
天氣和氣象
全球各地的氣候中心都依賴於強大的高效能運算 (HPC),而 Altair 可確保 HPC 實現可靠、高效運作。由於地球的天氣和氣象難以進行建模,因此您會需要協調極其複雜、具備強大工作負載 HPC 系統和軟體。氣候建模和氣象預報領域對全球經濟和社會都會產生重要影響。

天氣和氣象建模
Altair 的運算解決方案助力世界各地的主要氣象中心。 時間始終是一個關鍵因素,而結果對現實世界具有重大影響,因此我們的高性能計算 (HPC) 工作負載協調、資源管理、使用者訪問和分析解決方案可確保天氣和氣候建模與仿真工作負載快速高效地運作,並最大限度地運用 HPC 資源。
我們如何幫助您優化天氣建模和模擬?
聯繫我們
PBS Professional 和 Cylc 工作流引擎
Cylc 是用於循環系統的開源工作流引擎,可處理各種複雜的工作流。 它能根據詳細的時程表和依賴關係自動執行任務,在天氣和氣候建模、數值天氣預報、物理仿真和數據處理等領域特別有用。
Cylc 已成為世界各地主要天氣和氣候中心的常用工具,現在又成為澳大利亞氣象局的重要工具。 Cylc 最初是由紐西蘭國家水與大氣研究所 (NIWA) 開發的運作環境預測工具,現已成為 NIWA、英國氣象局 (Met Office),歐洲天氣和氣候仿真卓越中心 (Center of Excellence for Weather and Climate Simulation in Europe, ESiWACE)、氣象局及其他出資者之間的一種開源協作工具。
Altair、NIWA 和澳大利亞氣象局一起通過 Altair® PBS Professional®(我們行業領先的 HPC 和雲環境工作負載管理器和作業調度器)開發了一種用於監視多個 Cylc 工作流性能的生產環境。 Altair 能夠為基於社區的 Cylc 工作流引擎提供商業支援。

Altair 天氣解決方案
Altair 的天氣解決方案基於適用於 HPC 和雲環境的 Altair PBS Professional™ 工作負載管理器而構建,並與 Cylc 工作流引擎進行結合。 該解決方案讓使用者能夠監控 HPC 硬體、Cylc 套件和 PBS Professional 作業,還可提供簡明清晰的狀態報告。 Altair 天氣解決方案專用於應對複雜的氣候建模和預測工作負載,採用模組化設計,能夠輕鬆地直接進行部署,它包含以下功能:
- 消息查看器:查看 Cylc 和 PBS Professional 消息
- 事件檢查器:瞭解系統事件
- 消息修飾符:用於開發和支援
- 進度預測工具:工作流進度的視覺表現形式
特色資源

Sustainability and Climate Protection Through Digitalization
How to choose the right material, the right design, and the right process to develop solutions that minimize carbon emissions and bolster your bottom line.
A step-by-step guide for transforming your plastics engineering value chain to develop solutions that optimize performance, minimize carbon emissions, and gain a competitive edge.
What you will learn:
- Material selection and testing to minimize carbon emissions with Global Warming Potential evaluations alongside traditional performance and cost metrics
- A step-by-step simulation-driven approach to accelerate design space creation, explore variants, and better optimize lightweight plastic parts
- Ways to predict and prevent critical problems such as injection mold core fractures or failures to optimize production operations
- Improve manufacturability using multiscale material modeling
- How to gain a competitive edge with faster time to market through first-time-right injection molding methods, simulation and rapid variant analyses to optimize part, mold and parameters for manufacturing

HPC in Weather and Climate
Weather and climate centers worldwide rely on the power of high-performance computing (HPC) to track, analyze, simulate and predict the behavior of weather systems, climate patterns and other atmospheric phenomena. In this roundtable, experts from leading weather and climate centers across the globe will detail how they harness HPC to fuel this critically important research.
Topics will include:?
- Advanced workload orchestration to optimize resource utilization and accelerate research timelines?
- Maintaining critical uptime: alerting, support and redundancy measures?
- Storage and I/O optimization to manage massive data volumes?
- Machine learning and artificial intelligence to accelerate simulation timelines and improve prediction

Predicting Global Weather & Climate - New NCAR HPC System Expected to Achieve Over 3x Faster Performance
HPC systems at NCAR power the science behind global weather and climate prediction and research. In early 2021, NCAR announced that their Cheyenne supercomputer, located at the NCAR-Wyoming Supercomputing Center, will be replaced by a new HPE Cray EX system with 3.5x more processing power, up to 19.87 petaflops with a combination of CPU and GPU nodes. The new system will include efficient workload management and scheduling by Altair PBS Professional and Altair Accelerator Plus.

Job Scheduling on NCAR's Next-generation Supercomputer
Computational and Information Systems Laboratory (CISL) supports and advances this research in the geosciences with the world-class HPC and high-performance storage.
For workload management and resource scheduling, NCAR uses Altair PBS Professional to exploit features like cloud bursting, fairshare, power-user and maintenance reservations, resource assignment with control groups, high-throughput hierarchical scheduling, green provisioning, and energy-aware scheduling.
This presentation by Irfan Elahi, Brian Vanderwende, and John Blaas from the National Center for Atmospheric Research (NCAR) was aired during the 2021 Altair HPC Summit, and is about 29 minutes long.
