Beta
Subscribe

AI-HPC Convergence

Articles (12)

Genesis Mission logo: a white spiral emblem and wordmark above the U.S. Department of Energy seal, on a dark blue and magenta swirl background.
HPCanalysis

DOE's Genesis Mission Is a Federation Bet Built Over New Supercomputers

The first AmSC funding call centers on integration, even as Genesis expands DOE’s AI compute footprint.

Genesis MissionDepartment of Energy
SCN Staff·
Indigo assay signals arc from a wet-lab microplate into liquid-cooled compute racks, then return as a structured lattice to a microfluidic chip.
AIanalysis

Why Pharma Is Building Its Own AI Supercomputers

Biology models now train on proprietary experiments and feed results back to the lab. That closed loop strengthens the case for owning the machine.

AI InfrastructureNVIDIA
SCN Staff·
A monumental amber-lit lattice of stacked data cells standing directly beside a faceted alpine mountain peak rendered as a compute form, illustrating a national supercomputer and a vast data archive sitting side by side.
HPCnews

Why Switzerland Put a 100-Petabyte NASA Replica Beside It’s Alps Supercomputer

ETH Zurich’s year-long transfer was not a model launch. It was a decision to keep a continuously updated Earth-observation archive close to Swiss compute, and less dependent on upstream US data services.

Research ComputingAI Storage
SCN Staff·
In silico models can now reproduce clinical drug responses - but the runtime, hardware, and cost behind them go undisclosed, leaving half the picture in the dark.
HPCanalysis

ARPA-H Is Funding In Silico Drug Testing. Its Compute Footprint Is Still Opaque.

FDA is expanding non-animal methods while ARPA-H funds human drug-safety models. Public materials omit runtime, hardware, and deployment cost.

Research ComputingAI-HPC Convergence
SCN Staff·
Rendering of LANL's Mission supercomputer: a row of black HPE Cray cabinets with geometric artwork spelling "MISSION," carrying Los Alamos and NNSA logos.
HPCnews

The Mission Supercomputer Predates Genesis. The AI Framing Around It Does Too.

Mission and its AI framing predate Genesis; its public requirements remain centered on stockpile simulation and application performance.

National Labs & GovernmentAI-HPC Convergence
SCN Staff·
Orange-lit supercomputer rack plate with an indigo "AI CLUSTER" placard mounted over it, evoking scientific simulation hardware pitched as AI infrastructure.
HPCanalysis

Why Simulation Supercomputers Are Being Pitched as AI Infrastructure

Simulation still fills the world's research supercomputers. Funding and vendor roadmaps have tilted to AI, and some practitioners feel pressure to pitch their machines that way.

AI-HPC ConvergenceExascale Computing
SCN Staff·
A massive dark speedometer-style gauge emerging from haze, with only the first small segment of its scale glowing bright indigo where the needle points, while the rest of the dial recedes unlit into black.
AIopinion

Time to First Token Is a Real Metric. It Isn't the One That Defines the Era.

Time to first token is the AI industry's favorite candidate for era-defining metric. It is a real whole-stack signal, not a stand-in for business success.

AI InfrastructureInference Economics
Matt Walters·
Illustration of a graphics card dissolving into a CPU package.
HPCanalysis

The FP64 Debate Didn't Produce a Winner. It Produced a Joint Byline.

Dongarra, Hoefler, and Matsuoka now ask whether supercomputing needs GPUs at all, and the CPU-only machine atop the Top500 is their test vehicle.

Exascale ComputingDepartment of Energy
Matt Walters·
Abstract 3D rendering of a reconfigurable compute grid - ordered pale blocks in cool blue light, with cubes breaking formation on the right as glowing orange dataflow streams arc between them, evoking NextSilicon's runtime-reconfigurable Maverick-2 architecture inside the Spectra supercomputer at Sandia National Laboratories.
HPCnews

Spectra Clears Sandia's Supercomputer Acceptance. The Fall Mission-Code Gate Is the Real Test.

NextSilicon's Maverick-2 dataflow accelerator has met Sandia's Vanguard system-acceptance requirements on HPCG, LAMMPS, and SPARTA. The harder question comes this fall, when Sandia decides whether to move Spectra toward more demanding, mission-like ASC supercomputing workloads.

National Labs & GovernmentAI-HPC Convergence
SCN Staff·
Abstract three-quarter rendering of a two-tier network fabric - a row of leaf switches beneath a shorter row of spine switches, connected by hundreds of fine indigo parallel lines.
AInews

MRC Gives Open Ethernet Its First 75,000-GPU Production Proof Point

The 50-author MRC paper gives Ethernet its first multi-vendor, open-spec, production-trace answer to the one argument InfiniBand had left at frontier-training scale.

AI InfrastructureNVIDIA
SCN Staff·
Empty engineering workstation with dual monitors showing GPU kernel code, profiler traces, and pass/fail verification panels, with a small GPU rack visible through a glass partition behind the desk.
Emerginganalysis

The training stack is starting to optimize itself

Anthropic’s 2.9× to 51.9× training-optimization curve signals that AI training infrastructure is becoming machine-optimizable, raising rebound demand and control-plane risks for HPC operators.

AI InfrastructureAgentic AI
SCN Staff·
Interior of a Japanese supercomputing facility with open server rack showing GPU accelerators alongside a Fujitsu CPU module, representing FugakuNEXT's hybrid architecture pairing MONAKA-X CPUs with NVIDIA GPUs over NVLink Fusion.
HPCanalysis

Japan's Next Flagship Machine Abandons the Top500 Chase

FugakuNEXT pairs Fujitsu MONAKA-X CPUs with NVIDIA GPUs, ending Japan's all-Arm sovereign architecture and betting on throughput over benchmarks.

Exascale ComputingTop500
SCN Staff·