scholarly journals Salsa: An Automatic Tool to Improve the Numerical Accuracy of Programs

10.29007/j2fd ◽  
2018 ◽  
Author(s):  
Nasrine Damouche ◽  
Matthieu Martel
Keyword(s):  
Numerical Analysis ◽  
Embedded Systems ◽  
Software Architecture ◽  
Static Analysis ◽  
Abstract Interpretation ◽  
Experimental Results ◽  
Floating Point ◽  
Numerical Accuracy ◽  
Automatic Tool ◽  
Concise Description

This article describes Salsa, an automatic tool to improve the accuracy of the floating- point computations done in numerical codes. Based on static analysis methods by abstract interpretation, our tool takes as input an original program, applies to it a set of transformations and then generates an optimized program which is more accurate than the initial one. The original and the transformed programs are written in the same imperative language. This article is a concise description of former work on the techniques implemented in Salsa, extended with a presentation of the main software architecture, the inputs and outputs of the tool as well as experimental results obtained by applying our tool on a set of sample programs coming from embedded systems and numerical analysis.

scholarly journals FORWARD AND BACKWARD STATIC ANALYSIS FOR CRITICAL NUMERICAL ACCURACY IN FLOATING POINT PROGRAMS

2020 ◽  
Vol 21 (2) ◽  
Author(s):  
Somasundaram Kanagasabapathi ◽  
MG Thushara
Keyword(s):  
Static Analysis ◽  
Abstract Interpretation ◽  
Experimental Results ◽  
Floating Point ◽  
Numerical Accuracy ◽  
Main Approach ◽  
Backward Analysis

In this article, we introduce a new static analysis for numerical accuracy. Weaddress the problem of determining the minimal accuracy on the inputs and on the intermediary results of a program containing  foating-point computations in order to ensure a desired accuracy on the outputs. The main approach is to combine a forward and a backward static analysis, done by abstract interpretation. The backward analysis computes the minimal accuracy needed for the inputs and intermediary results of the program in order to ensure a desired accuracy on the results, specied by the user. In practice, the information collected by our analysis may help to optimize the formats used to represent the values stored in the variables of the program or to select the appropriate sensors. To illustrate our analysis, we have shown a prototype example with experimental results.

scholarly journals Static analysis of bond between prestressing strand and UHPC exposed to elevated temperatures

2021 ◽  
Vol 3 (1) ◽  
pp. 102-107
Author(s):  
David Čítek ◽  
Jindřich Čech ◽  
Petr Pokorný ◽  
Jiří Kolísko
Keyword(s):  
Numerical Analysis ◽  
Static Analysis ◽  
Material Properties ◽  
Elevated Temperatures ◽  
Cement Composite ◽  
Composite Model ◽  
Experimental Results ◽  
Reference Temperature ◽  
Pulling Force ◽  
Prestressing Strand

This article deals with the numerical analysis of the bond of a prestressing strand with UHPC heated to elevated temperatures and subsequently cooled. Numerical analysis was performed with experimentally determined material properties of UHPC at a reference temperature of 20 °C and further after heating to temperatures of 200, 400 and 600 °C and subsequent annealing to normal temperature. The resulting deflection depending on the pulling force from the numerical analysis were compared with the experimental results. The results are used to configure the cement composite model for more demanding simulations of structures and load cases.

scholarly journals Improving the Numerical Accuracy of High Performance Computing Programs by Process Specialization

10.29007/tfls ◽  
2018 ◽  
Author(s):  
Farah Benmouhoub ◽  
Nasrine Damouche ◽  
Matthieu Martel
Keyword(s):  
High Performance Computing ◽  
High Performance ◽  
Abstract Interpretation ◽  
Critical Level ◽  
Floating Point ◽  
Real Numbers ◽  
Numerical Accuracy ◽  
Finite Approximations ◽  
Performance Computing ◽  
Point Arithmetic

In high performance computing, nearly all the implementations and published experiments use floating-point arithmetic. However, since floating-point numbers are finite approximations of real numbers, it may result in hazards because of the accumulated errors. These round-off errors may cause damages whose gravity varies depending on the critical level of the application. To deal with this issue, we have developed a tool which im- proves the numerical accuracy of computations by automatically transforming programs in a source-to-source manner. Our transformation, relies on static analysis by abstract interpretation and operates on pieces of code with assignments, conditionals, loops, functions and arrays. In this article, we apply our techniques to optimize a parallel program representative of the high performance computing domain. Parallelism introduces new numerical accuracy problems due to the order of operations in this kind of systems. We are also interested in studying the compromise between execution time and numerical accuracy.

Compiler-Aided Run-Time Performance Speed-Up in Super-Scalar Processor

10.28945/3391 ◽  
2009 ◽  
Author(s):  
Moshe Pelleh
Keyword(s):  
Embedded Systems ◽  
Embedded System ◽  
General Purpose ◽  
Experimental Results ◽  
Medium Size ◽  
Organic System ◽  
Software Application ◽  
Organic Systems ◽  
Special Cases ◽  
Run Time

In our world, where most systems become embedded systems, the approach of designing embedded systems is still frequently similar to the approach of designing organic systems (or not embedded systems). An organic system, like a personal computer or a work station, must be able to run any task submitted to it at any time (with certain constrains depending on the machine). Consequently, it must have a sophisticated general purpose Operating System (OS) to schedule, dispatch, maintain and monitor the tasks and assist them in special cases (particularly communication and synchronization between them and with external devices). These OSs require an overhead on the memory, on the cache and on the run time. Moreover, generally they are task oriented rather than machine oriented; therefore the processor's throughput is penalized. On the other hand, an embedded system, like an Anti-lock Braking System (ABS), executes always the same software application. Frequently it is a small or medium size system, or made up of several such systems. Many small or medium size embedded systems, with limited number of tasks, can be scheduled by our proposed hardware architecture, based on the Motorola 500MHz MPC7410 processor, enhancing its throughput and avoiding the software OS overhead, complexity, maintenance and price. Encouraged by our experimental results, we shall develop a compiler to assist our method. In the meantime we will present here our proposal and the experimental results.

Inducing Frictional Force to Enhance the Transient Response in Beams

2019 ◽  
Vol 22 (2) ◽  
pp. 88-93
Author(s):  
Hamed Khanger Mina ◽  
Waleed K. Al-Ashtrai
Keyword(s):  
Numerical Analysis ◽  
Transient Response ◽  
Frictional Force ◽  
Damping Ratio ◽  
Experimental Results ◽  
Damping Capacity ◽  
Tightening Force ◽  
Contact Areas ◽  
Good Agreement ◽  
Ansys Workbench

This paper studies the effect of contact areas on the transient response of mechanical structures. Precisely, it investigates replacing the ordinary beam of a structure by two beams of half the thickness, which are joined by bolts. The response of these beams is controlled by adjusting the tightening of the connecting bolts and hence changing the magnitude of the induced frictional force between the two beams which affect the beams damping capacity. A cantilever of two beams joined together by bolts has been investigated numerically and experimentally. The numerical analysis was performed using ANSYS-Workbench version 17.2. A good agreement between the numerical and experimental results has been obtained. In general, results showed that the two beams vibrate independently when the bolts were loosed and the structure stiffness is about 20 N/m and the damping ratio is about 0.008. With increasing the bolts tightening, the stiffness and the damping ratio of the structure were also increased till they reach their maximum values when the tightening force equals to 8330 N, where the structure now has stiffness equals to 88 N/m and the damping ratio is about 0.062. Beyond this force value, increasing the bolts tightening has no effect on stiffness of the structure while the damping ratio is decreased until it returned to 0.008 when the bolts tightening becomes immense and the beams behave as one beam of double thickness.

Platform Independent Analysis of Probabilities on Multithreaded Programs

2013 ◽  
Vol 1 (3) ◽  
pp. 48-65
Author(s):  
Yuting Chen
Keyword(s):  
Operating System ◽  
Static Analysis ◽  
Main Idea ◽  
Experimental Results ◽  
Machine To Machine ◽  
Concurrent Program ◽  
Acceptable Accuracy ◽  
Multithreaded Programs ◽  
Independent Analysis ◽  
And Performance

A concurrent program is intuitively associated with probability: the executions of the program can produce nondeterministic execution program paths due to the interleavings of threads, whereas some paths can always be executed more frequently than the others. An exploration of the probabilities on the execution paths is expected to provide engineers or compilers with support in helping, either at coding phase or at compile time, to optimize some hottest paths. However, it is not easy to take a static analysis of the probabilities on a concurrent program in that the scheduling of threads of a concurrent program usually depends on the operating system and hardware (e.g., processor) on which the program is executed, which may be vary from machine to machine. In this paper the authors propose a platform independent approach, called ProbPP, to analyzing probabilities on the execution paths of the multithreaded programs. The main idea of ProbPP is to calculate the probabilities on the basis of two kinds of probabilities: Primitive Dependent Probabilities (PDPs) representing the control dependent probabilities among the program statements and Thread Execution Probabilities (TEPs) representing the probabilities of threads being scheduled to execute. The authors have also conducted two preliminary experiments to evaluate the effectiveness and performance of ProbPP, and the experimental results show that ProbPP can provide engineers with acceptable accuracy.

A Survey of Parametric Static Analysis

2021 ◽  
Vol 54 (7) ◽  
pp. 1-37
Author(s):  
Jihyeok Park ◽  
Hongki Lee ◽  
Sukyoung Ryu
Keyword(s):  
Recent Work ◽  
Static Analysis ◽  
Abstract Interpretation ◽  
Black Box ◽  
Parameter Selection ◽  
Survey Analysis ◽  
Future Directions ◽  
Analysis Techniques ◽  
Analysis Quality

Understanding program behaviors is important to verify program properties or to optimize programs. Static analysis is a widely used technique to approximate program behaviors via abstract interpretation. To evaluate the quality of static analysis, researchers have used three metrics: performance, precision, and soundness. The static analysis quality depends on the analysis techniques used, but the best combination of such techniques may be different for different programs. To find the best combination of analysis techniques for specific programs, recent work has proposed parametric static analysis . It considers static analysis as black-box parameterized by analysis parameters , which are techniques that may be configured without analysis details. We formally define the parametric static analysis, and we survey analysis parameters and their parameter selection in the literature. We also discuss open challenges and future directions of the parametric static analysis.

Numerical simulation and experimental performance research of cylindrical vane pump

2021 ◽  
pp. 095440622110200
Author(s):  
Yiqi Cheng ◽  
Xinhua Wang ◽  
Waheed Ur Rehman ◽  
Tao Sun ◽  
Hasan Shahzad ◽  
...  
Keyword(s):  
Numerical Analysis ◽  
Rotational Velocity ◽  
Geometric Theory ◽  
Mechanical Efficiency ◽  
Operating Conditions ◽  
Experimental Results ◽  
Field Analysis ◽  
Total Efficiency ◽  
Three Dimensional Model ◽  
Vane Pump

This study presents a novel cylindrical vane pump based on the traditional working principle. The efficiency of the cylindrical vane pump was verified by experimental validation and numerical analysis. Numerical analysis, such as kinematics analysis, was performed in Pro/Mechanism and unsteady flow-field analysis was performed using ANSYS FLUENT. The stator surface equations were derived using the geometric theory of the applied spatial triangulation function. A three-dimensional model of the cylindrical vane pump was established with the help of MATLAB and Pro/E. The kinematic analysis helped in developing kinematic equations for cylindrical vane pumps and proved the effectiveness of the structural design. The maximum inaccuracy error of the computational fluid dynamics (CFD) model was 5.7% compared with the experimental results, and the CFD results show that the structure of the pump was reasonable. An experimental test bench was developed, and the results were in excellent agreement with the numerical results of CFD. The experimental results show that the cylindrical vane pump satisfied the three-element design of a positive-displacement pump and the trend of changes in efficiency was the same for all types of efficiency under different operating conditions. Furthermore, the volumetric efficiency presented a nonlinear positive correlation with increased rotational velocity, the mechanical efficiency showed a nonlinear negative correlation, and the total efficiency first increased and then decreased. When the rotational velocity was 1.33[Formula: see text] and the discharge pressure was 0.68[Formula: see text], the total efficiency reached its maximum value.

Simulation of a Free Standing Riser Model and Validation With Experimental Results

2014 ◽  
Author(s):  
Marcio Yamamoto ◽  
Sotaro Masanobu ◽  
Satoru Takano ◽  
Shigeo Kanada ◽  
Tomo Fujiwara ◽  
...  
Keyword(s):  
Numerical Simulation ◽  
Numerical Analysis ◽  
Previous Experiment ◽  
Numerical Results ◽  
Experimental Results ◽  
Previous Article ◽  
Scale Model ◽  
Analysis Software ◽  
Free Standing ◽  
Reduced Scale

In this article, we present the numerical analysis of a Free Standing Riser. The numerical simulation was carried out using a commercial riser analysis software suit. The numerical model’s dimensions were the same of a 1/70 reduced scale model deployed in a previous experiment. The numerical results were compared with experimental results presented in a previous article [1]. Discussion about the model and limitations of the numerical analysis is included.

Nonlinear Numerical Analysis of SRC Frame Middle Joints

2013 ◽  
Vol 376 ◽  
pp. 231-235
Author(s):  
Cheng Li ◽  
Yun Zou ◽  
Jie Kong ◽  
Zhi Wei Wan
Keyword(s):  
Finite Element ◽  
Numerical Analysis ◽  
Bearing Capacity ◽  
Axial Load ◽  
Load Ratio ◽  
Experimental Results ◽  
Steel Ratio ◽  
Finite Element Software ◽  
Axial Load Ratio ◽  
Hysteretic Performance

Nonlinear numerical analysis for the force performance of frame middle joint is processed in this paper with the finite element software of ABAQUS. Compared with experimental results, numerical analysis results are found to be reasonable. Then the influence of factors such as shaped steel ratio and axial-load ratio are contrastively analyzed. The results show that shaped steel ratio has a greater influence on the bearing capacity and hysteretic performance of the structure, but the axial-load ratio has less influence.

Sign in / Sign up

Export Citation Format

Share Document