NVIDIA 加速的資料科學

唯一針對資料科學最佳化的硬體到應用軟體堆疊

GPU 加速你的資料科學工作流程

資料科學工作流程一直以來都十分緩慢且缺乏效率,仰賴 CPU 來載入、篩選和操控資料,以及訓練和部署模型。GPU 大幅降低基礎架構成本,並為使用 NVIDIA RAPIDS ™ 函式庫的端對端資料科學工作流程提供卓越效能。隨處都能使用 GPU 加速的資料科學,無論是桌上型電腦、資料中心、邊緣端及雲端都沒問題。

 

功能與優點

提供最佳生產力

提供最佳生產力

減少等待最寶貴分析洞見的時間,並加速投資報酬率。

成效更高

成效更高

加快機器學習速度最高達 215 倍,並執行更多迭代、增加實驗,以及進行更深度的探索。

經濟實惠

經濟實惠

降低資料科學基礎架構成本,同時提升資料中心效率。

RAPIDS 為 Apache Spark 3.0 的 GPU 加速

Apache Spark 3.0 是 Spark 系列的第一個版本,可為分析和人工智慧工作負載提供完全整合且順暢的 GPU 加速效能。無論在本機或雲端,都可以利用 Spark 3.0 (含 GPU ) 的強大功能,且無需變更程式碼。GPU 的突破性效能夠讓企業和研究人員更頻繁訓練更大的模型,最終借助人工智慧功能發揮巨量資料的價值。

使用 NVIDIA GPU 訓練 XGBOOST

GPU 加速 XGBoost 將為全球頂尖機器學習演算法的單一節點與分散式部署帶來顛覆的效能﹐資料科學團隊藉由比 CPU 大幅提升的訓練速度,能夠處理更龐大的資料集、迭代更快速並調整模型,以將預測精準度與商業價值極大化。

Data Prep

XGBoost

End-to-end

現在開始瞭解如何以 GPU 加速 XGBoost

適用於資料科學的 NVIDIA GPU 解決方案

透過多種不同的 NVIDIA GPU 解決方案,探索無與倫比的加速能力。

PC

著手使用機器學習。

工作站

全新的資料科學工作站。

資料中心

適用企業生產的人工智慧系統。

雲端

多功能加速機器學習。

透過 GPU 加速進行中的業務

讓機器學習工作流程展現最佳效能、生產力和投資報酬率。

RAPIDS: 資料科學函式庫套件

利用 NVIDIA CUDA-X AI 打造的 RAPIDS 運用 NVIDIA® CUDA® 逾 15 年的開發與機器學習專業知識。這款功能強大的應用軟體能在 GPU 中完整執行端對端資料科學訓練流程,將原本需要好幾天的訓練時間縮短為幾分鐘。

NVIDIA RAPIDS Flow
End-to-End Faster Speeds on RAPIDS

RAPIDS, a GPU-accelerated data science platform, is a next-generation computational ecosystem powered by Apache Arrow. The NVIDIA collaboration with Ursa Labs will accelerate the pace of innovation in the core Arrow libraries and help bring about major performance boosts in analytics and feature engineering workloads.

- Wes McKinney, Head of Ursa Labs and Creator of Apache Arrow and Pandas

At Databricks, we are excited about RAPIDS’ potential to accelerate Apache Spark workloads. We have multiple ongoing projects to integrate Spark better with native accelerators, including Apache Arrow support and GPU scheduling with Project Hydrogen. We believe that RAPIDS is an exciting new opportunity to scale our customers' data science and AI workloads.

- Matei Zaharia, co-founder and CTO of Databricks, and the original creator of Apache Spark

I got 24x speedup using RAPIDS XGBOOST and can now replace hundreds of CPU nodes, running my biggest ML workload on a single node with 8 GPUs. You made XGBOOST too fast!?

- Streaming Media Company

My previous bottleneck was I/O. …10 minutes to pull in data for 10 stores (about 1 million rows). With RAPIDS, we can pull in data for about 6000 stores (millions of rows) in less than 3 minutes. That scale could have easily taken us 4 days on legacy infrastructure … just plain awesome.

- A mid-market specialty retailer with 6000 stores

RAPIDS, a GPU-accelerated data science platform, is a next-generation computational ecosystem powered by Apache Arrow. The NVIDIA collaboration with Ursa Labs will accelerate the pace of innovation in the core Arrow libraries and help bring about major performance boosts in analytics and feature engineering workloads.

- Wes McKinney, Head of Ursa Labs and Creator of Apache Arrow and Pandas

At Databricks, we are excited about RAPIDS’ potential to accelerate Apache Spark workloads. We have multiple ongoing projects to integrate Spark better with native accelerators, including Apache Arrow support and GPU scheduling with Project Hydrogen. We believe that RAPIDS is an exciting new opportunity to scale our customers' data science and AI workloads.

- Matei Zaharia, co-founder and CTO of Databricks, and the original creator of Apache Spark

I got 24x speedup using RAPIDS XGBOOST and can now replace hundreds of CPU nodes, running my biggest ML workload on a single node with 8 GPUs. You made XGBOOST too fast!?

- Streaming Media Company

My previous bottleneck was I/O. …10 minutes to pull in data for 10 stores (about 1 million rows). With RAPIDS, we can pull in data for about 6000 stores (millions of rows) in less than 3 minutes. That scale could have easily taken us 4 days on legacy infrastructure … just plain awesome.

- A mid-market specialty retailer with 6000 stores

RAPIDS, a GPU-accelerated data science platform, is a next-generation computational ecosystem powered by Apache Arrow. The NVIDIA collaboration with Ursa Labs will accelerate the pace of innovation in the core Arrow libraries and help bring about major performance boosts in analytics and feature engineering workloads.

- Wes McKinney, Head of Ursa Labs and Creator of Apache Arrow and Pandas

At Databricks, we are excited about RAPIDS’ potential to accelerate Apache Spark workloads. We have multiple ongoing projects to integrate Spark better with native accelerators, including Apache Arrow support and GPU scheduling with Project Hydrogen. We believe that RAPIDS is an exciting new opportunity to scale our customers' data science and AI workloads.

- Matei Zaharia, co-founder and CTO of Databricks, and founder of Apache Spark

I got 24x speedup using RAPIDS XGBOOST and can now replace hundreds of CPU nodes, running my biggest ML workload on a single node with 8 GPUs. You made XGBOOST too fast!?

- Streaming Media Company

My previous bottleneck was I/O. …10 minutes to pull in data for 10 stores (about 1 million rows). With RAPIDS, we can pull in data for about 6000 stores (millions of rows) in less than 3 minutes. That scale could have easily taken us 4 days on legacy infrastructure … just plain awesome.

- A mid-market specialty retailer with 6000 stores

合作夥伴生態系統

RAPIDS 開放所有人使用,並且在資料科學與分析領域獲得全球廣泛採用。我們的合作夥伴透過 GPU 加速分析、機器學習和深度學習的進展,共同改變傳統巨量資料分析的生態系統。

ANACONDA
BlazingDB
Chainer
Datalogue
DataBricks
DellEMC
FastData
Graphistry
H20.ai
HPE
IBM
Kinetica
MAPR
NetApp
Omni Sci
Oracle
Pure Storage
PyTorch
SAP
Sas
Sqream
ZILLIZ
ANACONDA
BlazingDB
Chainer
Datalogue
DataBricks
DellEMC
FastData
Graphistry
H20.ai
HPE
IBM
Kinetica
MAPR
NetApp
Omni Sci
Oracle
Pure Storage
PyTorch
SAP
Sas
Sqream
ZILLIZ

線上研討會

透過 RAPIDS 加速資料科學工作流程

透過 XGBoost 提升機器學習效能與生產力

適用於健康照護產業 GPU 加速資料科學的 RAPIDS

透過 RAPIDS 與 DGX-2 加速端對端資料科學

探索 GPU 加速硬體解決方案