數值軟體開發

The course introduces the art of building numerical software, i.e., computer programs applying numerical methods for solving mathematical, scientific, or engineering problems. We will be using Python, C++, and other tools (bash, git, make, etc.) to show the development process. You will learn: * practical software development by starting an open-source project * real-world applications of C++ * software engineering for scientific and high-performance computing * essentials of professional software development

Critical discussion is required for the practical course and the project development. Students are required to attend all meetings. If you don't feel comfortable expressing yourself publicly, this course is not designed for you. Read the full syllabus at https://yyc.solvcon.net/en/latest/nsd/schedule/26au_nycu/schedule26au.html before taking the course. Students should have taken computer architecture and engineering mathematics or equivalents. Working knowledge of Linux and Unix-like is required. Prior knowledge to numerical methods is good to have. The instructor uses English in the lectures and discussions. Students are required to use English for the homework assignments and term project, including the presentation, and pro-actively speak English. The lecture meetings are not only for lecturing but also for discussing. Students are required to attend every meeting to discuss ideas and progress of the project. Discussion requires hard-learning skills, and students are **strongly recommended** to also take the course Academic English Writing (permanent course code GELT00014: https://timetable.nycu.edu.tw/?r=main/crsoutline&Acy=114&Sem=2&CrsNo=562044) if they have not.

無備註

The instructor will use English. Students are required to use English for the homework assignments and term project, including the presentation, and pro-actively speak English. Students should attend all meetings. Computer program for homework and project should be developed against the latest Ubuntu LTS system.

* You are expected to learn programming languages yourself. Students are encouraged to form study groups for practicing C++, and discuss with the instructor and/or the teaching assistant. * Grading: homework 30%, mid-term exam: 30%, term project: 40%. * There are 12 lectures for the subjects of numerical software developing using Python and C++. * There will be 6 homework assignments for you to exercise. Programming in Python and/or C++ is required. * Mid-term examination will be conducted to assess students' understandings to the analytical materials. * Term project will be used to assess students' overall coding skills. Presentation is required. Failure to present results in 0 point for this part.

• Introduction

  What is numerical software.

  講授：

  3

• Fundamental engineering

  A large chunk of efforts is spent in the infrastructure for coding. The key to the engineering system is automation.

  講授：

  3

• Python and numpy

  Python is a popular choice for the scripting engine that makes the numerical software work as a platform.

  講授：

  3

• C++ and computer architecture

  The low-level code of numerical software must be high-performance. The industries chose C++ because it can take advantage of everything that a hardware architecture offers while using any level of abstraction.

  講授：

  3

• Matrix operations

  Matrices are everywhere in numerical analysis. Arrays are the fundamental data structure and used for matrix-vector, matrix-matrix, and other linear algebraic operations.

  講授：

  3

• Cache optimization

  How cache works, its importance to performance, and optimization with cache.

  講授：

  3

• SIMD (vector processing)

  Parallelism and x86 assembly for SIMD.

  講授：

  3

• Memory management

  Numerical software tends to use as much memory as a workstation has. The memory has two major uses: (i) to hold the required huge amount of data, and (ii) to gain speed.

  講授：

  3

• Ownership and smart pointers

  Ownership and memory management using C++ smart pointers.

  講授：

  3

• Modern C++

  Copy elision and move semantics. Variadic template and perfect forwarding. Closure.

  講授：

  3

• C++ and C for Python

  Use C++ and C to control the CPython interpreter.

  講授：

  3

• Array code in C++

  Dissect the array-based code and element-based code and when to use them.

  講授：

  3

• Advanced Python

  Advanced topics in Python programming.

  講授：

  3

Course notes: https://yyc.solvcon.net/en/latest/nsd/ Textbook: None References: * Computer Systems: A Programmer's Perspective: https://csapp.cs.cmu.edu/ * Python documentation: https://docs.python.org/3/ * Cppreference: https://en.cppreference.com/ * Effective Modern C++, Scott Meyer, O'Reilly, 2014 * modmesh (example system of this course) source code: https://github.com/solvcon/modmesh * pybind11 source code: https://github.com/pybind/pybind11 * cpython source code: https://github.com/python/cpython * numpy source code: https://github.com/numpy/numpy

地點

教師未提供此項資料

時間

Use the sciwork meetup at NYCU (see https://sciwork.dev/meetup/ )

聯絡方式

Send email to yyc at sciwork.dev, with "[nsd-course]" prefix in the subject line. Or go to sciwork meetup: https://sciwork.dev/meetup/ .