skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: DAiSES: Dynamic Adaptivity in Support of Extreme Scale Department of Energy Project No. ER25622 Prime Contract No. DE-FG02-04ER25622 Final Report for September 15, 2004-September 14, 2008

Abstract

The DAiSES project [Te04] was focused on enabling conventional operating systems, in particular, those running on extreme scale systems, to dynamically customize system resource management in order to offer applications the best possible environment in which to execute. Such dynamic adaptation allows operating systems to modify the execution environment in response to changes in workload behavior and system state. The main challenges of this project included determination of what operating system (OS) algorithms, policies, and parameters should be adapted, when to adapt them, and how to adapt them. We addressed these challenges by using a combination of static analysis and runtime monitoring and adaptation to identify a priori profitable targets of adaptation and effective heuristics that can be used to dynamically trigger adaptation. Dynamic monitoring and adaptation of the OS was provided by either kernel modifications or the use of KernInst and Kperfmon [Wm04]. Since Linux, an open source OS, was our target OS, patches submitted by kernel developers and researchers often facilitated kernel modifications. KernInst operates on unmodified commodity operating systems, i.e., Solaris and Linux; it is fine-grained, thus, there were few constraints on how the underlying OS can be modified. Dynamically adaptive functionality of operating systems, both inmore » terms of policies and parameters, is intended to deliver the maximum attainable performance of a computational environment and meet, as best as possible, the needs of high-performance applications running on extreme scale systems, while meeting system constraints. DAiSES research endeavored to reach this goal by developing methodologies for dynamic adaptation of OS parameters and policies to manage stateful and stateless resources [Te06] and pursuing the following two objectives: (1) Development of mechanisms to dynamically sense, analyze, and adjust common performance metrics, fluctuating workload situations, and overall system environment conditions. (2) Demonstration, via Linux prototypes and experiments, of dynamic self-tuning and self-provisioning in HPC environments. From a high level, the DAiSES methodology, depicted in Figure 1, includes characterization of application resource usage patterns, identification of candidate (profitable) adaptation targets, determination of feasible adaptation ranges, definition of heuristics to trigger adaptation, design and implementation of OS monitoring, triggering, and adaptation code, and quantification of performance gains.« less

Authors:
Publication Date:
Research Org.:
The University of Texas at El Paso
Sponsoring Org.:
USDOE Office of Science (SC)
OSTI Identifier:
951983
Report Number(s):
DOE/ER/25622-1
TRN: US201002%%341
DOE Contract Number:
FG02-04ER25622
Resource Type:
Technical Report
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING; ALGORITHMS; DESIGN; IMPLEMENTATION; KERNELS; METRICS; MODIFICATIONS; MONITORING; PERFORMANCE; RESOURCE MANAGEMENT; TARGETS

Citation Formats

PI: Patricia J. Teller, Ph.D. University of Texas-El Paso Department of Computer Science. DAiSES: Dynamic Adaptivity in Support of Extreme Scale Department of Energy Project No. ER25622 Prime Contract No. DE-FG02-04ER25622 Final Report for September 15, 2004-September 14, 2008. United States: N. p., 2009. Web. doi:10.2172/951983.
PI: Patricia J. Teller, Ph.D. University of Texas-El Paso Department of Computer Science. DAiSES: Dynamic Adaptivity in Support of Extreme Scale Department of Energy Project No. ER25622 Prime Contract No. DE-FG02-04ER25622 Final Report for September 15, 2004-September 14, 2008. United States. doi:10.2172/951983.
PI: Patricia J. Teller, Ph.D. University of Texas-El Paso Department of Computer Science. Tue . "DAiSES: Dynamic Adaptivity in Support of Extreme Scale Department of Energy Project No. ER25622 Prime Contract No. DE-FG02-04ER25622 Final Report for September 15, 2004-September 14, 2008". United States. doi:10.2172/951983. https://www.osti.gov/servlets/purl/951983.
@article{osti_951983,
title = {DAiSES: Dynamic Adaptivity in Support of Extreme Scale Department of Energy Project No. ER25622 Prime Contract No. DE-FG02-04ER25622 Final Report for September 15, 2004-September 14, 2008},
author = {PI: Patricia J. Teller, Ph.D. University of Texas-El Paso Department of Computer Science},
abstractNote = {The DAiSES project [Te04] was focused on enabling conventional operating systems, in particular, those running on extreme scale systems, to dynamically customize system resource management in order to offer applications the best possible environment in which to execute. Such dynamic adaptation allows operating systems to modify the execution environment in response to changes in workload behavior and system state. The main challenges of this project included determination of what operating system (OS) algorithms, policies, and parameters should be adapted, when to adapt them, and how to adapt them. We addressed these challenges by using a combination of static analysis and runtime monitoring and adaptation to identify a priori profitable targets of adaptation and effective heuristics that can be used to dynamically trigger adaptation. Dynamic monitoring and adaptation of the OS was provided by either kernel modifications or the use of KernInst and Kperfmon [Wm04]. Since Linux, an open source OS, was our target OS, patches submitted by kernel developers and researchers often facilitated kernel modifications. KernInst operates on unmodified commodity operating systems, i.e., Solaris and Linux; it is fine-grained, thus, there were few constraints on how the underlying OS can be modified. Dynamically adaptive functionality of operating systems, both in terms of policies and parameters, is intended to deliver the maximum attainable performance of a computational environment and meet, as best as possible, the needs of high-performance applications running on extreme scale systems, while meeting system constraints. DAiSES research endeavored to reach this goal by developing methodologies for dynamic adaptation of OS parameters and policies to manage stateful and stateless resources [Te06] and pursuing the following two objectives: (1) Development of mechanisms to dynamically sense, analyze, and adjust common performance metrics, fluctuating workload situations, and overall system environment conditions. (2) Demonstration, via Linux prototypes and experiments, of dynamic self-tuning and self-provisioning in HPC environments. From a high level, the DAiSES methodology, depicted in Figure 1, includes characterization of application resource usage patterns, identification of candidate (profitable) adaptation targets, determination of feasible adaptation ranges, definition of heuristics to trigger adaptation, design and implementation of OS monitoring, triggering, and adaptation code, and quantification of performance gains.},
doi = {10.2172/951983},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Tue May 05 00:00:00 EDT 2009},
month = {Tue May 05 00:00:00 EDT 2009}
}

Technical Report:

Save / Share:
  • This report summarizes the activities of the INRA Water Research Consortium (IWRC) for the period beginning September 15, 2005 and ending December 16, 2010. This report compares accomplishments to project objectives, documents the activities associated with this project, and lists products developed during the course of the project. The Water Resources Research Needs Assessment team received funding from the Inland Northwest Research Alliance Water Resources Steering Committee to facilitate a structured needs assessment process that could provide a basis for future targeted research efforts to improve regional water resources management in the Inland Northwest region. The original INRA proposal specificallymore » mentions the need to conduct a detailed assessment of the information and research needs of policy makers and water user groups during a period of increasing competition for scarce water supplies. A particular focus of this assessment would be to understand what types of research might facilitate water resource management during periods of drought. The specific goals of the Needs Assessment project were to: (1) Quickly ascertain the perceptions of diverse stakeholders in this region; (2) Condense this complex information into a format that can be shared with the INRA scientific panel, and (3) Develop of a realistic set of research needs & priorities that can shape future INRA-funded research activities.« less
  • This program investigated the potential for Alkali Metal Thermal to Electric Converter (AMTEC) technology to be useful in automotive power system applications. AMTEC, a thermally regenerative electrochemical energy conversion system, converts heat into electricity from a heat source at 750 C to 850 C and a radiator at 200 C to 350 C. AMTEC uses external combustion with correspondingly low emission of NO{sub x} and hydrocarbons, and can tolerate essentially any hydrocarbon fuel. Efficiencies of 20% to 30% are projected to be feasible for systems of 25 kWe to 40 kWe peak output. The research program has shown that theremore » are significant advantages to be achieved if AMTEC systems can be made cost effective for vehicle applications. Among these are (1) higher efficiency at part load than IC engines can yield, (2) omnifuel capability, and (3) low noise and low emission of pollutants. Demonstrated lifetimes already above 12,000 hours should be adequate for most vehicle applications. In major production, AMTEC costs are projected to reach $1/Watt, a value still too high for widespread automotive main power application. AMTEC's unique capabilities for low emissions, all-fuel operation, and insensitivity to ambient temperature, however, do make it a potential option for specialized vehicle applications needing these properties.« less
  • In order to understand how high performance computing (HPC) programs are developed, a series of experiments, using students in graduate level HPC classes and various research centers, were conducted at various locations in the US. In this report, we discuss this research, give some of the early results of those experiments, and describe a web-based Experiment Manager we are developing that allows us to run studies more easily and consistently at universities and laboratories, allowing us to generate results that more accurately reflect the process of building HPC programs.
  • This Phase II SBIR program was conducted to demonstrate the ability of a circulating fluidized bed (CFB) of flyash to cause particle agglomeration and consequent reduction in the quantity of fine particulate emissions from the system. Another objective was to show that carbon addition to the bed would result in the removal of mercury compounds from the flue gas at carbon utilization levels significantly better than duct injection of activated carbon. The pilot-scale testing was carried out in 1997. The pilot-scale fluid bed reactor was a 1000 CFM system, drawing gas from a slipstream of the exhaust of a 325more » MW coal-fired boiler. Flue gas for the pilot unit was drawn downstream of the air preheater and returned to the same unit. Particle agglomeration testing was carried out for which the parameters of gas flow rate and water evaporation rate were varied, and the particle size distribution leaving the fluid bed system was monitored. The bed was able to cause a reduction in total particulate concentration by a factor of 10 and in fine particulate concentration by a factor of 5, and it was found that the best agglomeration of particles was obtained with simultaneous water spray evaporation. Tests were then carried out in which activated carbon was added to the fluid bed for mercury adsorption. Carbon addition in the bed was shown to yield both higher Hg removal and higher carbon utilization than normal carbon addition with the bed. The fluid bed fly ash alone, without the injection of activated carbon, will capture 50% of the inlet Hg vapor. A total of 80% removal of Hg vapor is achieved with the addition to the bed of 1000 g iodine impregnated activated carbon per gram of inlet Hg. The ability of the fluid bed to capture SO{sub 2} and HCl was also evaluated, using hydrated lime added to the bed. It was found that the fluid bed alone, without lime injection, removed 16% of the SO{sub 2}. Complete utilization of hydrated lime is achieved for SO{sub 2} removal at mole ratios up to 0.66, and concomitant to the SO{sub 2} removal, over 90% of the HCl was removed. It is clear that this multipollutant control device can address both PM{sub 2.5} issues, as well as potential Hg vapor emission regulations.« less