Precise Data Locality Optimization of Nested Loops

A significant source for enhancing application performance and for reducing power consumption in embedded processor applications is to improve the usage of the memory hierarchy. In this paper, a temporal and spatial locality optimization framework of nested loops is proposed, driven by parameterized cost functions. The considered loops can be imperfectly nested. New data layouts are propagated through the connected references and through the loop nests as constraints for optimizing the next connected reference in the same nest or in the other ones. Unlike many existing methods, special attention is paid to TLB (Translation Lookaside Buffer) effectiveness since TLB misses can take from tens to hundreds of processor cycles. Our approach only considers active data, that is, array elements that are actually accessed by a loop, in order to prevent useless memory loads and take advantage of storage compression and temporal locality. Moreover, the same data transformation is not necessarily applied to a whole array. Depending on the referenced data subsets, the transformation can result in different data layouts for a same array. This can significantly improve the performance since a priori incompatible references can be simultaneously optimized. Finally, the process does not only consider the innermost loop level but all levels. Hence, large strides when control returns to the enclosing loop are avoided in several cases, and better optimization is provided in the case of a small index range of the innermost loop.     

Insights into the behavior of LWR steel containment buildings during severe accidents

Investigations into the performance of steel containment subject to pressure and temperature greater than their design basis loads are discussed. The timing, mechanism, and location of a containment failure, i.e., release of radioactive material, have an important impact on the consequences of a severe accident. We review the results of experiments on steel containment models pressurized to failure, on aged and unaged seals subjected to elevated temperature and pressure, and on electrical penetration assemblies tested for leakage. Based on the results, the important features and details of analytical methods that can be used to predict containment performance are identified. Finally, we speculate on the performance of steel containments in severe accident conditions.     

Amplifying four-wavelength combiner, based on erbium/ytterbium-doped waveguide amplifiers and integrated splitters

We report the successful system demonstration of a four-wavelength integrated-optics amplifying combiner. The arrangement consists of an all-connectorized 4/spl times/1 glass splitter followed by a 4.5-cm-long Er/Yb-doped waveguide amplifier. When injecting 120 mW of 975-nm laser diode pump, we record, in the amplifying section, 11.6 dB of net gain in the single pass configuration and 23 dB in the double pass as well as a noise figure of 4.5 dB. These results show the potentiality of ion-exchange technology for the fabrication of lossless telecommunication devices.     

Application of Symbolic Approach to the Bernstein Expansion for Program Analysis and Optimization

Mathematical packages for static analysis of programs have recently been developed. Although these packages are widely used, they have a number of limitations. In particular, they do not support multivariate polynomials with integer coefficients, which are often met in programs and used for the analysis of systems. Some methods to overcome this difficulty have already been suggested, but, unfortunately, they can be applied to only a subclass of such expressions. In this paper, we suggest a more general approach based on the Bernstein expansion, which facilitates the analysis of integer multivariate polynomials.     

Parametric Analysis of Polyhedral Iteration Spaces

In the area of automatic parallelization of programs, analyzing and transforming loop nests with parametric affine loop bounds requires fundamental mathematical results. The most common geometrical model of iteration spaces, called the polytope model, is based on mathematics dealing with convex and discrete geometry, linear programming, combinatorics and geometry of numbers.In this paper, we present automatic methods for computing the parametric vertices and the Ehrhart polynomial, i.e., a parametric expression of the number of integer points, of a polytope defined by a set of parametric linear constraints.These methods have many applications in analysis and transformations of nested loop programs. The paper is illustrated with exact symbolic array dataflow analysis, estimation of execution time, and with the computation of the maximum available parallelism of given loop nests.     

Leak rate test of personnel airlock for LWR containment subject to pressures and temperatures beyond design

As part of the U.S. Nuclear Regulatory Commission's (USNRC) Containment Integrity Program, a full-size personnel airlock for a nuclear containment building was subjected to conditions simulating a severe accident.The objective of the test was to characterize the performance of an airlock when subjected to conditions that exceeded design. The gasket tested was a “double dog-ear” configuration made from an elastomer known as EPDM E603. The data obtained from this test will be used by SNL as a benchmark for development of analytical methods.Strain, temperature, displacements, pressure, and leak rate data were measured and recorded from over 330 transducers. The test lasted approximately 60 hours. Data were recorded at regular intervals during heating, pressurization and depressurization.The airlock was originally designed for 340°F and 60 psig. The airlock inner door and bulkhead were exposed to a maximum air temperature of approximately 850°F and a maximum air pressure of 300 psig. Two heating and pressurization cycles were planned; one to heat the air to 400°F and pressurize to 300 psig, and the second to heat to 800°F and pressurize to 300 psig. No significant leakage was recorded during these two cycles. A third cycle was added to the test program. The air temperature was increased to approximately 850°F and held at this temperature for nearly 12 hours. Pressure was increased and the inner door seal failed at a pressure of 150.5 psig. The maximum leak rate recorded past the inner door seal was 706 SCFM. The outer door seal did not fail.     

Ruby masers for maximum G/Top

The Deep-Space Network (DSN) includes world-wide networks of 26-, 34- and 70-m antennas in Australia, Spain, and the USA. Ruby masers are used on the 34and 70-m antennas to maximize the system operating noise temperature and thereby maximize the DSN receiving system figure of merit, antenna gain divided by system operating noise temperature. These systems are used for deep-space telecommunications, solar system radar, and radio astronomy. Cavity, traveling-wave, and reflected-wave maser designs and performance characteristics from 960 MHz to 34 GHz are summarized. Effective noise temperatures of ruby masers are addressed with emphasis on a 33.68-GHz maser where quantum noise is a major source of noise