Purdue 5.2 Petaflop Supercomputer

Purdue University has a new 5.2 petaflop AI Anvil supercomputer. It will enable new discoveries in fields ranging from hypersonics and deep learning to archeological reconstruction and bone biomechanics. Anvil was funded by a $22.5 million grant from the National Science Foundation. Early users are working on projects in key areas of research such as:…
Purdue 5.2 Petaflop Supercomputer


Purdue University has a new 5.2 petaflop AI Anvil supercomputer. It will enable new discoveries in fields ranging from hypersonics and deep learning to archeological reconstruction and bone biomechanics. Anvil was funded by a $22.5 million grant from the National Science Foundation.

Early users are working on projects in key areas of research such as:

Hypersonics: Addressing the scientific challenges of predicting and controlling the extreme aerothermal loads that occur in flight at high Mach numbers.


Deep learning theory: Combining traditional graph algorithms and neural network methods to learn latent vertex representation in a graph. Applications include recommender systems, object relationship reasoning robotic, drug design, and logical rule extraction.


Archaeological reconstruction: A novel procedural (or generative) approach to creating detailed 3D models of existing or previously existing urban spaces for use in various exploratory, what-if design scenarios.


Bone biomechanics: Research combining structural geometry and material science. The research will study enhanced bone hydration as novel pathway to reduce bone fragility.

It has 1,000 nodes with two 64-core third-generation AMD Epyc processors each, and will eventually deliver over 1 billion CPU core hours to XSEDE each year, with a peak performance of 5.3 petaflops. Anvil’s nodes are interconnected with 100 Gbps Mellanox HDR InfiniBand. The supercomputer ecosystem also includes 32 large memory nodes, each with 1 TB of RAM, and 16 nodes each with four NVIDIA A100 Tensor Core GPUs providing 1.5 PF of single-precision performance to support machine learning and artificial intelligence applications.

Anvil uses a diverse set of storage technologies. It has 10-plus PB parallel filesystem and over 3 PB of flash disk. Novel workflows will benefit from block and object storage systems also supported by Anvil.

SOURCES -Purdue University


Written by Brian Wang, Nextbigfuture.com

Subscribe on Google News

Read More

Total
0
Shares
Leave a Reply

Your email address will not be published.

Related Posts
Theresa Baumgartner Presents: Vandals (Excerpt)
Read More

Theresa Baumgartner Presents: Vandals (Excerpt)

Theresa Baumgartner presents an excerpt from I do not wish to be known as a Vandal, a collaboration with Grammy-award winning musician Sam Slater. I do not wish to be known as a Vandal is a 40-minute, audiovisual installation work from musician Sam Slater, performer Lukas Malkowski and this month’s Fact Resident, artist Theresa Baumgartner.…
Microsoft AI Co-pilot for the Web
Read More

Microsoft AI Co-pilot for the Web

Microsoft is bringing into the browser. The AI will summarize any pdf document (including scientific papers) and integrate data from current web and use it to create content. Merging in the AI has created the largest jump in search relevance in two decades. AI will supercharge search and boost content creation. Brian Wang is a…
Finland and Sweden Moving Ahead Towards NATO Membership
Read More

Finland and Sweden Moving Ahead Towards NATO Membership

Finland and Sweden are moving ahead on separate tracks to quickly move ahead to join NATO. Finland had indicated it would prefer a solution that would see the two Nordic unaligned states “jump together” into NATO. Sweden is looking at a range of enhanced security-related options, including deepening Nordic defense cooperation and urging the European…
Smithsonian Has the Original Star Trek Enterprise Model
Read More

Smithsonian Has the Original Star Trek Enterprise Model

The Smithsonian has the original Star Trek Enterprise model. It has been restored by special effects experts from ILM (Industrial Light and Magic). It shows paint details that were not visible from the resolution of old TVs. Adam Savage went and saw it. It was made of wood. The Smithsonian did a great deal of…