Loop perforation

Last updated

Loop perforation is an approximate computing technique that allows to regularly skip some iterations of a loop. [1] [2] [3]

Contents

It relies on one parameter: the perforation rate. The perforation rate can be interpreted as the number of iteration to skip each time or the number of iterations to perform before skipping one.

Variants of loop perforation include those that skip iterations deterministically at regular intervals, those that skip iterations at the beginning or the end of the loop, and those that skip a random sample of iterations. The compiler may select the perforation variant at the compile-time, or include instrumentation that allows the runtime system to adaptively adjust the perforation strategy and perforation rate to satisfy the end-to-end accuracy goal.

Loop perforation techniques were first developed by MIT senior researchers Martin C. Rinard and Stelios Sidiroglou.

Code examples

The examples that follows provide the result of loop perforation applied on this C-like source code

for(inti=0;i<N;i++){// do things}

Skip n iterations each time

for(inti=0;i<N;i++){// do thingsi=i+skip_factor;}

Skip one iteration after n

intcount=0;for(inti=0;i<N;i++){if(count==skip_factor){count=0;}else{// do thingscount++;}}

See also

  1. ↑ Henry Hoffmann, Sasa Misailovic, Stelios Sidiroglou, Anant Agarwal, Martin Rinard "Using Code Perforation to Improve Performance, Reduce Energy Consumption, and Respond to Failures" MIT CSAIL Tech. Report 2009-042, September 2009
  2. ↑ Sasa Misailovic, Stelios Sidiroglou, Henry Hoffmann, Martin C. Rinard "Quality of Service Profiling" 32nd International Conference on Software Engineering (ICSE 2010). May 2010.
  3. ↑ Steilos Sidiroglou, Sasa Misailovic, Henry Hoffmann, and Martin Rinard. "Managing Performance vs. Accuracy Trade-offs With Loop Perforation." ESEC/FSE. September, 2011


Related Research Articles

<span class="mw-page-title-main">Insertion sort</span> Sorting algorithm

Insertion sort is a simple sorting algorithm that builds the final sorted array (or list) one item at a time by comparisons. It is much less efficient on large lists than more advanced algorithms such as quicksort, heapsort, or merge sort. However, insertion sort provides several advantages:

In computing, an optimizing compiler is a compiler that tries to minimize or maximize some attributes of an executable computer program. Common requirements are to minimize a program's execution time, memory footprint, storage size, and power consumption.

In computer science, control flow is the order in which individual statements, instructions or function calls of an imperative program are executed or evaluated. The emphasis on explicit control flow distinguishes an imperative programming language from a declarative programming language.

In computing, a vector processor or array processor is a central processing unit (CPU) that implements an instruction set where its instructions are designed to operate efficiently and effectively on large one-dimensional arrays of data called vectors. This is in contrast to scalar processors, whose instructions operate on single data items only, and in contrast to some of those same scalar processors having additional single instruction, multiple data (SIMD) or SWAR Arithmetic Units. Vector processors can greatly improve performance on certain workloads, notably numerical simulation and similar tasks. Vector processing techniques also operate in video-game console hardware and in graphics accelerators.

Standard ML (SML) is a general-purpose, modular, functional programming language with compile-time type checking and type inference. It is popular among compiler writers and programming language researchers, as well as in the development of theorem provers.

In computer science, divide and conquer is an algorithm design paradigm. A divide-and-conquer algorithm recursively breaks down a problem into two or more sub-problems of the same or related type, until these become simple enough to be solved directly. The solutions to the sub-problems are then combined to give a solution to the original problem.

<span class="mw-page-title-main">OpenMP</span> Open standard for parallelizing

OpenMP is an application programming interface (API) that supports multi-platform shared-memory multiprocessing programming in C, C++, and Fortran, on many platforms, instruction-set architectures and operating systems, including Solaris, AIX, FreeBSD, HP-UX, Linux, macOS, and Windows. It consists of a set of compiler directives, library routines, and environment variables that influence run-time behavior.

<span class="mw-page-title-main">For loop</span> Control flow statement for repeated execution

In computer science a for-loop or for loop is a control flow statement for specifying iteration. Specifically, a for loop functions by running a section of code repeatedly until a certain condition has been satisfied.

In computer science, a generator is a routine that can be used to control the iteration behaviour of a loop. All generators are also iterators. A generator is very similar to a function that returns an array, in that a generator has parameters, can be called, and generates a sequence of values. However, instead of building an array containing all the values and returning them all at once, a generator yields the values one at a time, which requires less memory and allows the caller to get started processing the first few values immediately. In short, a generator looks like a function but behaves like an iterator.

In computer science and particularly in compiler design, loop nest optimization (LNO) is an optimization technique that applies a set of loop transformations for the purpose of locality optimization or parallelization or another loop overhead reduction of the loop nests. One classical usage is to reduce memory access latency or the cache bandwidth necessary due to cache reuse for some common linear algebra algorithms.

Cilk, Cilk++, Cilk Plus and OpenCilk are general-purpose programming languages designed for multithreaded parallel computing. They are based on the C and C++ programming languages, which they extend with constructs to express parallel loops and the fork–join idiom.

Loop unrolling, also known as loop unwinding, is a loop transformation technique that attempts to optimize a program's execution speed at the expense of its binary size, which is an approach known as space–time tradeoff. The transformation can be undertaken manually by the programmer or by an optimizing compiler. On modern processors, loop unrolling is often counterproductive, as the increased code size can cause more cache misses; cf. Duff's device.

In compiler theory, loop optimization is the process of increasing execution speed and reducing the overheads associated with loops. It plays an important role in improving cache performance and making effective use of parallel processing capabilities. Most execution time of a scientific program is spent on loops; as such, many compiler optimization techniques have been developed to make them faster.

Automatic vectorization, in parallel computing, is a special case of automatic parallelization, where a computer program is converted from a scalar implementation, which processes a single pair of operands at a time, to a vector implementation, which processes one operation on multiple pairs of operands at once. For example, modern conventional computers, including specialized supercomputers, typically have vector operations that simultaneously perform operations such as the following four additions :

In computer science, software pipelining is a technique used to optimize loops, in a manner that parallels hardware pipelining. Software pipelining is a type of out-of-order execution, except that the reordering is done by a compiler instead of the processor. Some computer architectures have explicit support for software pipelining, notably Intel's IA-64 architecture.

<span class="mw-page-title-main">Recursion (computer science)</span> Use of functions that call themselves

In computer science, recursion is a method of solving a computational problem where the solution depends on solutions to smaller instances of the same problem. Recursion solves such recursive problems by using functions that call themselves from within their own code. The approach can be applied to many types of problems, and recursion is one of the central ideas of computer science.

The power of recursion evidently lies in the possibility of defining an infinite set of objects by a finite statement. In the same manner, an infinite number of computations can be described by a finite recursive program, even if this program contains no explicit repetitions.

In computer science, a loop variant is a mathematical function defined on the state space of a computer program whose value is monotonically decreased with respect to a (strict) well-founded relation by the iteration of a while loop under some invariant conditions, thereby ensuring its termination. A loop variant whose range is restricted to the non-negative integers is also known as a bound function, because in this case it provides a trivial upper bound on the number of iterations of a loop before it terminates. However, a loop variant may be transfinite, and thus is not necessarily restricted to integer values.

Tracing just-in-time compilation is a technique used by virtual machines to optimize the execution of a program at runtime. This is done by recording a linear sequence of frequently executed operations, compiling them to native machine code and executing them. This is opposed to traditional just-in-time (JIT) compilers that work on a per-method basis.

Task skipping is an approximate computing technique that allows to skip code blocks according to a specific boolean condition to be checked at run-time.

<span class="mw-page-title-main">Plotting algorithms for the Mandelbrot set</span> Algorithms and methods of plotting the Mandelbrot set on a computing device

There are many programs and algorithms used to plot the Mandelbrot set and other fractals, some of which are described in fractal-generating software. These programs use a variety of algorithms to determine the color of individual pixels efficiently.

This page is based on this Wikipedia article
Text is available under the CC BY-SA 4.0 license; additional terms may apply.
Images, videos and audio are available under their respective licenses.