Cerebras CS-2, the world’s most powerful AI system, is the main compute element of Neocortex.
Unlocking Interactive AI Development for Rapidly Evolving Research
Neocortex is a highly innovative resource that targets the acceleration of AI-powered scientific discovery by greatly shortening the time required for deep learning training and by fostering greater integration of artificial deep learning with scientific workflows. The revolutionary hardware in Neocortex facilitates the development of more efficient algorithms for artificial intelligence and graph analytics.
Neocortex democratizes access to game-changing compute power, otherwise only available to tech giants, for students, postdocs, faculty, and others who require faster turnaround on training to analyze data and integrate AI with simulation. It also inspires the research community to scale their AI-based research and integrate AI advances into their research workflows.
With Neocortex, users can apply more accurate models and larger training data, scale model parallelism to unprecedented levels, and avoid the need for expensive and time-consuming hyperparameter optimization. This innovative AI platform enables the development of new algorithms in machine learning and graph analytics.
For more information about Neocortex, please email firstname.lastname@example.org.
Neocortex: An Innovative Resource for Accelerating AI and HPC Development for Rapidly Evolving Research
All Campus Champions Community Call Presentation
This presentation gives an overview of Neocortex for the Campus Champions community. Neocortex is an NSF-funded AI supercomputer at PSC. Neocortex targets the acceleration of AI-powered scientific discovery by vastly shortening the time required for deep learning training and fostering greater integration of deep learning with scientific workflows.
This webinar presents the Spring 2023 Call for Proposals and gives a system overview of Neocortex.
This webinar gives an overview of the recent Neocortex System upgrade to now feature two Cerebras CS-2 systems, in order to help researchers better understand the benefits of the new servers and changes to the system.
Neocortex features two Cerebras CS-2 systems and an HPE Superdome Flex HPC server robustly provisioned to drive the CS-2 systems simultaneously at maximum speed and support the complementary requirements of AI and HPDA workflows. Neocortex projects can also use Bridges-2, PSC’s flagship computing resource.
Each CS-2 features a Cerebras WSE-2 (Wafer Scale Engine 2), the largest chip ever built.
AI processor: Cerebras Wafer Scale Engine 2
- 850,000 Sparse Linear Algebra Compute (SLAC) Cores
- 2.6 trillion transistors
- 46,225 mm² 40 GB SRAM on-chip memory
- 20 PB/s aggregate memory bandwidth
- 220 Pb/s interconnect bandwidth
System I/O: 1.2 Tb/s (12 × 100 GbE ports)
HPE Superdome Flex
|32 x Intel Xeon Platinum 8280L, 28 cores, 56 threads each, 2.70-4.0 GHz, 38.5 MB cache (more info)
|24 TiB RAM, aggregate memory bandwidth of 4.5TB/s
32 x 6.4TB NVMe SSDs
|Network to CS systems
24 x 100 GbE interfaces
|Interconnect to Bridges-2
16 Mellanox HDR-100 InfiniBand adapters
|Red Hat Enterprise Linux
Neocortex is federated with PSC’s flagship computing system, Bridges-2, which provides users with:
- Access to the Bridges-2 filesystem for management of persistent data
- General purpose computing for complementary data wrangling and preprocessing
- High bandwidth connectivity to other ACCESS sites, campus, labs, and clouds
Acknowledgment in publications
Please use the following citation when acknowledging the use of computational time on Neocortex:
Buitrago P.A., Nystrom N.A. (2021) Neocortex and Bridges-2: A High Performance AI+HPC Ecosystem for Science, Discovery, and Societal Good. In: Nesmachnow S., Castro H., Tchernykh A. (eds) High Performance Computing. CARLA 2020. Communications in Computer and Information Science, vol 1327. Springer, Cham. https://doi.org/10.1007/978-3-030-68035-0_15
Neocortex in action
This material is based upon work supported by the National Science Foundation under Grant Number 2005597. Any opinions, findings, conclusions, or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.