The 2nd NSTCCore Computing Service User Training Workshop in 2026: AI Applications with AMD

Wednesday 27 May 2026, 09:00 → 15:00 Asia/Taipei

1F Auditorium (Institute of Physics, Academia Sinica)

2026年 NSTCCore 高效能核心科學計算服務第二場教育訓練工作坊：AI應用（邀請AMD技術分享）

2026年度第二場 「NSTCCore 高效能核心科學計算服務教育訓練工作坊」 將在 2026年 5 月 27 日（三） 於 中央研究院物理研究所（台北）舉行。本工作坊由 國科會高效能核心科學計算服務計畫 支持，並由 中央研究院物理研究所網格計算中心（ASGC）執行， 特別感謝 超微半導體 參與技術分享。

本次特別邀請 超微半導體（AMD）講者，分享 AI 與高效能運算（HPC）之發展趨勢，協助研究團隊掌握最新技術與應用方向。

誠摯邀請各大專院校與研究機構，具科學計算與大數據分析需求的團隊踴躍參加。

活動重點

• 協助研究團隊便利使用 NSTCCore 核心計算服務
• AI／HPC 發展趨勢與產業觀點
• 跨 CPU／GPU／NPU 平台之 LLM 應用與部署
• 促進產學交流與跨領域合作機會

活動資訊

• 時間：2026 年 5 月 27 日（三）上午 9:00 ~ 下午 1:30
• 地點：台北中央研究院物理研究所 1F 演講廳
• 報名（5月24日截止）：https://indico4.twgrid.org/event/73/registrations/36/
• 問卷（5月27日開放）：https://indico4.twgrid.org/event/73/surveys/13

課程內容

• 高效能計算服務內容介紹
  。資源現況與合作模式
  。計算與儲存服務內容
  。使用者應用案例展示
   
• 基礎實作訓練
  。SLURM 與 DiCOSApp 計算服務操作（Batch 模式與 SaaS 模式）
  。資料儲存服務與操作示範
   
• AI 主題技術分享（AMD 講者）
  。AI 與高效能運算（HPC）趨勢及從資料中心到邊緣端的技術演進
  。基於 GPU 之 AI 訓練與加速技術（ROCm 最佳化）及異質運算實務應用
  。跨 CPU／GPU／NPU 平台之 LLM 部署方法與產業實務案例分享

注意事項

• 本次活動以中文為主，簡報資料提供中／英文版。
• 活動開放線上參與，報名時請選擇「線上參與」選項，連結將於會前寄送至報名者信箱。
• 受限於當天場地的網路環境，現場實作演練將以講者示範操作為主。本次仍將提供測試帳號，會前將統一寄送至報名者所填寫之電子郵件信箱；另亦可於現場提供，供與會者於會後使用。測試帳號之有效期限為 7 天。
• 為教學、推廣及成果報告用途，活動將進行錄音、錄影及拍照，若不希望入鏡或被錄音，請關閉攝影機與麥克風。
• 為響應一次用產品減量，鼓勵所有實體與會者自備環保杯與環保餐具。

如有任何疑問，請來信 dicos-support@twgrid.org，期待與您在工作坊相見！

---

The 2nd NSTCCore Computing Service User Training Workshop in 2026: AI Applications (AMD Technical Session)

The second NSTCCore High-Performance Computing (HPC) User Training Workshop of 2026 will be held on Wednesday, May 27, 2026, at Institute of Physics, Academia Sinica, Taipei. This workshop is supported by the NSTCCore HPC Service Project and organized by the Academia Sinica Grid Computing Centre (ASGC),  with the technical contribution of Advanced Micro Devices, Inc. (AMD). 

This workshop will feature an invited talk by AMD on AI and HPC trends, offering insights into emerging technologies and real-world applications.

We sincerely invite research teams from universities and research institutes with needs in scientific computing and big data analysis to actively participate in this event.

Workshop Highlights

• Overview of the NSTCCore program, services, resources, and collaboration models.
• Industry insights on AI and HPC trends.
• Efficient LLM deployment and inference across CPU, GPU, and NPU platforms.
• Foster collaboration and knowledge exchange between academia and industry.

Event Information

• Date & Time: 9:00AM ~ 1:30PM, Wednesday, May 27, 2026
• Venue: 1F Auditorium, Institute of Physics, Academia Sinica, Taipei
• Registration (Deadline: May 24): https://indico4.twgrid.org/event/73/registrations/36/
• Questionnaire (Open on May 27): https://indico4.twgrid.org/event/73/surveys/13

Program

• Introduction to HPC Services
  。Overview of resources and collaboration models
  。Computing and storage service features
  。User case demonstrations
   
• Hands-on Training
  。Practical introduction to SLURM & DiCOSApp (Batch and SaaS modes)
  。Data storage services and transfer demonstrations
   
• AI Technical Session (Invited Speaker from AMD)
  。AI & HPC trends and datacenter-to-edge evolution
  。GPU-based AI training and acceleration (ROCm optimization)
  。LLM deployment across CPU/GPU/NPU with industry use cases

Notes

• The workshop will be conducted primarily in Chinese, with presentation materials provided in Chinese or English.
• Online participation is available. Please select the “Online Participation” option when registering. The access link will be sent to each participant’s registered email address prior to the event.
• Due to network limitations at the venue, the hands-on session will be conducted primarily as a live demonstration by the speaker. Test accounts will still be provided and sent to each participant’s registered email address before the event, or distributed on-site for post-event use. Each test account will be valid for 7 days.
• This event will be recorded (audio and video) and photographed for training, outreach and reporting purposes. If you prefer not to be recorded, please turn off your camera and microphone.
• To support our efforts in reducing the use of disposable products, all in-person attendees are encouraged to bring their own reusable cups and cutlery.

For inquiries, please contact us at dicos-support@twgrid.org. We sincerely look forward to your participation!

 

 

Poster-NSTCCore Computing Service User Training Workshop on MAY 27, 2026.pdf

09:00 → 09:15 報到｜Registration

09:15 → 09:35 一、NSTCCore 高效能運算服務介紹 暨 AMD 受邀技術演講引言｜Introduction to NSTCCore Computing Services and Transition to the AMD Invited Technical Talk

• 計算服務架構 Introduction of Computing Service
• 運算資源、服務內容和計價模式 Resource, Service & Pricing
• 帳號申請與2FA Account Application & 2FA

Convener: Prof. Eric YEN (ASGC)

09:35 → 10:25 AMD 技術專題：從資料中心到客戶端：以 AMD AI 軟體、ROCm 與 NPU 部署大型語言模型（LLM）｜Invited Technical Talk: From Datacenter to Client: Deploying LLMs with AMD AI Software, ROCm, and NPUs (Presented by AMD)

This session explores how AMD’s AI software stack enables efficient LLM deployment from Datacenter to client across CPUs, GPUs, and NPUs. We will cover model development and optimization using ROCm in the data center, followed by seamless deployment to client platforms with AMD Ryzen™ AI NPUs. Using Lemonade, an open-source, OpenAI-compatible local LLM server, attendees will learn how to deploy LLMs with minimal code changes, leveraging heterogeneous execution across GPU and NPU to achieve low-latency, power-efficient, and private on-device generative AI experiences on PCs and edge devices.

Conveners: Mr Micky CHENG (AMD) , Mr Simon CHANG (AMD)

10:25 → 10:40 休息 Break

10:40 → 11:00 三、虛擬化SaaS雲端計算服務 - DiCOSApp｜Virtualized SaaS Cloud Computing Service – DiCOSApp

• 計算資源
• 服務架構
• 軟體部署

Convener: Ms Ming-Jyuan YANG (ASGC)

11:00 → 11:30 四、SLURM 計算服務｜SLURM Computing Service

• SLURM執行工作操作
• SLURM參數介紹
• 多核心程式編程及操作 Multi-Core Jobs

Convener: Ms Yi-Ru CHEN (ASGC)

11:30 → 12:00 五、資料服務｜Data Service

• 資料傳輸
• 資料操作

Convener: Mr Siang-Yu YANG (ASGC)

12:00 → 12:15 六、綜合討論與交流（問卷填寫）｜Discussion & Exchange (Questionnaire)

12:15 → 13:30 午餐 Lunch @ P7A Meeting Room

13:30 → 15:00 交流（內部會議）｜Networking Session (Closed Meeting)