國外 Julia 核心開發者 Katharine Hyatt 邀請演講
A tour of GPGPU and Julia
Julia claims to solve the so-called "two-language problem", allowing programmers to use high level constructs and enjoy expressive, straightforward syntax without sacrificing performance. Can we achieve this promise for general purpose GPU programming, which has often required either lots of opaque boilerplate code or writing functions which look like language chimeras? I'll give an introduction to a few parts of Julia's GPU ecosystem, especially CuArrays.jl and the native Julia CUDA kernel writing system, with some examples of how we can write performant code which is still readable. Finally, I'll present some directions we're currently working on which provide opportunities for new contributors.
講者
Katharine Hyatt
Julia 的核心開發者之一，為 Simons Foundation 下，Flatiron Institute 的計算量子物理中心的博士後研究員。她專注於開發數值計算方法，她的作品為 ITensors.jl，還參與 CuArrays.jl 等眾多套件的實作。
議程
18：30 餐點時間及自由交流
19：30 邀請演講
20：30 會眾心得分享與討論
Julia's Knights 連線直播
當天晚上會與成功大學學生社團 Julia's Knights 連線直播活動現場。無法到台北的朋友可以考慮到台南成功大學一起參與。
地點：成功大學統計系系館 62223 教室
議程
18：30 入場
19：00 連線直播
P.S. 感謝天瓏書局的場地支持，天瓏書局的場地需要場地費，為每人一百元，在入場時繳交至櫃台。
