As High-End Computing machines continue to grow in size, issues such as fault tolerance and reliability limit application scalability. Current techniques to ensure progress across faults, like checkpoint-restart, are unsuitable at these scale due to excessive overheads predicted to more than double an applications time to solution. Redundant computation, long used in distributed and mission critical systems, has been suggested as an alternative to checkpoint-restart on its own. In this paper we describe the rMPI library which enables portable and transparent redundant computation for MPI applications. We detail the design of the library as well as two replica consistency protocols, outline the overheads of this library at scale on a number of real-world applications, and finally outline the significant increase in an applications time to solution at extreme scale as well as show the scenarios in which redundant computation makes sense.
The annual program report provides detailed information about all aspects of the Sandia National Laboratories, California (SNL/CA) Waste Management Program. It functions as supporting documentation to the SNL/CA Environmental Management System Program Manual. This annual program report describes the activities undertaken during the past year, and activities planned in future years to implement the Waste Management (WM) Program, one of six programs that supports environmental management at SNL/CA.
The annual program report provides detailed information about all aspects of the SNL/CA Pollution Prevention Program for a given calendar year. It functions as supporting documentation to the SNL/CA Environmental Management System Program Manual. The program report describes the activities undertaken during the past year, and activities planned in future years to implement the Pollution Prevention Program, one of six programs that supports environmental management at SNL/CA. Pollution Prevention supports the goals and objectives to increase the procurement and use of environmentally friendly products and materials and minimize the generation of waste (nonhazardous, hazardous, radiological, wastewater). Through participation on the Interdisciplinary Team P2 provides guidance for integration of environmentally friendly purchasing and waste minimization requirements into projects during the planning phase. Table 7 presents SNL's corporate objectives and targets that support the elements of the Pollution Prevention program.
The National Water, Energy and Carbon Sequestration simulation model (WECSsim) is being developed to address the question, 'Where in the current and future U.S. fossil fuel based electricity generation fleet are there opportunities to couple CO{sub 2} storage and extracted water use, and what are the economic and water demand-related impacts of these systems compared to traditional power systems?' The WECSsim collaborative team initially applied this framework to a test case region in the San Juan Basin, New Mexico. Recently, the model has been expanded to incorporate the lower 48 states of the U.S. Significant effort has been spent characterizing locations throughout the U.S. where CO{sub 2} might be stored in saline formations including substantial data collection and analysis efforts to supplement the incomplete brine data offered in the NatCarb database. WECSsim calculates costs associated with CO{sub 2} capture and storage (CCS) for the power plant to saline formation combinations including parasitic energy costs of CO{sub 2} capture, CO{sub 2} pipelines, water treatment options, and the net benefit of water treatment for power plant cooling. Currently, the model can identify the least-cost deep saline formation CO{sub 2} storage option for any current or proposed coal or natural gas-fired power plant in the lower 48 states. Initial results suggest that additional, cumulative water withdrawals resulting from national scale CCS may range from 676 million gallons per day (MGD) to 30,155 MGD depending on the makeup power and cooling technologies being utilized. These demands represent 0.20% to 8.7% of the U.S. total fresh water withdrawals in the year 2000, respectively. These regional and ultimately nation-wide, bottom-up scenarios coupling power plants and saline formations throughout the U.S. can be used to support state or national energy development plans and strategies.
This paper reports the methodology for calculating present worth of system and operating costs for a number of energy storage technologies for representative electric utility applications. The values are an update from earlier reports, categorized by application use parameters. This work presents an update of energy storage system costs assessed previously and separately by the U.S. Department of Energy (DOE) Energy Storage Systems Program. The primary objective of the series of studies has been to express electricity storage benefits and costs using consistent assumptions, so that helpful benefit/cost comparisons can be made. Costs of energy storage systems depend not only on the type of technology, but also on the planned operation and especially the hours of storage needed. Calculating the present worth of life-cycle costs makes it possible to compare benefit values estimated on the same basis.
Node-based architecture (NBA) designs for future satellite projects hold the promise of decreasing system development time and costs, size, weight, and power and positioning the laboratory to address other emerging mission opportunities quickly. Reconfigurable field programmable gate array (FPGA)-based modules will comprise the core of several of the NBA nodes. Microprocessing capabilities will be necessary with varying degrees of mission-specific performance requirements on these nodes. To enable the flexibility of these reconfigurable nodes, it is advantageous to incorporate the microprocessor into the FPGA itself, either as a hard-core processor built into the FPGA or as a soft-core processor built out of FPGA elements. This document describes the evaluation of three reconfigurable FPGA-based soft-core processors for use in future NBA systems: the MicroBlaze (uB), the open-source Leon3, and the licensed Leon3. Two standard performance benchmark applications were developed for each processor. The first, Dhrystone, is a fixed-point operation metric. The second, Whetstone, is a floating-point operation metric. Several trials were run at varying code locations, loop counts, processor speeds, and cache configurations. FPGA resource utilization was recorded for each configuration.
Fast-curing impression materials are sometimes used to cast negative-mold replications of physical defects on material surfaces. The negative-mold impressions can then be used for further measurements to record the nature of the defect. These impression materials have been designed to cure quickly, and with very low adhesion, so that they can be easily removed from the surface leaving little residual contamination. Unfortunately, some contaminant is retained by the substrate material. This investigation seeks to identify the composition and quantity of the remaining material upon removal of Microset Synthetic Rubber Replicating Compound from several material surfaces. Coe-Flex was used as a relative comparison to Microset. On fifteen different substrate materials the Microset leaves no visible trace of contaminant, however, X-ray photoelectron spectroscopy shows evidence of a thin silicone-based contaminant film of approximately 2 nm thickness.