Christian Hundt – NVIDIA Technical Blog News and tutorials for developers, data scientists, and IT admins 2023-10-25T21:14:09Z http://www.open-lab.net/blog/feed/ Christian Hundt <![CDATA[Modeling Earth��s Atmosphere with Spherical Fourier Neural Operators]]> http://www.open-lab.net/blog/?p=68433 2023-10-25T21:14:09Z 2023-07-27T16:06:59Z Machine learning-based weather prediction has emerged as a promising complement to traditional numerical weather prediction (NWP) models. Models such as NVIDIA...]]>

Machine learning-based weather prediction has emerged as a promising complement to traditional numerical weather prediction (NWP) models. Models such as NVIDIA FourCastNet have demonstrated that the computational time for generating weather forecasts can be reduced from hours to mere seconds, a significant improvement to current NWP-based workflows. Traditional methods are formulated from…

Source

]]>
2
Christian Hundt <![CDATA[Maximizing Performance with Massively Parallel Hash Maps on GPUs]]> http://www.open-lab.net/blog/?p=61480 2023-05-23T23:50:12Z 2023-03-06T17:30:00Z Decades of computer science history have been devoted to devising solutions for efficient storage and retrieval of information. Hash maps (or hash tables) are a...]]>

Decades of computer science history have been devoted to devising solutions for efficient storage and retrieval of information. Hash maps (or hash tables) are a popular data structure for information storage given their amortized, constant-time guarantees for the insertion and retrieval of elements. However, despite their prevalence, hash maps are seldom discussed in the context of GPU…

Source

]]>
1
Christian Hundt <![CDATA[Machine Learning Frameworks Interoperability, Part 3: Zero-Copy in Action using an E2E Pipeline]]> http://www.open-lab.net/blog/?p=36553 2022-08-21T23:52:32Z 2021-09-01T15:00:00Z Introduction Efficient pipeline design is crucial for data scientists. When composing complex end-to-end workflows, you may choose from a wide variety of...]]>

Efficient pipeline design is crucial for data scientists. When composing complex end-to-end workflows, you may choose from a wide variety of building blocks, each of them specialized for a dedicated task. Unfortunately, repeatedly converting between data formats is an error-prone and performance-degrading endeavor. Let’s change that! In this blog series, we discuss different aspects of…

Source

]]>
0
Christian Hundt <![CDATA[Machine Learning Frameworks Interoperability, Part 2: Data Loading and Data Transfer Bottlenecks]]> http://www.open-lab.net/blog/?p=35948 2022-08-21T23:52:27Z 2021-08-17T16:30:00Z Efficient pipeline design is crucial for data scientists. When composing complex end-to-end workflows, you may choose from a wide variety of building blocks,...]]>

Efficient pipeline design is crucial for data scientists. When composing complex end-to-end workflows, you may choose from a wide variety of building blocks, each of them specialized for a dedicated task. Unfortunately, repeatedly converting between data formats is an error-prone and performance-degrading endeavor. Let’s change that! In this post series, we discuss different aspects of…

Source

]]>
0
Christian Hundt <![CDATA[Machine Learning Frameworks Interoperability, Part 1: Memory Layouts and Memory Pools]]> http://www.open-lab.net/blog/?p=35645 2022-08-21T23:52:23Z 2021-08-09T15:00:00Z Introduction Efficient pipeline design is crucial for data scientists. When composing complex end-to-end workflows, you may choose from a wide variety of...]]>

Efficient pipeline design is crucial for data scientists. When composing complex end-to-end workflows, you may choose from a wide variety of building blocks, each of them specialized for a dedicated task. Unfortunately, repeatedly converting between data formats is an error-prone and performance-degrading endeavor. Let’s change that! In this post series, we discuss different aspects of…

Source

]]>
0
Christian Hundt <![CDATA[Aligning Time Series at the Speed of Light]]> http://www.open-lab.net/blog/?p=30974 2022-08-21T23:51:28Z 2021-04-29T22:45:29Z To say it with the words of Eamonn Keogh: ��Time series is a ubiquitous and increasingly prevalent type of data [��]��. Virtually any incrementally measured...]]>

To say it with the words of Eamonn Keogh: “Time series is a ubiquitous and increasingly prevalent type of data […]”. Virtually any incrementally measured signal, be it along a time axis or a linearly ordered set, can be treated as time series. Examples include electrocardiograms, temperature or voltage measurements, audio, server logs, but also heavy-weight data such as video and time-resolved MRI…

Source

]]>
1
Christian Hundt <![CDATA[Streaming Interactive Deep Learning Applications at Peak Performance]]> http://www.open-lab.net/blog/?p=20528 2022-08-21T23:40:37Z 2020-09-01T17:25:39Z Imagine that you have just finished implementing an awesome, interactive, deep learning pipeline on your NVIDIA-accelerated data science workstation using...]]>

Imagine that you have just finished implementing an awesome, interactive, deep learning pipeline on your NVIDIA-accelerated data science workstation using OpenCV for capturing your webcam stream and rendering the output. A colleague of yours mentions that exploiting the novel TF32 compute mode of the Ampere microarchitecture third-generation Tensor Cores might significantly accelerate your…

Source

]]>
2
���˳���97caoporen����