Altair at SC Asia 2021
March 2-4, 2021
Altair is the market leader in HPC workload management and optimization solutions. Top500 systems and small to mid-sized computing environments alike rely on Altair to keep infrastructure running smoothly.
We’ve made monumental updates and additions to our enterprise computing solutions in 2021 that are helping organizations drive innovation forward even in a year full of unprecedented challenges.
At SCA21 the Altair team is standing by and thrilled to share what we’ve been working hard on all year! Come back to this page during the virtual event and sign to access exclusive SCA content.
Acquiring New Technology:
Altair announced our acquisition of workload management leader, Univa, and I/O profiling company - Ellexus, putting more advanced, diverse HPC optimization solutions at our customers’ fingertips and bringing some of the foremost HPC talent in the world together under one Altair roof.
Read the blog post by Altair CEO James Scapa about the recent acquisitions.
2020 saw some of the most significant updates to the world’s leading HPC workload manager, Altair® PBS Professional®, to date. Advancements in cloud bursting, high-throughput scheduling, budgeting, and workload simulation are empowering organizations across the globe to achieve unprecedented scale, flexibility, and ROI across computing infrastructure.
Enter the Contest
Challenge “An Ideal System Architecture for the Convergence of HPC and A”
The adoption of machine learning (ML) has skyrocketed in the past few years, creating new computing demands and challenges. HPC simulations produce large amounts of data, which ML can rely on and benefit from. HPC systems should be designed to serve both scientific and ML applications. Traditionally, HPC has larger compute, storage, and fast network requirements. ML codes can run on both CPUs and accelerators (GPUs), but they are proven to perform faster and more cost-effectively on GPUs.
Containers are playing a major role in delivering ML software, but HPC scientific applications have yet to adapt to a container-based software delivery model. As a result, there are mixed hardware and software requirements when HPC and ML need to coexist. Many sites have proven that it is possible to have a converged infrastructure, but everyone has their own challenges.
This competition is about putting together your ideas for an ideal unified HPC and AI system.
- Create an A4-sized document briefly describing the architecture of an ideal unified HPC and AI system.
- Upload the completed pdf or Word document using the form on this page.
Altair Presentation Lineup
Cloud Bursting: State of the Art GUIs for Seamless Scaling
PBS Professional 2020.1
American Axle: Driving Efficiency with High-performance Computing
Hierarchical Scheduling for High-throughput Computing Workloads
Rapid Chip Design in the Cloud
Kyoto University's HPC Resources Power Japan's Research Community
Fireside Chat: How PING Golf Stays Ahead of the Innovation Curve with Altair HPC Appliance Powered by the Latest Intel Xeon Processor
Cloud Roundtable: Is Cloud Officially Inevitable?
Second Generation AMD EPYC Processors Speed Up Altair Radioss™ Crash and Multiphysics Simulations
The Next Level of AI in Science Powered by NVIDIA GPUs: How to Get Started with the DGX A100 in Less Than One Week
GPU Acceleration of Altair AcuSolve
Fireside Chat with Dell and Intel
Altair Access - Job submission and remote visualization via desktop, mobile and web
Altair Control - Manage and optimize HPC resources
Altair Control - Analyze and simulate HPC workloads
Altair Mistral - Live system telemetry and I/O monitoring
Altair Breeze - Detailed analysis and I/O profiling
Altair Allocator & Altair Monitor - Sharing licenses across multiple sites and cloud locations
Altair Accelerator & Accelerator Plus - Workload scalability
Altair FlowTracer - Change propagation and flow branching