18 645 how to write fast code

The fast evolution and increasing complexity of computing platforms pose a major challenge for developers of high performance software for engineering, science, and consumer applications: it becomes increasingly harder to harness the available computing power. Straightforward implementations may lose as much as one or two orders of magnitude in performance. On the other hand, creating optimal implementations requires the developer to have an understanding of algorithms, capabilities and limitations of compilers, and the target platforms architecture and microarchitecture. This interdisciplinary course introduces the student to the foundations and state-of-the-art techniques in high performance software development using important functionality such as linear algebra kernels, transforms, filters, and others as examples.

Course Information

The fast evolution and increasing complexity of computing platforms pose a major challenge for developers of high performance software for engineering, science, and consumer applications: it becomes increasingly harder to harness the available computing power. Straightforward implementations may lose as much as one or two orders of magnitude in performance. On the other hand, creating optimal implementations requires the developer to have an understanding of algorithms, capabilities and limitations of compilers, and the target platforms architecture and microarchitecture.

This interdisciplinary course introduces the student to the foundations and state-of-the-art techniques in high performance software development using important functionality such as linear algebra kernels, transforms, filters, and others as examples. The course will explain how to optimize for the memory hierarchy, take advantage of special instruction sets, and how to write parallel code for multicore, manycore, and cluster platforms, based on state-of-the-art research.

Further, a general strategy for performance analysis and optimization is introduced that the students will apply in group projects that accompany the course. Finally, the course will introduce the students to the recent field of automatic performance tuning. Description The fast evolution and increasing complexity of computing platforms pose a major challenge for developers of high performance software for engineering, science, and consumer applications: it becomes increasingly harder to harness the available computing power.

Instructors Ian Lane Jike Chong. Terms: Spring. Ask a Question Please login to post questions. No questions have been asked yet! Post your Review. Have you taken this course? Help others by posting your experience in this course.

Please keep your review less than characters. Specific comments about the course will be more useful than general assesments. Eg: 'People without programming may face this course find if they take it in the first semester. John Doe is the best!!! Please login to post reviews.

How to Write Fast Code. Units: Description: The fast evolution and increasing complexity of computing platforms pose a major. Homework submission. Submit the homework as pdf. Name your file '​parrotsprint.co.nz' where. 'userid' is your andrew user id. parrotsprint.co.nz file must include.

The fast evolution and increasing complexity of computing platforms pose a major challenge for developers of high performance software for engineering, science, and consumer applications: it becomes increasingly harder to harness the available computing power. Straightforward implementations may lose as much as one or two orders of magnitude in performance. On the other hand, creating optimal implementations requires the developer to have an understanding of algorithms, capabilities and limitations of compilers, and the target platform's architecture and microarchitecture.

Embed Size px x x x x Idea, benefits, reasons, restrictionsHistory and state-of-the-art floating-point SIMD extensionsHow to use it: compiler vectorization, class library, intrinsics, inline assembly.

The fast evolution and increasing complexity of computing platforms pose a major challenge for developers of high performance software for engineering, science, and consumer applications: it becomes increasingly harder to harness the available computing power. Straightforward implementations may lose as much as one or two orders of magnitude in performance. On the other hand, creating optimal implementations requires the developer to have an understanding of algorithms, capabilities and limitations of compilers, and the target platform's architecture and microarchitecture.

How to Write Fast Numerical Code: A Small Introduction

The complexity of modern computing platforms has made it extremely difficult to write numerical code that achieves the best possible performance. Straightforward implementations based on algorithms that minimize the operations count often fall short in performance by at least one order of magnitude. Further, program generators are discussed as a way to reduce the implementation and optimization effort. Two running examples are used to demonstrate these techniques: matrix-matrix multiplication and the discrete Fourier transform. Unable to display preview. Download preview PDF.