Simulation Overhead Research Articles

Multiprocessor enhancements of the SimpleScalar tool set

This paper describes multiprocessor enhancements of the SimpleScalar tool set. The core simulation code has been modified to support multiprocessing, and a run-time library has been introduced for thread creation and synchronization. Measurements using the SPLASH-2 parallel benchmark suite [13] indicate that the multiprocessor enhancements introduce simulation overhead of 30%-50% relative to the original uniprocessor simulator. An idealized multiprocessor cache simulator has also been developed, and a dynamic visualization tool for cache coherence has also been developed. These multiprocessor enhancements are available at the WWW site for the SimpleScalar tool set.

Switching activity generation with automated BIST synthesis for performance testing of interconnects

A novel scheme of synthesizing nonlinear feedback shift register structures that can be superimposed on the boundary of the component of a system under test to generate interconnect switching activities that resemble real life interconnect switching profiles is proposed. The goal is to perform at-speed interconnect test while simultaneously capturing the dynamic switching effects such as crosstalk and ground bounce, as accurately as possible during interconnect built-in self-test. A library of nonlinear feedback shift register structures called precharacterized test pattern generators (P-TPGs) is constructed. Components of P-TPGs can be modeled using Markov chain and can be interconnected together in specific ways to recreate the switching activity profile of the interconnections being tested. The unique advantage of this scheme is that there is no simulation overhead since P-TPG components are precharacterized by solving Markov equations analytically. An integrated genetic algorithm-based search and optimization technique for finding the best P-TPG component among various possible implementations and matching its activity profiles with those of the interconnections under test has been designed and implemented synthesis for testability allows generation of the worst case interconnect switching activities. Experimental results confirm the validity of our approach.

An investigation into the attenuation of wind speed by the use of windbreaks in the vicinity of overhead wires

An experimental and computational analysis of the efficacy of windbreaks for the attenuation of wind speed in the vicinity of overhead wires is presented. This work has been specifically applied to the case of oscillation of overhead catenary contact wires in a cross-wind. A wind tunnel subscale model of a windbreak/overhead wire configuration is presented and discussed with regard to the feasibility of reducing the aerodynamically induced oscillations encountered in such situations. This porous blockage placed upstream of the simulated overhead wire attempts to represent the influence of a windbreak/shelterbelt on the reduction of the vibration phenomenon known as galloping. In conjunction with this experimental study, a computational fluid dynamics (CFD) analysis has been carried out in order to assess the influence of the shelterbelt on the catenary wires in a typical full-scale scenario. The results indicate that a dense shelterbelt of height H positioned between 2-8 tree heights, H, upstream of the suspended cables provides a sufficient reduction in wind speed to reduce significantly the amplitude of wire vibration.

Scaling simulation of the fusing-restricted reconfigurable mesh

This paper deals with the ability of a model to adapt algorithm instances of different sizes to run on a given model size without significant loss of efficiency. The overhead in simulating a step of a large instance of the model on a smaller instance can quantify this ability. A reconfigurable mesh (R-Mesh) can use its bus structure as a computational resource, presenting an obstacle to efficiently scaling down algorithms to run on a smaller R-Mesh. We construct a scaling simulation of a Fusing-Restricted Reconfigurable Mesh (FR-Mesh), a version of the R-Mesh. The overhead of this simulation depends only on the simulating machine size and not on the simulated machine size. Previously, the R-Mesh was not known to admit such a simulation overhead without significantly reducing its computational power. The small overhead holds importance for flexibility in algorithm design and for running algorithms with various input sizes on an available model of given size. The results of this paper extend to a variety of concurrent write rules and also translate to an improved scaling simulation of an unrestricted R-Mesh.

Analysis of one-line ground fault surge on simulated overhead distribution line

One-line fault location techniques of intermittent ground fault points on nongrounded distribution lines are necessary to supply steady electric power for customers. This paper proposes a novel method that uses frequency spectra of transient waveforms at a sending end when an intermittent ground fault occurs. In this method, if a ground fault position differs, it causes a different frequency spectrum pattern for the current or voltage transients at a send end (decided by surge reflection phenomena between the sending end and the ground fault point). Comparison between experimental data of one-line ground fault tests on a simulated overhead line (373 m line length, two branch points) and digital simulation analyses shows that the grounded capacitance of a power source transformer was useful in locating the ground fault point. © 1997 Scripta Technica, Inc. Electr Eng Jpn, 118 (1): 52–60, 1997

Analysis of one‐line ground fault surge on simulated overhead distribution line

Analysis of one‐line ground fault surge on simulated overhead distribution line

Improving conservative VHDL simulation performance by reduction of feedback

This paper describes two forms of feedback in the simulation runtime of VHDL circuits that greatly influences performance. While circuit feedback and strongly connected components have been observed and documented as detrimental influences to conservative parallel discrete event simulation (PDES) efficiency, that influence has never been quantified. Moreover, in this study, the phenomenon of induced feedback was observed to diminish speedup to the same degree as explicit feedback. In this paper the influence of feedback on simulation runtime is analyzed and an algorithm for its elimination is presented. In addition, a metric for the quantification of feedback is introduced. By measuring feedback, it is possible to balance its influence on simulation runtime with that of other factors (e.g. load balance, number of processors, machine granularity, etc. ) through the use of a cost-based partitioning approach. This paper reports significant improvements in runtime for three circuits due to the prevention of feedback using the partitioning algorithm presented. In addition, strong correlation between the feedback metric and conservative parallel simulation overhead is demonstrated.

Performance analysis of parallel simulation on distributed systems

This paper presents an analytical model to evaluate the performance of parallel simulation on distributed computing platforms. The proposed model is formalized by two important time components in parallel and distributed processing: computation time and communication time. A conservative parallel simulation of multistage interconnection networks is used as an example in our analytical model. Performance metrics such as elapsed time, speedup and simulation bandwidth associated with different schemes for partitioning/mapping parallel simulation onto distributed processors are evaluated. Our mathematical analysis identifies the major constituents of simulation overheads in these mapping strategies necessary for improving parallel simulation efficiency. We also show that a perfectly balanced workload distribution may not necessarily translate into better performance. On the contrary, we have shown that a balanced mapping of workload may increase communication overheads resulting in a longer simulation elapsed time. Our performance model has been validated against implementation results from a parallel simulation model. The analytical framework is also practical to evaluate the runtime efficiency of other simulation applications which are based on the conservative paradigm.

模擬架空配電線における一線地絡事故サージの解析

模擬架空配電線における一線地絡事故サージの解析

Cyclic Sprinkler Irrigation and Pre-irrigation Substrate Water Content Affect Water and N Leaching from Containers

Abstract A series of sprinkler irrigation experiments were conducted to determine the influences of water application rate (WAR), pre-irrigation substrate water content (PSWC), and cyclic irrigation on water and N leaching from container-grown plants. Prior to experiments, Marigold (Tagetes erecta L. ‘Apollo’), were glass house-grown in pine bark-filled 3.8 liter (1 gal) containers. Prior to treatment, substrate was dried via evapotranspiration (ET) to targeted PSWCs. A simulated overhead irrigation system applied the daily water allotment in a single continuous application or cyclically (multiple applications); in most cases the respective ET volumes were applied to the substrate. Water application efficiency (WAE; water vol retained in substrate + water vol applied to substrate) was determined, and in some experiments, leachates were analyzed for EC, NO3-N and NH4-N. A negative linear relationship existed between WAR and WAE. Leachate NO3-N and NH4-N concentrations were unaffected by WAR, however, total N leached increased with increasing WAR. WAE of cyclic irrigation was 4% higher (absolute basis) than with continuous irrigation; WAE increased as the time interval between cyclic applications increased from 20 to 60 min. Regardless of how water was applied, WAE was inversely related to PSWC and application volume. These experiments showed that the most effective method to increase WAE is to irrigate at relatively low PSWCs; if irrigation occurs at relatively high PSWCs, then relatively low volumes should be applied.

INTERMITTENT SPRINKLER IRRIGATION INCREASES IRRIGATION APPLICATION EFFICIENCY

Experiments were conducted to determine if applying a container-grown plant's daily water allotment in multiple applications (intermittently) increased application efficiency relative to the allotment applied in a single application (continuous). Water was applied (simulated overhead sprinkler system) to marigold plants in a pine bark substrate. Time interval between intermittent applications and water movement in the substrate were investigated. Application efficiency was greater when the water allotment was applied intermittently compared to a single application. Efficiency was also greater when the interval between applications was increased from one to two h. Sectioning substrate into top, middle, and bottom thirds showed that the bottom layer of the intermittent treatment contained more water than the bottom layer for the continuous treatment. There were no differences in water content in the top and middle layers between the two treatments.

A class of systems with nearly zero distributed simulation overhead

We identify a class of systems that can be simulated by a distributed simulator with almost no overhead messages. In general terms, a system in this class is a network of processes that communicate with each other solely via messages in an asynchronous manner. The communication delays in this network are allowed to be arbitrary, though finite, and the communication lines are assumed to be first in, first out (FIFO). The class of systems defined in this paper includes many cyclic systems. To show that these systems can be simulated with nearly zero overhead, we present such a distributed simulation algorithm for systems in this class. By identifying this class of systems and designing a distributed simulation algorithm specifically geared for it, we have been able to avoid almost all overhead messages. By contrast, other distributed simulation algorithms are usually designed for more general systems and they usually require too many overhead messages. In particular, in simulating the class of systems defined in this paper, these general-purpose schemes would usually involve much more overhead than our distributed simulation algorithm.

Systems with low distributed simulation overhead

Distributed simulation often involves a large number of overhead messages. This may lead to severe performance degradation of the simulator as pointed out in several previous works. In this work special classes of systems which can be simulated with low overhead-in most systems identified here the total number of overhead messages on any given communication line is at most one, irrespective of the time up to which the physical system is to be simulated-are presented. This includes acyclic systems, some closed queuing networks, some open cyclic queuing networks, and other special types of cyclic networks. Distributed simulation algorithms specially tailored to these systems are given. By isolating these classes of systems and designing distributed simulation algorithms specifically geared for them, a large number of overhead messages can be avoided. By contrast, other distributed simulation algorithms are usually designed for more general systems and they usually require too many overhead messages, even in simulating these special classes of systems. >

Yield performance of rainfed lowland rice as affected by nursery fertilization under conditions of intermediate deepwater (15?50 cm) and flash floods

The effect of simulated overhead flash flooding for 10 days at an early vegetative stage of rice grown from transplanted seedlings with and without fertilizer N applied to the nursery soil, was studied under intermediate deepwater conditions at Cuttack, India. Application of 100 kg fertilizer N per ha to nursery soil at sowing or in splits improved seedling vigor measured as dry weight and N concentration of the plant tissue at transplanting. The crop raised with healthy and vigorous N-fertilized seedlings was more tolerant to ravages of flooding and showed a better stand with more profuse tillering and less mortality. The grain yield of the semi-tall cultivar Suresh was not affected by N applied to the nursery soil under partial submergence of 70±2 cm water depth. However, the semi-dwarf cultivar Gayatri planted with N-fertilized seedlings established better in the early stages and produced more panicles and a two- to five-time higher grain yield. The beneficial effect of nursery fertilization was less marked under natural flooding conditions (25–40 cm). Transplanting semi-dwarf rice with healthy and vigorous seedlings obtained by fertilizing nursery seed-bed could therefore, be a very productive low-cost technology for farmers growing rice under conditions of intermediate deepwater and flash floods.

Wear in simulated railway overhead current collection systems

The wear rates of the pantograph contact strips and the overhead contact copper wire are investigated using a wear machine simulating a railway current collection system. The contact strip material used for the tests is the actual copper and lead-impregnated carbon used by British Rail (MY7D Grade) on all pantographs. Wear rates of the contact strip and copper disk (simulating the overhead contact wire) are presented and discussed. Disk wear is measured in situ by the spherical indentations method. The wear performance at speeds from 60 to 250 km h −1 is examined with and without current collection. The dependence of the wear rate on the current (0–300 A both a.c. and d.c.), the static contact force (1.0–250 N) and the dwell time for intermittent sliding has been determined. A comparison with previous results is given where applicable. The dwell time, representing the time lapse between pantograph passes, is shown to have a crucial effect on the copper wear depth per passage owing to the formation and subsequent removal of an oxide layer on the copper surface.