News from the AI & ML world
DeeperML - #supercomputer
|
@singularityhub.com
//
Sandia National Laboratories has officially deployed a SpiNNaker2-based neuromorphic supercomputer, a brain-inspired system designed by SpiNNcloud. This supercomputer, capable of simulating between 150 and 180 million neurons, represents a significant advancement in energy-efficient computing, with potential applications in national security. The SpiNNaker2 system utilizes a highly parallel architecture, featuring 48 SpiNNaker2 chips per server board, each containing 152 Arm-based cores and specialized accelerators, enabling efficient, event-driven computation.
This brain-inspired approach offers an alternative to traditional GPU-based systems, which while powerful, consume significant energy. The SpiNNaker2 mimics the brain's communication method using bursts of electricity, activating neurons only when they have important information to transmit. This contrasts with conventional neural networks, where neurons activate every time data is processed, potentially leading to substantial energy savings. Sandia research scientist Craig Vineyard noted that the new system delivers both impressive performance and substantial efficiency gains to Sandia’s neuromorphic capabilities. SpiNNcloud’s SpiNNaker2 is among the top five largest brain-inspired computing platforms globally. Sandia is exploring the new system’s capabilities for nuclear deterrence and various other energy-efficient AI applications. The AI Hypercomputer combines purpose-built hardware, optimized software and frameworks, and flexible consumption models to create a strong AI infrastructure.
Share:
References :
Classification:
Dashveenjit Kaur@TechHQ
//
Dell Technologies has secured a contract with the U.S. Department of Energy to construct the next-generation NERSC-10 supercomputer, a project powered by NVIDIA's Vera Rubin architecture. This new system, dubbed "Doudna" after Nobel laureate Jennifer Doudna, a pioneer in CRISPR gene-editing technology, is poised to be a major federal investment in scientific computing infrastructure. Energy Secretary Chris Wright announced the contract during a visit to Lawrence Berkeley National Laboratory, emphasizing that the deployment in 2026 is crucial for maintaining American technological leadership amidst increasing global competition in AI and quantum computing.
The "Doudna" supercomputer, also known as NERSC-10, aims to significantly accelerate scientific research across multiple domains, including fusion energy, astronomy, and life sciences. Designed to serve 11,000 researchers, it represents an integration of artificial intelligence, quantum workflows, and real-time data streaming from experimental facilities. Unlike traditional supercomputers, Doudna’s architecture emphasizes coherent memory access between CPUs and GPUs, facilitating efficient data sharing between heterogeneous processors which is essential for modern AI-accelerated scientific workflows. The Doudna system is expected to deliver a 10x increase in scientific output compared to its predecessor, Perlmutter, while only consuming 2-3x the power, translating to a 3-5x improvement in performance per watt. Nick Wright, advanced technologies group lead and Doudna chief architect at NERSC, stated, "We’re not just building a faster computer, we’re building a system that helps researchers think bigger and discover sooner." NVIDIA's Vera Rubin platform introduces hardware-level optimizations specifically designed for the convergence of simulation, machine learning, and quantum algorithm development, marking a significant advancement in cutting-edge research capabilities.
Share:
References :
Classification:
@quantumcomputingreport.com
//
NVIDIA is significantly advancing quantum and AI research through strategic collaborations and cutting-edge technology. The company is partnering with Japan’s National Institute of Advanced Industrial Science and Technology (AIST) to launch ABCI-Q, a new supercomputing system focused on hybrid quantum-classical computing. This research-focused system is designed to support large-scale operations, utilizing the power of 2,020 NVIDIA H100 GPUs interconnected with NVIDIA’s Quantum-2 InfiniBand platform. The ABCI-Q system will be hosted at the newly established Global Research and Development Center for Business by Quantum-AI Technology (G-QuAT).
The ABCI-Q infrastructure integrates CUDA-Q, an open-source platform that orchestrates large-scale quantum-classical computing, enabling researchers to simulate and accelerate quantum applications. This hybrid setup combines GPU-based simulation with physical quantum processors from vendors such as Fujitsu (superconducting qubits), QuEra (neutral atom qubits), and OptQC (photonic qubits). This modular architecture will allow for testing quantum error correction, developing algorithms, and refining co-design strategies, which are all critical for future quantum systems. The system serves as a testbed for evaluating quantum-GPU workflows and advancing practical use cases across multiple hardware modalities. NVIDIA is also expanding its presence in Taiwan, powering a new supercomputer at the National Center for High-Performance Computing (NCHC). This supercomputer is projected to deliver eight times the AI performance compared to the center's previous Taiwania 2 system. The new supercomputer will feature NVIDIA HGX H200 systems with over 1,700 GPUs, two NVIDIA GB200 NVL72 rack-scale systems, and an NVIDIA HGX B300 system built on the NVIDIA Blackwell Ultra platform, all interconnected by NVIDIA Quantum InfiniBand networking. This enhanced infrastructure is expected to significantly boost research in AI development, climate science, and quantum computing, fostering technological autonomy and global AI leadership for Taiwan.
Share:
References :
Classification:
|
BenchmarksBlogsResearch Tools |