organized as follows. We discuss the most relevant pieces of related work in Section 2. Section 3 describes LANL's data center in terms of its power characteristics and the main supercomputers it hosts. The main section of the paper is Section 4, where we present our power measurements and associated analyses. Section 5 briefly describes some prospects for follow-on research. Finally, we draw some conclusions from our findings in Section 6. RELATED WORKFan, Weber, and Barroso quantify the power usage of three workloads that run at one of Google's large data centers: Google Web search, GMail, and various offline MapReduce jobs [4]. As in our work, they examine power characteristics at the rack, sub-cluster, and full-cluster levels. The key characteristic that distinguishes our work from theirs is that we focus on a production scientific workload in which applications tend to be more tightly coupled than search and email services and MapReduce jobs. Scienti...
As computer simulations continue to grow in size and complexity, they present a particularly challenging class of big data problems. at this scale can generate output that exceeds both the storage capacity and the bandwidth available for transfer to storage, making post-processing and analysis challenging. One approach is to embed some analyses in the simulation while the simulation is running -a strategy often called in situ analysis -to reduce the need for transfer to storage. Another strategy is to save only a reduced set of time steps rather than the full simulation. Typically the selected time steps are evenly spaced, where the spacing can be defined by the budget for storage and transfer. This paper combines both of these ideas to introduce an online in situ method for identifying a reduced set of time steps of the simulation to save. Our approach significantly reduces the data transfer and storage requirements, and it provides improved fidelity to the simulation to facilitate post-processing and reconstruction. We illustrate the method using a computer simulation that supported NASA's 2009 Lunar Crater Observation and Sensing Satellite mission.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.