Alan Gray – NVIDIA Technical Blog News and tutorials for developers, data scientists, and IT admins 2025-04-23T02:44:05Z http://www.open-lab.net/blog/feed/ Alan Gray <![CDATA[Spotlight: University of Tokyo Uses NVIDIA Grace Hopper for Groundbreaking Energy-Efficient Seismic Research]]> http://www.open-lab.net/blog/?p=96178 2025-04-23T02:44:05Z 2025-02-20T16:00:00Z Supercomputers are the engines of groundbreaking discoveries. From predicting extreme weather to advancing disease research and designing safer, more efficient...]]>

Supercomputers are the engines of groundbreaking discoveries. From predicting extreme weather to advancing disease research and designing safer, more efficient infrastructures, these machines simulate complex systems that are impractical to test in the real world due to their size, cost, and material requirements. Since the introduction of the GPU in 1999, NVIDIA has continually pushed the…

Source

]]>
Alan Gray <![CDATA[Optimizing llama.cpp AI Inference with CUDA Graphs]]> http://www.open-lab.net/blog/?p=86845 2024-11-14T16:03:17Z 2024-08-07T20:00:00Z The open-source llama.cpp code base was originally released in 2023 as a lightweight but efficient framework for performing inference on Meta Llama models....]]>

The open-source llama.cpp code base was originally released in 2023 as a lightweight but efficient framework for performing inference on Meta Llama models. Built on the GGML library released the previous year, llama.cpp quickly became attractive to many users and developers (particularly for use on personal workstations) due to its focus on C/C++ without the need for complex dependencies.

Source

]]>
Alan Gray <![CDATA[A Guide to CUDA Graphs in GROMACS 2023]]> http://www.open-lab.net/blog/?p=63250 2023-06-09T22:31:08Z 2023-04-14T18:10:14Z GPUs continue to get faster with each new generation, and it is often the case that each activity on the GPU (such as a kernel or memory copy) completes very...]]>

GPUs continue to get faster with each new generation, and it is often the case that each activity on the GPU (such as a kernel or memory copy) completes very quickly. In the past, each activity had to be separately scheduled (launched) by the CPU, and associated overheads could accumulate to become a performance bottleneck. The CUDA Graphs facility addresses this problem by enabling multiple GPU…

Source

]]>
1
Alan Gray <![CDATA[Massively Improved Multi-node NVIDIA GPU Scalability with GROMACS]]> http://www.open-lab.net/blog/?p=60772 2024-11-04T23:15:21Z 2023-02-09T17:00:00Z GROMACS, a scientific software package widely used for simulating biomolecular systems, plays a crucial role in comprehending important biological processes...]]>

GROMACS, a scientific software package widely used for simulating biomolecular systems, plays a crucial role in comprehending important biological processes important for disease prevention and treatment. GROMACS can use multiple GPUs in parallel to run each simulation as quickly as possible. Over the past several years, NVIDIA and the core GROMACS developers have collaborated on a series of…

Source

]]>
0
Alan Gray <![CDATA[Maximizing GROMACS Throughput with Multiple Simulations per GPU Using MPS and MIG]]> http://www.open-lab.net/blog/?p=37951 2022-08-21T23:52:43Z 2021-10-08T16:20:08Z GROMACS, a simulation package for biomolecular systems, is one of the most highly used scientific software applications worldwide, and a key tool in...]]>

GROMACS, a simulation package for biomolecular systems, is one of the most highly used scientific software applications worldwide, and a key tool in understanding important biological processes including those underlying the current COVID-19 pandemic. In a previous post, we showcased recent optimizations, performed in collaboration with the core development team, that enable GROMACS to…

Source

]]>
11
Alan Gray <![CDATA[Creating Faster Molecular Dynamics Simulations with GROMACS 2020]]> http://www.open-lab.net/blog/?p=16519 2023-07-12T23:16:45Z 2020-02-25T23:28:06Z GROMACS��one of the most widely used HPC applications�� has received a major upgrade with the release of GROMACS 2020. The new version includes exciting new...]]>

GROMACS—one of the most widely used HPC applications— has received a major upgrade with the release of GROMACS 2020. The new version includes exciting new performance improvements resulting from a long-term collaboration between NVIDIA and the core GROMACS developers. As a simulation package for biomolecular systems, GROMACS evolves particles using the Newtonian equations of motion.

Source

]]>
15
Alan Gray <![CDATA[Getting Started with CUDA Graphs]]> http://www.open-lab.net/blog/?p=15559 2023-02-10T20:49:15Z 2019-09-05T23:39:08Z The performance of GPU architectures continue to increase with every new generation. Modern GPUs are so fast that, in many cases of interest, the time taken by...]]>

The performance of GPU architectures continue to increase with every new generation. Modern GPUs are so fast that, in many cases of interest, the time taken by each GPU operation (e.g. kernel or memory copy) is now measured in microseconds. However, there are overheads associated with the submission of each operation to the GPU – also at the microsecond scale – which are now becoming significant…

Source

]]>
11
���˳���97caoporen����