GTC HPC Presentations

Starting on October 5-9, This fall’s GTC will run continuously for five days, across seven time zones. The conference will showcase the latest breakthroughs in…

Nefi Alarcon
4 min readintermediate
--
View Original

Overview

The GTC HPC conference, occurring from October 5-9, will highlight advancements in High-Performance Computing (HPC) and GPU technologies. Attendees can participate in live events or access a catalog of on-demand content focused on innovative HPC applications.

What You'll Learn

1

How to leverage NVIDIA Ampere GPU architecture for scientific discovery

2

Why the Summit supercomputer is effective for drug discovery during pandemics

3

How to implement GPUDirect Storage for I/O-bound workflows

4

When to use InfiniBand for high-speed data throughput in supercomputing

Key Questions Answered

How does the NVIDIA Ampere GPU architecture enhance scientific discovery?
The NVIDIA Ampere GPU architecture integrates traditional simulations with deep learning and big data analytics, enabling researchers to solve complex scientific problems more efficiently. Real-world examples demonstrate its application in accelerating discoveries across various fields.
What is the role of the Summit supercomputer in COVID-19 drug discovery?
The Summit supercomputer facilitated an in silico drug discovery pipeline that screened two billion drug candidates against the SARS-CoV-2 main protease. This innovative approach was developed in response to the COVID-19 pandemic and is currently in experimental testing phases.
What benefits does GPUDirect Storage provide for HPC workloads?
GPUDirect Storage allows a direct data path between local storage and GPU memory, eliminating CPU bottlenecks and enhancing performance for I/O-bound workflows. This technology is crucial for accelerating computing on GPUs without the need to fit datasets into local memory.
What advantages does InfiniBand offer for supercomputing networks?
InfiniBand provides high-speed data throughput and low latency, essential for connecting compute elements in supercomputers. Its in-network computing capabilities enable faster data analysis and scalability, making it a preferred choice for leading supercomputing systems.

Technologies & Tools

Hardware
Nvidia Ampere GPU Architecture
Used to accelerate scientific discovery through advanced computing techniques.
Hardware
Summit Supercomputer
Facilitates drug discovery by screening billions of drug candidates.
Technology
Gpudirect Storage
Enables direct data paths between storage and GPU memory for enhanced performance.
Network Technology
Infiniband
Provides high-speed data throughput and low latency for supercomputing networks.

Key Actionable Insights

1
Leverage the NVIDIA Ampere GPU architecture to enhance your research capabilities.
By integrating deep learning and big data analytics with traditional simulations, researchers can solve complex problems more effectively, leading to faster scientific discoveries.
2
Utilize the Summit supercomputer for drug discovery efforts, especially in response to pandemics.
The end-to-end drug discovery pipeline developed on the Summit supercomputer showcases its potential in rapidly screening drug candidates, which can significantly expedite the development of treatments.
3
Implement GPUDirect Storage to optimize I/O-bound workflows in your HPC applications.
By allowing direct data access from storage to GPU memory, GPUDirect Storage can significantly reduce latency and improve performance, making it essential for data-intensive tasks.