# GPU

A specialized processor designed for highly parallel computation, originally for graphics rendering and now used to accelerate computational workloads.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## Vulkan Rendering Is Now The Default on Linux Wayland

DevFeed: [Vulkan Rendering Is Now The Default on Linux Wayland](<https://devfeed.tech/articles/vulkan-rendering-is-now-the-default-on-linux-wayland-146494.md>)

Original publisher: [Read original article](<https://blog.jetbrains.com/platform/2026/11/vulkan-rendering-is-now-the-default-on-linux-wayland/>)

Author: Ekaterina Valeeva

Published: 2026-11-20T16:05:48Z

Content type: release

Language: en

Sources: [JetBrains](<https://devfeed.tech/sources/the-jetbrains-blog.md>)

Topics: [GPU rendering](<https://devfeed.tech/topics/gpu-rendering.md>), [Wayland](<https://devfeed.tech/topics/wayland.md>), [GPU](<https://devfeed.tech/topics/gpu.md>), [Linux](<https://devfeed.tech/topics/linux.md>), [IntelliJ IDEA](<https://devfeed.tech/topics/intellij-idea.md>)

Tags: [2026-3](<https://devfeed.tech/tags/2026-3.md>), [acceleration](<https://devfeed.tech/tags/acceleration.md>), [bug-fix](<https://devfeed.tech/tags/bug-fix.md>), [color](<https://devfeed.tech/tags/color.md>), [compatibility](<https://devfeed.tech/tags/compatibility.md>), [cpu](<https://devfeed.tech/tags/cpu.md>), [crashes](<https://devfeed.tech/tags/crashes.md>), [desktop](<https://devfeed.tech/tags/desktop.md>), [development](<https://devfeed.tech/tags/development.md>), [display](<https://devfeed.tech/tags/display.md>), [distributions](<https://devfeed.tech/tags/distributions.md>), [drivers](<https://devfeed.tech/tags/drivers.md>), [ides](<https://devfeed.tech/tags/ides.md>), [jetbrains](<https://devfeed.tech/tags/jetbrains.md>), [jetbrains-runtime](<https://devfeed.tech/tags/jetbrains-runtime.md>), [linux](<https://devfeed.tech/tags/linux.md>), [rendering](<https://devfeed.tech/tags/rendering.md>), [vulkan](<https://devfeed.tech/tags/vulkan.md>), [wayland](<https://devfeed.tech/tags/wayland.md>), [x11](<https://devfeed.tech/tags/x11.md>), [xwayland](<https://devfeed.tech/tags/xwayland.md>)

### AI overview

JetBrains IDEs enable Vulkan rendering by default in native Linux Wayland sessions starting with 2026.3, shifting more UI drawing from the CPU to the GPU while retaining Wayland behavior; no configuration is required. Performance and power benefits vary by GPU, driver, desktop environment, and workload. Remote Development is not yet supported, with compatibility work planned for a 2026.3 bugfix release. Some configurations may still show missing primitives, blurry text at fractional scaling, presentation problems, or other rendering issues. Users who encounter problems can disable Vulkan with `-Dsun.java2d.vulkan=false` and report them with IDE/build, system and GPU details, logs, reproduction steps, and visual evidence. X11, XWayland, and other operating systems are unchanged.

### Source excerpt

JetBrains IDEs are adopting a new rendering path on Linux. As you may remember, our IDEs started running in native Wayland mode in Linux Wayland sessions with 2026.1. Starting from 2026.3, the Vulkan rendering pipeline is enabled by default in this setup. Nothing changes for X11, XWayland, or other operating systems. Vulkan rendering is not [...]

## Amazon SageMaker AI now detects latency drift, idle GPUs, and missed optimizations on your inference endpoints

DevFeed: [Amazon SageMaker AI now detects latency drift, idle GPUs, and missed optimizations on your inference endpoints](<https://devfeed.tech/articles/amazon-sagemaker-ai-now-detects-latency-drift-idle-gpus-and-missed-optimizations-on-your-inference-endpoints-144471.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/machine-learning/amazon-sagemaker-ai-now-detects-latency-drift-idle-gpus-and-missed-optimizations-on-your-inference-endpoints/>)

Author: Hazim Qudah

Published: 2026-11-19T21:28:18Z

Content type: tutorial

Language: en

Sources: [Artificial Intelligence](<https://devfeed.tech/sources/artificial-intelligence.md>)

Topics: [Amazon SageMaker](<https://devfeed.tech/topics/amazon-sagemaker.md>), [LLM optimization](<https://devfeed.tech/topics/llm-optimization.md>), [gpu monitoring](<https://devfeed.tech/topics/gpu-monitoring.md>), [Benchmark](<https://devfeed.tech/topics/benchmark.md>), [performance-engineering](<https://devfeed.tech/topics/performance-engineering.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [GPU](<https://devfeed.tech/topics/gpu.md>), [Amazon Web Services](<https://devfeed.tech/topics/aws.md>), [vllm](<https://devfeed.tech/topics/vllm.md>)

Tags: [advanced-300](<https://devfeed.tech/tags/advanced-300.md>), [amazon-sagemaker-ai](<https://devfeed.tech/tags/amazon-sagemaker-ai.md>), [amazon-sagemaker-studio](<https://devfeed.tech/tags/amazon-sagemaker-studio.md>), [announcements](<https://devfeed.tech/tags/announcements.md>), [aws-user-notifications](<https://devfeed.tech/tags/aws-user-notifications.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [drift](<https://devfeed.tech/tags/drift.md>), [generative-ai](<https://devfeed.tech/tags/generative-ai.md>), [gpu-utilization](<https://devfeed.tech/tags/gpu-utilization.md>), [gpus](<https://devfeed.tech/tags/gpus.md>), [inference-endpoints](<https://devfeed.tech/tags/inference-endpoints.md>), [latency](<https://devfeed.tech/tags/latency.md>), [optimizations](<https://devfeed.tech/tags/optimizations.md>), [performance](<https://devfeed.tech/tags/performance.md>), [serving](<https://devfeed.tech/tags/serving.md>), [studio](<https://devfeed.tech/tags/studio.md>)

### AI overview

Amazon SageMaker AI monitors eligible generative AI inference endpoints by default when they use vLLM serving containers and GPU instances. Using aggregated latency, GPU utilization, and serving configuration signals, it can flag week-over-week latency drift, mostly idle GPUs, or supported optimizations and updates that are missing. Monitoring does not read request or response content and has no GPU cost. Findings arrive through AWS Health and email; users can open them in SageMaker Studio and benchmark alternatives on separate compute, comparing response time, throughput, and cost per million tokens. No endpoint change occurs until the user chooses to apply a result, and keeping the current configuration is also an option. Benchmarks incur standard SageMaker charges or use an ML reservation. Monitoring can be disabled by setting OPT_OUT_TRACKING on a model container; doing so prevents telemetry collection and notifications for that endpoint.

### Source excerpt

Amazon SageMaker AI now monitors your generative AI inference endpoints by default and tells you when a change could improve performance or lower cost. It tracks latency, GPU utilization, and serving configuration, then sends a finding through AWS Health with a one-click path to benchmark alternatives in SageMaker Studio before you apply any change.

## Intel says its graphics-free D270 GPU beats Nvidia's RTX Pro 6000 by up to 5.9x in benchmarks

DevFeed: [Intel says its graphics-free D270 GPU beats Nvidia's RTX Pro 6000 by up to 5.9x in benchmarks](<https://devfeed.tech/articles/intel-says-its-graphics-free-d270-gpu-beats-nvidia-s-rtx-pro-6000-by-up-to-5-9x-in-benchmarks-143755.md>)

Original publisher: [Read original article](<https://www.techspot.com/news/114624-intel-graphics-free-d270-gpu-beats-nvidia-rtx.html>)

Author: Zo Ahmed

Published: 2026-11-19T14:49:00Z

Content type: news

Language: en

Sources: [TechSpot](<https://devfeed.tech/sources/techspot.md>)

Topics: [GPU](<https://devfeed.tech/topics/gpu.md>), [RTX PRO](<https://devfeed.tech/topics/rtx-pro.md>), [intel](<https://devfeed.tech/topics/intel.md>), [Nvidia](<https://devfeed.tech/topics/nvidia.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-models](<https://devfeed.tech/tags/ai-models.md>), [banks](<https://devfeed.tech/tags/banks.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [blackwell](<https://devfeed.tech/tags/blackwell.md>), [contracts](<https://devfeed.tech/tags/contracts.md>), [crescent-island](<https://devfeed.tech/tags/crescent-island.md>), [data-center](<https://devfeed.tech/tags/data-center.md>), [financial](<https://devfeed.tech/tags/financial.md>), [games](<https://devfeed.tech/tags/games.md>), [geforce-rtx-5090](<https://devfeed.tech/tags/geforce-rtx-5090.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [graphics](<https://devfeed.tech/tags/graphics.md>), [graphics-card](<https://devfeed.tech/tags/graphics-card.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [intel](<https://devfeed.tech/tags/intel.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [rtx](<https://devfeed.tech/tags/rtx.md>), [rtx-pro](<https://devfeed.tech/tags/rtx-pro.md>), [rtx-pro-6000](<https://devfeed.tech/tags/rtx-pro-6000.md>)

### AI overview

Intel reports that its pre-production GPU D270 beat Nvidia's RTX Pro 6000 Blackwell Server Edition by 1.92-5.9x across four bank-focused financial-contract workloads, all using FP64 calculations, while rated at 350W versus Nvidia's 600W. The excerpt attributes much of the advantage to the D270's full-speed FP64 support; Nvidia's card handles FP64 at a fraction of its usual speed. These are Intel's own results, and the cards ran in different systems, with Nvidia's in a larger dual-socket server, so the comparison is not controlled. Intel reported only modest gains in its AI tests, including a 12% throughput edge on one language model. Customers are testing the D270 ahead of a planned 2027 launch; Intel has not announced pricing.

### Source excerpt

In a post, Intel put the D270 up against Nvidia's RTX Pro 6000 Blackwell Server Edition. Instead of games or AI models, Intel went with something a little more niche. It picked four workloads that banks rely on to put a price on financial contracts. Read Entire Article

## Multilevel Graph Partitioning on GPU Supercomputers

DevFeed: [Multilevel Graph Partitioning on GPU Supercomputers](<https://devfeed.tech/articles/multilevel-graph-partitioning-on-gpu-supercomputers-142978.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.24264>)

Author: Michael S. Gilbert, Erik G. Boman, Kamesh Madduri

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Distributed Computing](<https://devfeed.tech/sources/arxiv-distributed-computing.md>)

Topics: [Multi-GPU](<https://devfeed.tech/topics/multi-gpu.md>), [High-Performance Computing](<https://devfeed.tech/topics/high-performance-computing.md>), [Distributed Systems & Parallel Computing](<https://devfeed.tech/topics/distributed-systems-parallel-computing.md>), [GPU](<https://devfeed.tech/topics/gpu.md>), [graph theory](<https://devfeed.tech/topics/graph-theory.md>)

Tags: [cs-dc](<https://devfeed.tech/tags/cs-dc.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [graph](<https://devfeed.tech/tags/graph.md>), [multi-gpu](<https://devfeed.tech/tags/multi-gpu.md>), [optimize](<https://devfeed.tech/tags/optimize.md>), [parallel](<https://devfeed.tech/tags/parallel.md>), [partitioning](<https://devfeed.tech/tags/partitioning.md>), [scalability](<https://devfeed.tech/tags/scalability.md>), [scaling](<https://devfeed.tech/tags/scaling.md>), [supercomputers](<https://devfeed.tech/tags/supercomputers.md>)

### AI overview

The work presents a distributed multi-GPU graph partitioner based on the single-GPU Jet partitioner, aiming to improve scalability while preserving solution quality. It uses CUDA-aware MPI, one-sided communication, and a bulk-synchronous parallel structure. The authors report solution quality comparable to Jet and weak scaling up to 256 GPUs, allowing it to partition instances larger than a single-GPU partitioner can handle.

### Source excerpt

arXiv:2611.24264v1 Announce Type: new Abstract: Graph partitioning is a key enabler in computational science, with applications to sparse linear algebra, load-balanced parallel computation, and graph computations. In recent years, graphics processing units (GPUs) have become the preferred accelerator on the world's top supercomputers. Despite this, most existing distributed graph partitioners do not utilize GPUs. In this work, we develop a new multi-GPU distributed partitioner based on Jet, a prior single-GPU partitioner. We seek to optimize scalability without sacrificing quality. Our partitioner utilizes CUDA-aware MPI, one-sided communication routines, and an overall bulk-synchronous parallel structure. We demonstrate solution quality comparable to Jet. Our experiments show excellent weak scaling up to 256 GPUs, enabling partitioning of instances significantly larger than a single-GPU partitioner can handle.

## Harvesting Expert-Parallel Bubbles for Fine-Tuning

DevFeed: [Harvesting Expert-Parallel Bubbles for Fine-Tuning](<https://devfeed.tech/articles/harvesting-expert-parallel-bubbles-for-fine-tuning-142803.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.24496>)

Author: Oana Balmau, Saurav Banka, Jiaxuan Chen, Anne-Marie Kermarrec, Ramya Narasimha Prabhu, Rafael Pires, Mathis Randl, Martijn de Vos

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Distributed Computing](<https://devfeed.tech/sources/arxiv-distributed-computing.md>)

Topics: [Fine-tuning](<https://devfeed.tech/topics/fine-tuning.md>), [LLM Techniques](<https://devfeed.tech/topics/llm-techniques.md>), [LLM optimization](<https://devfeed.tech/topics/llm-optimization.md>), [Mixture of Experts (MoE)](<https://devfeed.tech/topics/mixture-of-experts-moe.md>), [GPU optimization](<https://devfeed.tech/topics/gpu-optimization.md>), [distributed-training](<https://devfeed.tech/topics/distributed-training.md>), [Multi-GPU](<https://devfeed.tech/topics/multi-gpu.md>), [GPU](<https://devfeed.tech/topics/gpu.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>)

Tags: [a100](<https://devfeed.tech/tags/a100.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [balancing](<https://devfeed.tech/tags/balancing.md>), [bubble](<https://devfeed.tech/tags/bubble.md>), [capacity](<https://devfeed.tech/tags/capacity.md>), [colocation](<https://devfeed.tech/tags/colocation.md>), [compiler](<https://devfeed.tech/tags/compiler.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-dc](<https://devfeed.tech/tags/cs-dc.md>), [cs-os](<https://devfeed.tech/tags/cs-os.md>), [cuda](<https://devfeed.tech/tags/cuda.md>), [engine](<https://devfeed.tech/tags/engine.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [expert-parallel](<https://devfeed.tech/tags/expert-parallel.md>), [expert-parallelism](<https://devfeed.tech/tags/expert-parallelism.md>), [experts](<https://devfeed.tech/tags/experts.md>), [fine-tuning](<https://devfeed.tech/tags/fine-tuning.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [gpus](<https://devfeed.tech/tags/gpus.md>), [inference](<https://devfeed.tech/tags/inference.md>), [language-model](<https://devfeed.tech/tags/language-model.md>), [large-language-model](<https://devfeed.tech/tags/large-language-model.md>)

### AI overview

BubbleTea uses per-GPU idle periods created by uneven expert-parallel token routing to run fine-tuning alongside LLM serving. It predicts each GPU's available bubble at every layer and schedules short fine-tuning chunks that reuse serving weights; compiler passes derive the chunks from the served model without porting its architecture to a separate training engine. On three MoE models served across four A100 GPUs, BubbleTea delivered 1.9-6.2x the fine-tuning throughput of LLMStation and kept p90 inference latency within 10% of serving alone in 50 of 63 workloads, versus 44 for LLMStation. On bursty production traces, it trained 3.3x faster than LLMStation and 14x faster than a stock trainer under CUDA MPS, with p90 latency within 7% of serving alone.

### Source excerpt

arXiv:2611.24496v1 Announce Type: cross Abstract: GPUs are expensive, yet latency-critical large language model (LLM) inference leaves part of them idle. Colocation reclaims this capacity by running best-effort work, such as fine-tuning, alongside serving. Many leading LLMs now use a Mixture-of-Experts (MoE) architecture, which adds a new source of idle time. Under expert parallelism, MoE experts are spread across GPUs, and because the router sends uneven numbers of tokens to each GPU, lightly loaded GPUs wait for the most loaded one. Even after applying load balancing solutions, such bubbles still take 8--13% of prefill time. We present BubbleTea, the first colocation system that leverages expert-parallel bubbles for fine-tuning. Unlike existing colocation systems, which make one scheduling decision for all GPUs per forward pass, BubbleTea predicts each GPU's bubble at every layer from the router's token assignment, and fills it with short chunks of fine-tuning work that reuse the weights already loaded for serving. We also introduce compiler passes that automatically derive these chunks from the served LLM as-is, without porting each model architecture to a separate training engine. Our evaluation on three MoE models, served on four A100 GPUs, shows that fine-tuning throughput with BubbleTea is 1.9--6.2$\times$ faster than with LLMStation, a state-of-the-art colocation system. BubbleTea keeps p90 inference latency within 10% of serving alone in 50 of 63 workloads, compared to 44 for LLMStation. On bursty production traces, BubbleTea trains 3.3$\times$ faster than LLMStation and 14$\times$ faster than a stock trainer colocated under CUDA MPS, with p90 inference latency staying within 7% of serving alone.

## AWS Lambda now supports GPU instances for Event Functions on Lambda Managed Instances

DevFeed: [AWS Lambda now supports GPU instances for Event Functions on Lambda Managed Instances](<https://devfeed.tech/articles/aws-lambda-now-supports-gpu-instances-for-event-functions-on-lambda-managed-instances-141088.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/aws/aws-lambda-now-supports-gpu-instances-for-event-functions-on-lambda-managed-instances/>)

Author: Daniel Abib

Published: 2026-11-18T19:39:07Z

Content type: release

Language: en

Sources: [AWS News Blog](<https://devfeed.tech/sources/aws-news-blog.md>)

Topics: [AWS Lambda](<https://devfeed.tech/topics/aws-lambda.md>), [GPU](<https://devfeed.tech/topics/gpu.md>), [function as a service](<https://devfeed.tech/topics/function-as-a-service.md>), [GPU rendering](<https://devfeed.tech/topics/gpu-rendering.md>), [Amazon EventBridge](<https://devfeed.tech/topics/amazon-eventbridge.md>), [Transformers](<https://devfeed.tech/topics/transformers.md>)

Tags: [all](<https://devfeed.tech/tags/all.md>), [announcements](<https://devfeed.tech/tags/announcements.md>), [auto-scaling](<https://devfeed.tech/tags/auto-scaling.md>), [aws](<https://devfeed.tech/tags/aws.md>), [aws-lambda](<https://devfeed.tech/tags/aws-lambda.md>), [compute](<https://devfeed.tech/tags/compute.md>), [concurrent](<https://devfeed.tech/tags/concurrent.md>), [driver](<https://devfeed.tech/tags/driver.md>), [ec2](<https://devfeed.tech/tags/ec2.md>), [event](<https://devfeed.tech/tags/event.md>), [functions](<https://devfeed.tech/tags/functions.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [gpu-utilization](<https://devfeed.tech/tags/gpu-utilization.md>), [inference](<https://devfeed.tech/tags/inference.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [lambda](<https://devfeed.tech/tags/lambda.md>), [learning](<https://devfeed.tech/tags/learning.md>), [like](<https://devfeed.tech/tags/like.md>), [machine](<https://devfeed.tech/tags/machine.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [managed](<https://devfeed.tech/tags/managed.md>), [management](<https://devfeed.tech/tags/management.md>), [serverless](<https://devfeed.tech/tags/serverless.md>)

### AI overview

AWS Lambda Managed Instances now support GPU-accelerated Event Functions on EC2 G-series instances. Deploy a container image and connect it to a Lambda capacity provider; Lambda handles provisioning, scaling, GPU drivers, routing, and AWS event integration. Functions support concurrent requests per execution environment, and Lambda reports GPU utilization and memory metrics. The feature is aimed at inference and other GPU workloads, with support for images up to 32 GB, up to 32 GB ephemeral storage, and timeouts up to 90 minutes for specified asynchronous and event source mapping invocations. The example deploys a Hugging Face model using Lambda's vLLM image and an OpenAI-compatible request interface; publishing a version provisions environments and warms the model before callers invoke it. Users may need to request G-series quota. Pricing is EC2 instance cost, plus a 9.75% GPU management fee and standard Lambda request charges; Savings Plans and Reserved Instances apply.

### Source excerpt

AWS Lambda now supports GPU instances (G-series) for Event Functions running on Lambda Managed Instances, enabling serverless GPU workloads like machine learning inference, media processing, and rendering. Lambda handles all infrastructure tasks, including instance lifecycle, driver management, and auto scaling, while maximizing GPU utilization across concurrent requests with standard EC2 pricing plus a management fee

## Magflow: Differentiable Simulation of Industrial-Grade 6D Magnetic Levitation

DevFeed: [Magflow: Differentiable Simulation of Industrial-Grade 6D Magnetic Levitation](<https://devfeed.tech/articles/magflow-differentiable-simulation-of-industrial-grade-6d-magnetic-levitation-141836.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23050>)

Author: Philipp Hartmann, Jannick Strangh\\"oner, Klaus Neumann

Published: 2026-11-18T05:00:00Z

Content type: news

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Robotics Simulation](<https://devfeed.tech/topics/robotics-simulation.md>), [Simulation](<https://devfeed.tech/topics/simulation.md>), [Physical AI](<https://devfeed.tech/topics/physical-ai.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Robotics](<https://devfeed.tech/topics/robotics.md>), [GPU](<https://devfeed.tech/topics/gpu.md>)

Tags: [co-design](<https://devfeed.tech/tags/co-design.md>), [control](<https://devfeed.tech/tags/control.md>), [controller](<https://devfeed.tech/tags/controller.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [cs-sy](<https://devfeed.tech/tags/cs-sy.md>), [eess-sy](<https://devfeed.tech/tags/eess-sy.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [industrial](<https://devfeed.tech/tags/industrial.md>), [learning](<https://devfeed.tech/tags/learning.md>), [magnetic](<https://devfeed.tech/tags/magnetic.md>), [neural](<https://devfeed.tech/tags/neural.md>), [physics](<https://devfeed.tech/tags/physics.md>), [real-time](<https://devfeed.tech/tags/real-time.md>), [simulation](<https://devfeed.tech/tags/simulation.md>), [simulator](<https://devfeed.tech/tags/simulator.md>), [topology](<https://devfeed.tech/tags/topology.md>), [zero-shot](<https://devfeed.tech/tags/zero-shot.md>)

### AI overview

Magflow uses a fully differentiable, GPU-parallel simulator to model 6D magnetic levitation directly, avoiding hand-crafted electromagnetic force approximations used in industrial systems. Those approximations help make kilohertz control tractable but constrain the operating workspace and supported mover topologies. The authors report training an industrial-grade neural controller purely in simulation that transfers zero-shot to real hardware, meets 4 kHz hard real-time constraints, exhibits new behaviors, and levitates a mover topology lacking a hand-crafted solution. These results suggest hardware-software co-design is not physically necessary for this control task.

### Source excerpt

arXiv:2611.23050v1 Announce Type: cross Abstract: Magnetic levitation with six-degrees-of-freedom enables frictionless transport and micrometer-scale positioning in principle, but low-level control is inherently challenging due to its complex and unstable dynamics. Industrial systems typically rely on deep hardware-software co-design, where mover and stator geometries are jointly designed to enable a compact, hand-crafted approximation of the electromagnetic force computation, thereby keeping control computationally tractable at kilohertz rates. However, this approach imposes overly conservative constraints on the operating workspace and the range of supported mover topologies. Instead of reducing the underlying physics to a hand-crafted approximation, we compute the physics directly. We introduce Magflow, a fully differentiable, GPU-parallel simulator for 6D magnetic levitation. It is used to train the first neural controller for an industrial-grade 6D magnetic levitation system that is learned purely from simulated physics. We show that the learned neural controller transfers zero-shot from simulation to real hardware while satisfying hard real-time constraints at 4 kHz. The learned controller achieves qualitatively new behaviors and even levitates a mover topology for which no hand-crafted solution exists. These results demonstrate that co-design is not a physical necessity for 6D magnetic levitation control. More broadly, control learning from simulated physics can find solutions beyond the reach of control engineering in demanding industrial domains. Source code and demonstration videos are publicly available at https://sites.google.com/view/magflow.

## Introducing SUNK Auto Repair: Safe, Automatic Recovery for Drained GPU Nodes

DevFeed: [Introducing SUNK Auto Repair: Safe, Automatic Recovery for Drained GPU Nodes](<https://devfeed.tech/articles/introducing-sunk-auto-repair-safe-automatic-recovery-for-drained-gpu-nodes-138560.md>)

Original publisher: [Read original article](<https://wf.coreweave.com/blog/introducing-sunk-auto-repair-safe-automatic-recovery-for-drained-gpu-nodes>)

Author: CoreWeave

Published: 2026-11-17T18:57:30Z

Content type: release

Language: en

Sources: [Coreweave Blog Posts](<https://devfeed.tech/sources/coreweave-blog-posts.md>)

Topics: [GPU](<https://devfeed.tech/topics/gpu.md>), [health check](<https://devfeed.tech/topics/health-check.md>)

Tags: [automated](<https://devfeed.tech/tags/automated.md>), [automatic](<https://devfeed.tech/tags/automatic.md>), [automatically](<https://devfeed.tech/tags/automatically.md>), [blog](<https://devfeed.tech/tags/blog.md>), [capacity](<https://devfeed.tech/tags/capacity.md>), [cluster](<https://devfeed.tech/tags/cluster.md>), [clusters](<https://devfeed.tech/tags/clusters.md>), [collective](<https://devfeed.tech/tags/collective.md>), [compute](<https://devfeed.tech/tags/compute.md>), [coreweave](<https://devfeed.tech/tags/coreweave.md>), [crash](<https://devfeed.tech/tags/crash.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [nodes](<https://devfeed.tech/tags/nodes.md>), [recovery](<https://devfeed.tech/tags/recovery.md>), [repair](<https://devfeed.tech/tags/repair.md>), [safe](<https://devfeed.tech/tags/safe.md>), [safely](<https://devfeed.tech/tags/safely.md>), [service](<https://devfeed.tech/tags/service.md>), [sunk](<https://devfeed.tech/tags/sunk.md>), [waiting](<https://devfeed.tech/tags/waiting.md>)

### AI overview

CoreWeave SUNK Auto Repair restores drained GPU nodes only when a curated policy identifies the drain reason as recoverable and a fresh HPC Verification check confirms the node is healthy. Unknown reasons and failed checks leave nodes drained for human review. CoreWeave manages repair cooldowns and daily attempt caps to limit repeated recovery of flapping nodes; resuming a node makes it available for new jobs without disrupting jobs already running. User-defined policies are on the roadmap, not currently available. The feature is available for SUNK clusters and requires contacting a CoreWeave representative to enable.

### Source excerpt

SUNK Auto Repair re-verifies drained GPU nodes and safely returns recoverable ones to service, so healthy hardware doesn't sit idle waiting for a human.

## RoboKit: Accelerating Parallel Differentiable Optimization for Robot Learning

DevFeed: [RoboKit: Accelerating Parallel Differentiable Optimization for Robot Learning](<https://devfeed.tech/articles/robokit-accelerating-parallel-differentiable-optimization-for-robot-learning-137544.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.20185>)

Author: Jianglong Ye, Xuanbin Peng, Conor Mc Gartoll, Changwei Jing, Lai Wei, Jiali Chen, Xueyan Zou, Xiaolong Wang

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Robot Manipulation](<https://devfeed.tech/topics/robot-manipulation.md>), [GPU](<https://devfeed.tech/topics/gpu.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [High-Performance Computing](<https://devfeed.tech/topics/high-performance-computing.md>)

Tags: [accelerate](<https://devfeed.tech/tags/accelerate.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [high-performance](<https://devfeed.tech/tags/high-performance.md>), [hybrid](<https://devfeed.tech/tags/hybrid.md>), [integration](<https://devfeed.tech/tags/integration.md>), [learning](<https://devfeed.tech/tags/learning.md>), [optimization](<https://devfeed.tech/tags/optimization.md>), [parallel](<https://devfeed.tech/tags/parallel.md>), [performance](<https://devfeed.tech/tags/performance.md>), [physics](<https://devfeed.tech/tags/physics.md>), [rendering](<https://devfeed.tech/tags/rendering.md>), [rl](<https://devfeed.tech/tags/rl.md>), [robot](<https://devfeed.tech/tags/robot.md>), [robotics](<https://devfeed.tech/tags/robotics.md>), [sampling](<https://devfeed.tech/tags/sampling.md>), [simulation](<https://devfeed.tech/tags/simulation.md>), [simulator](<https://devfeed.tech/tags/simulator.md>), [toolkit](<https://devfeed.tech/tags/toolkit.md>)

### AI overview

RoboKit combines GPU-parallel multi-seed solvers with sampling and flexible optimizers, reporting 2-60x faster performance than task-specific baselines across seven robotics tasks. Hybrid and coarse-to-fine representations accelerate collision handling, differentiable rendering, and simulation; the abstract reports optimizing 4,096 grasps in 2.97 seconds. Because its costs are differentiable and simulation is parallel, the toolkit also supports policy distillation and reinforcement learning: distilled policies run 9-240x faster at comparable accuracy, while RL reaches a GPU baseline's scores up to 2.1x sooner on the same simulator. These are abstract-level reported results; the supplied text does not provide task details or evaluation methods.

### Source excerpt

arXiv:2611.20185v1 Announce Type: cross Abstract: We present RoboKit, a high-performance, general-purpose toolkit for parallel differentiable optimization in robotics, built on three principles: (i) Performance and flexibility: GPU-parallel multi-seed solvers pair sampling with versatile optimizers, running 2-60$\times$ faster than task-specific baselines across seven tasks; (ii) Geometry, rendering and physics: hybrid and coarse-to-fine representations accelerate collision, differentiable rendering and simulation, optimizing 4096 grasps in 2.97 s; (iii) Integration with learning: costs are differentiable and simulation is parallel, so RoboKit both distills optimization into policies and trains RL; distilled policies run 9-240$\times$ faster at comparable accuracy, and RL reaches a GPU baseline's scores up to 2.1$\times$ sooner on the same simulator.

## Integrated GPU Core for Mamba-3

DevFeed: [Integrated GPU Core for Mamba-3](<https://devfeed.tech/articles/integrated-gpu-core-for-mamba-3-139044.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.20409>)

Author: Shiva Raja, William Desueza, Ajay Joshi

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Mamba](<https://devfeed.tech/topics/mamba.md>), [GPU](<https://devfeed.tech/topics/gpu.md>), [Tensor Cores](<https://devfeed.tech/topics/tensor-cores.md>)

Tags: [core](<https://devfeed.tech/tags/core.md>), [cs-ar](<https://devfeed.tech/tags/cs-ar.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data-movement](<https://devfeed.tech/tags/data-movement.md>), [decoding](<https://devfeed.tech/tags/decoding.md>), [energy](<https://devfeed.tech/tags/energy.md>), [execution](<https://devfeed.tech/tags/execution.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [hybrid](<https://devfeed.tech/tags/hybrid.md>), [inference](<https://devfeed.tech/tags/inference.md>), [kernel](<https://devfeed.tech/tags/kernel.md>), [long-context](<https://devfeed.tech/tags/long-context.md>), [mamba](<https://devfeed.tech/tags/mamba.md>), [memory-latency](<https://devfeed.tech/tags/memory-latency.md>), [mimo](<https://devfeed.tech/tags/mimo.md>), [model](<https://devfeed.tech/tags/model.md>), [pipelines](<https://devfeed.tech/tags/pipelines.md>), [power](<https://devfeed.tech/tags/power.md>), [power-modeling](<https://devfeed.tech/tags/power-modeling.md>), [prefill](<https://devfeed.tech/tags/prefill.md>), [recurrent](<https://devfeed.tech/tags/recurrent.md>), [rtl](<https://devfeed.tech/tags/rtl.md>)

### AI overview

RescuCore is a proposed GPU-integrated execution unit for Mamba-3 inference, targeting inefficiency from fragmented kernels, data movement, and register pressure. It shares SM sub-partition resources with CUDA and Tensor Cores: dense projections stay on conventional GPU resources, while RescuCore pipelines discretization, normalization, RoPE, MIMO computation, recurrent state updates, and output gating. The authors report that BF16 preserves model accuracy, with an estimated area overhead of 5.9% of an A100 die at state size 32 and MIMO rank 4. In simulation and synthesis-informed evaluation, RescuCore achieves 1.4-9.4x speedup and 1.4-5.0x energy reduction in prefill, and 1.3-2.8x speedup and 1.2-2.6x energy reduction in autoregressive decoding, compared with conventional GPU execution.

### Source excerpt

arXiv:2611.20409v1 Announce Type: cross Abstract: Mamba-3, the latest Mamba state-space model (SSM), enables efficient long-context modeling, matching Mamba-2 perplexity with half the state size and increasing decoding arithmetic intensity. However, its multi-input multi-output (MIMO) computation, exponential-trapezoidal discretization, and rotary positional embeddings (RoPE) introduce dense, element-wise, and sequential operations. Despite kernel fusion, conventional GPUs execute these operations inefficiently. Our characterization of Mamba-3 hybrid models reveals numerous kernel launches, substantial data movement, and increased register pressure that reduces warp occupancy and memory-latency hiding. We propose RescuCore, a specialized GPU execution unit for Mamba-3 inference that retains other hybrid model layers on existing GPU resources. Integrated alongside CUDA Cores and Tensor Cores within each Streaming Multiprocessor (SM) sub-partition, RescuCore shares the register file, shared memory, and warp schedulers. Dense projections remain on conventional GPU resources, while RescuCore pipelines discretization, normalization, RoPE, MIMO computation, recurrent state updates, and output gating to reduce kernel fragmentation and intermediate data movement. We select BF16 after evaluating multiple numerical formats, preserving model accuracy with an estimated area overhead of 5.9% of an A100 die for state size 32 and MIMO rank 4. We evaluate RescuCore using cycle-accurate Accel-Sim simulation, RTL synthesis, and synthesis-informed AccelWattch power modeling. Compared with conventional GPU execution, RescuCore achieves 1.4-9.4x speedup and 1.4-5.0x energy reduction during prefill, and 1.3-2.8x speedup and 1.2-2.6x energy reduction during autoregressive decoding. These results demonstrate the potential of GPU-integrated specialized execution for SSM-based hybrid models.

## Accelerate Molecular Dynamics Simulations with Accurate, Flexible Primitives Using NVIDIA cuPotential

DevFeed: [Accelerate Molecular Dynamics Simulations with Accurate, Flexible Primitives Using NVIDIA cuPotential](<https://devfeed.tech/articles/accelerate-molecular-dynamics-simulations-with-accurate-flexible-primitives-using-nvidia-cupotential-136676.md>)

Original publisher: [Read original article](<https://developer.nvidia.com/blog/accelerate-molecular-dynamics-simulations-with-accurate-flexible-primitives-using-nvidia-cupotential/>)

Author: Yutong Zhao

Published: 2026-11-16T22:30:00Z

Content type: tutorial

Language: en

Sources: [NVIDIA Developer](<https://devfeed.tech/sources/nvidia-developer.md>), [NVIDIA Technical Blog](<https://devfeed.tech/sources/nvidia-technical-blog.md>)

Topics: [Molecular Dynamics](<https://devfeed.tech/topics/molecular-dynamics.md>), [Computational Chemistry / Materials Science](<https://devfeed.tech/topics/computational-chemistry-materials-science.md>), [GPU](<https://devfeed.tech/topics/gpu.md>), [RTX PRO](<https://devfeed.tech/topics/rtx-pro.md>)

Tags: [accelerate](<https://devfeed.tech/tags/accelerate.md>), [api](<https://devfeed.tech/tags/api.md>), [biology](<https://devfeed.tech/tags/biology.md>), [chemistry](<https://devfeed.tech/tags/chemistry.md>), [computational-chemistry-materials-science](<https://devfeed.tech/tags/computational-chemistry-materials-science.md>), [cuda-x](<https://devfeed.tech/tags/cuda-x.md>), [developer-tools-techniques](<https://devfeed.tech/tags/developer-tools-techniques.md>), [materials-science](<https://devfeed.tech/tags/materials-science.md>), [molecular-dynamics](<https://devfeed.tech/tags/molecular-dynamics.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [physics](<https://devfeed.tech/tags/physics.md>), [sc26](<https://devfeed.tech/tags/sc26.md>), [science](<https://devfeed.tech/tags/science.md>), [simulation-modeling-design](<https://devfeed.tech/tags/simulation-modeling-design.md>), [simulations](<https://devfeed.tech/tags/simulations.md>)

### AI overview

NVIDIA cuPotential provides modular GPU primitives for molecular dynamics, targeting long-range electrostatics and dispersion bottlenecks. Its OpenMM plugin replaces the nonbonded force and can also supply a Langevin integrator and Monte Carlo barostat; the excerpt gives Linux build steps using Conda or Micromamba, CUDA 13.1, and OpenMM 8.4.0. Reported NPT benchmarks show over 3x end-to-end speedups on selected RTX PRO 6000 Blackwell systems and gains often around 2x or more on other tested GPUs, for standard non-alchemical systems. The text also reports reduced force error and energy drift under specified comparisons. The plugin targets solvated, roughly uniform-density systems in rectangular periodic boxes; it does not support LJPME, triclinic boxes, or particle and exception parameter offsets. These results and instructions are from the supplied excerpt.

### Source excerpt

Molecular dynamics (MD) is an indispensable tool across physics, chemistry, biology, and materials science. Owing to its arithmetic structure, particularly the...

## NVIDIA Expands Slurm to Bring Quantum Computing Into Supercomputing

DevFeed: [NVIDIA Expands Slurm to Bring Quantum Computing Into Supercomputing](<https://devfeed.tech/articles/nvidia-expands-slurm-to-bring-quantum-computing-into-supercomputing-136626.md>)

Original publisher: [Read original article](<https://blogs.nvidia.com/blog/slurm-quantum-computing-supercomputing/>)

Author: Sam Stanwyck

Published: 2026-11-16T16:00:47Z

Content type: release

Language: en

Sources: [NVIDIA](<https://devfeed.tech/sources/nvidia-blog.md>)

Topics: [Quantum Computing](<https://devfeed.tech/topics/quantum-computing.md>), [Supercomputing](<https://devfeed.tech/topics/supercomputing.md>), [scheduling](<https://devfeed.tech/topics/scheduling.md>), [GPU](<https://devfeed.tech/topics/gpu.md>)

Tags: [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [computing](<https://devfeed.tech/tags/computing.md>), [deployments](<https://devfeed.tech/tags/deployments.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [quantum](<https://devfeed.tech/tags/quantum.md>), [quantum-computing](<https://devfeed.tech/tags/quantum-computing.md>), [slurm](<https://devfeed.tech/tags/slurm.md>), [software](<https://devfeed.tech/tags/software.md>), [supercomputing](<https://devfeed.tech/tags/supercomputing.md>)

### AI overview

Slurm's expansion is intended to reduce the bespoke work required to connect quantum processors (QPUs) to supercomputing centers. QPU builders can configure a single hardware interface in a file and reuse integrations across deployments; centers can add processors to existing scheduling and resource-management workflows. The expansion supports QRMI and QDMI, is being adopted by Alice & Bob, IBM, IQM, Pasqal, Quantinuum and MQSC, and is planned for Slurm 26.11. The excerpt also describes CUDA-Q platform updates, including quantum error-correction prototyping, fault-tolerant primitives, reusable application components and pulse-shaping tools. These are presented as parts of NVIDIA's open quantum-GPU platform; the excerpt does not establish comparative performance across providers or tools.

### Source excerpt

Quantum processor deployments at supercomputing centers have grown nearly sevenfold over the past five years, with nearly 30 supercomputing centers around the world now in the process of deploying over 50 quantum processors. Many of these centers use Slurm to run scientific supercomputing workloads, so expanding its support for quantum processors builds on familiar workflows [...]

## NVIDIA Announces Rubin HPC -- A New GPU to Accelerate Agents for Science, Engineering and Analytics

DevFeed: [NVIDIA Announces Rubin HPC -- A New GPU to Accelerate Agents for Science, Engineering and Analytics](<https://devfeed.tech/articles/nvidia-announces-rubin-hpc-a-new-gpu-to-accelerate-agents-for-science-engineering-and-analytics-136625.md>)

Original publisher: [Read original article](<https://blogs.nvidia.com/blog/rubin-hpc/>)

Author: Dan Ernst

Published: 2026-11-16T16:00:01Z

Content type: release

Language: en

Sources: [NVIDIA](<https://devfeed.tech/sources/nvidia-blog.md>)

Topics: [GPU](<https://devfeed.tech/topics/gpu.md>), [High-Performance Computing](<https://devfeed.tech/topics/high-performance-computing.md>), [CUDA-X](<https://devfeed.tech/topics/cuda-x.md>), [NVLink](<https://devfeed.tech/topics/nvlink.md>), [AI for science](<https://devfeed.tech/topics/ai-for-science.md>), [NVIDIA Vera Rubin](<https://devfeed.tech/topics/nvidia-vera-rubin.md>), [Simulation and Design](<https://devfeed.tech/topics/simulation-and-design.md>)

Tags: [accelerate](<https://devfeed.tech/tags/accelerate.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [analytics](<https://devfeed.tech/tags/analytics.md>), [cuda-x](<https://devfeed.tech/tags/cuda-x.md>), [data](<https://devfeed.tech/tags/data.md>), [data-processing](<https://devfeed.tech/tags/data-processing.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [hpc](<https://devfeed.tech/tags/hpc.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [nvidia-rubin](<https://devfeed.tech/tags/nvidia-rubin.md>), [nvidia-vera-rubin](<https://devfeed.tech/tags/nvidia-vera-rubin.md>), [nvlink](<https://devfeed.tech/tags/nvlink.md>), [rubin](<https://devfeed.tech/tags/rubin.md>), [rubin-hpc](<https://devfeed.tech/tags/rubin-hpc.md>), [sc26](<https://devfeed.tech/tags/sc26.md>), [science](<https://devfeed.tech/tags/science.md>), [simulation](<https://devfeed.tech/tags/simulation.md>), [simulations](<https://devfeed.tech/tags/simulations.md>), [supercomputing](<https://devfeed.tech/tags/supercomputing.md>), [test](<https://devfeed.tech/tags/test.md>), [vera-rubin](<https://devfeed.tech/tags/vera-rubin.md>), [workflows](<https://devfeed.tech/tags/workflows.md>)

### AI overview

NVIDIA says Rubin HPC, a GPU for the Vera Rubin platform announced at SC26, is designed to speed simulations and data processing used in science, engineering and analytics. It combines more than 50,000 CUDA cores with sparse data acceleration to keep cores supplied with data, plus seventh-generation NVLink for GPU-to-GPU connectivity. The design targets recurring workloads such as sparse matrix operations, stencil computations, sorting and reductions, balancing data movement with compute. NVIDIA CUDA-X libraries provide optimized algorithms intended to work with the GPU. Faster simulations could give agents and researchers quicker feedback as they test and refine candidate ideas. Rubin HPC is expected to be available in 2028.

### Source excerpt

Announced today at SC26, NVIDIA Rubin HPC is a new GPU in the NVIDIA Vera Rubin platform designed to accelerate simulation and data processing for science, engineering and analytics workflows. As agents help scientists and engineers explore more candidate materials, molecules and designs, they rely on simulations to test those ideas and guide their next [...]

## Zero-SM NCCL: A Systematic Guide to Designing and Deploying SM-Free GPU Communication

DevFeed: [Zero-SM NCCL: A Systematic Guide to Designing and Deploying SM-Free GPU Communication](<https://devfeed.tech/articles/zero-sm-nccl-a-systematic-guide-to-designing-and-deploying-sm-free-gpu-communication-134894.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.18075>)

Author: Zhenhao He, Siyuan Shen, Ke Wen, Prajwal Singhania, Shiqing Fan, Youngeun Kwon, Sangkug Lym, Vasudevan Rengasamy, Vishnuswaroop Ramesh, Yicheng Qiang, Craig Belusar, Jeff Hammond, Torsten Hoefler

Published: 2026-11-16T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Distributed Computing](<https://devfeed.tech/sources/arxiv-distributed-computing.md>)

Topics: [NCCL](<https://devfeed.tech/topics/nccl.md>), [rdma](<https://devfeed.tech/topics/rdma.md>), [Multi-GPU](<https://devfeed.tech/topics/multi-gpu.md>), [Distributed Systems & Parallel Computing](<https://devfeed.tech/topics/distributed-systems-parallel-computing.md>), [NVLink](<https://devfeed.tech/topics/nvlink.md>), [GPU](<https://devfeed.tech/topics/gpu.md>)

Tags: [accelerate](<https://devfeed.tech/tags/accelerate.md>), [apis](<https://devfeed.tech/tags/apis.md>), [bandwidth](<https://devfeed.tech/tags/bandwidth.md>), [building](<https://devfeed.tech/tags/building.md>), [communication](<https://devfeed.tech/tags/communication.md>), [copy](<https://devfeed.tech/tags/copy.md>), [cpu](<https://devfeed.tech/tags/cpu.md>), [cs-dc](<https://devfeed.tech/tags/cs-dc.md>), [cuda](<https://devfeed.tech/tags/cuda.md>), [data](<https://devfeed.tech/tags/data.md>), [data-movement](<https://devfeed.tech/tags/data-movement.md>), [deep](<https://devfeed.tech/tags/deep.md>), [designing](<https://devfeed.tech/tags/designing.md>), [domain](<https://devfeed.tech/tags/domain.md>), [domains](<https://devfeed.tech/tags/domains.md>), [free](<https://devfeed.tech/tags/free.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [guide](<https://devfeed.tech/tags/guide.md>), [nccl](<https://devfeed.tech/tags/nccl.md>), [sm](<https://devfeed.tech/tags/sm.md>), [zero](<https://devfeed.tech/tags/zero.md>)

### AI overview

Zero-SM NCCL avoids SM kernels by synchronizing with CUDA stream memory operations, using copy engines for transfers within a scale-up domain, and driving network RDMA through a stream-ordered CPU proxy. It supports AllGather and AlltoAll, plus one-sided ncclPutSignal and ncclWaitSignal APIs. On GB200 NVL72 and B200 NVL8, collectives achieved up to 1.36x NCCL's legacy AllGather bandwidth and reduced median overlapped GEMM slowdown from 33% to 4%, with higher per-operation latency. In Megatron-LM, FSDP improved by up to 3.5% and pipeline parallelism by up to 15% with one-sided puts. The gains depend on workload: the approach benefits large transfers in GPU-bound workloads and deep pipelines with large bubbles, but loses on small messages and host-bound workloads.

### Source excerpt

arXiv:2611.18075v1 Announce Type: new Abstract: GPU communication libraries such as NCCL use Streaming Multiprocessors (SMs) for data movement and synchronization, slowing overlapped computation at a cost that grows with interconnect bandwidth. We present Zero-SM NCCL, which communicates within and across scale-up domains without launching any SM kernel. It synchronizes through CUDA stream memory operations, moves data asynchronously with copy engines within a scale-up domain, and drives network RDMA from a stream-ordered CPU proxy. It supports data-movement collectives such as AllGather and AlltoAll, plus one-sided ncclPutSignal and ncclWaitSignal APIs. On GB200 NVL72 and B200 NVL8 systems, Zero-SM collectives reach up to 1.36x higher AllGather bandwidth than NCCL's legacy SM-based kernels and cut the median slowdown of overlapped GEMMs from 33% to 4%, at a higher latency per operation. In Megatron-LM training, they accelerate FSDP by up to 3.5%, and one-sided puts accelerate pipeline parallelism by up to 15%. Zero-SM communication is not a universal win, however. Through an end-to-end study across models and parallelism strategies, we show that it pays off for large transfers in GPU-bound workloads and for deep pipelines with large bubbles, but loses on small messages and host-bound workloads. We distill these findings into guidelines for enabling zero-SM communication in NCCL and building it into other libraries.

## FlashSampling: Fast and Memory-Efficient Exact Sampling

DevFeed: [FlashSampling: Fast and Memory-Efficient Exact Sampling](<https://devfeed.tech/articles/flashsampling-fast-and-memory-efficient-exact-sampling-134969.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2603.15854>)

Author: Tomas Ruiz, Zhen Qin, Yifan Zhang, Xuyang Shen, Yiran Zhong, Mengdi Wang

Published: 2026-11-16T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [GPU optimization](<https://devfeed.tech/topics/gpu-optimization.md>), [GPU](<https://devfeed.tech/topics/gpu.md>), [vllm](<https://devfeed.tech/topics/vllm.md>)

Tags: [bandwidth](<https://devfeed.tech/tags/bandwidth.md>), [communication](<https://devfeed.tech/tags/communication.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [decode](<https://devfeed.tech/tags/decode.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [kernel](<https://devfeed.tech/tags/kernel.md>), [memory](<https://devfeed.tech/tags/memory.md>), [sampling](<https://devfeed.tech/tags/sampling.md>), [vllm](<https://devfeed.tech/tags/vllm.md>)

### AI overview

FlashSampling fuses exact categorical sampling into the language-model head matrix multiplication, avoiding materialization of the logits tensor in HBM. It processes logits in on-chip tiles, adds Gumbel noise, retains each tile's maximizer, then reduces across tiles. For tensor-parallel decoding, it replaces logits all-gather with streaming peer-to-peer writes that overlap communication with computation and memory loads across up to eight GPUs; the abstract reports near-ideal scaling at large batch sizes. The method's exactness relies on argmax decomposition and hierarchical factorization for grouped variants. Kernel speedups are reported across H100, H200, B200, and B300 GPUs; end-to-end vLLM tests reduce time per output token by up to 10% on the tested models.

### Source excerpt

arXiv:2603.15854v4 Announce Type: replace-cross Abstract: Sampling from a categorical distribution is mathematically simple, but in large-vocabulary decoding, it often triggers extra memory traffic and extra kernels after the LM head. We present FlashSampling, an exact sampling primitive that fuses sampling into the LM-head matmul and never materializes the logits tensor in HBM. The method is simple: compute logits tile-by-tile on chip, add Gumbel noise, keep only one maximizer per row and per vocabulary tile, and finish with a small reduction over tiles. In tensor-parallel decoding, FlashSampling replaces the all-gather of logits with streaming peer-to-peer writes: This overlaps GPU-to-GPU communication with computation and HBM loads across up to 8 GPUs, with near-ideal scaling at large batch sizes. Our kernel is exact because argmax decomposes over partitions; grouped variants for online and tensor-parallel settings are exact by hierarchical factorization of the categorical distribution. FlashSampling demonstrates kernel-level speedups on decode workloads across 4 different datacenter GPUs (H100, H200, B200, B300), and in end-to-end vLLM experiments, it reduces time per output token by up to $10%$ on the models we test. These results show that exact sampling, with no approximation, can be integrated into the matmul itself, consolidating the bandwidth-bound sampling step in an efficient epilogue.

## Compiling Specialized GPU-Initiated Networking

DevFeed: [Compiling Specialized GPU-Initiated Networking](<https://devfeed.tech/articles/compiling-specialized-gpu-initiated-networking-134896.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.18145>)

Author: Javid Baydamirli, Ismayil Ismayilov, Kaan Oktay, Didem Unat

Published: 2026-11-16T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Distributed Computing](<https://devfeed.tech/sources/arxiv-distributed-computing.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [High-Performance Computing](<https://devfeed.tech/topics/high-performance-computing.md>), [GPU](<https://devfeed.tech/topics/gpu.md>), [Multi-GPU](<https://devfeed.tech/topics/multi-gpu.md>), [vllm](<https://devfeed.tech/topics/vllm.md>), [NCCL](<https://devfeed.tech/topics/nccl.md>)

Tags: [amd](<https://devfeed.tech/tags/amd.md>), [compiler](<https://devfeed.tech/tags/compiler.md>), [cs-dc](<https://devfeed.tech/tags/cs-dc.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-pl](<https://devfeed.tech/tags/cs-pl.md>), [deepep](<https://devfeed.tech/tags/deepep.md>), [device-code](<https://devfeed.tech/tags/device-code.md>), [faster](<https://devfeed.tech/tags/faster.md>), [gin](<https://devfeed.tech/tags/gin.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [h100](<https://devfeed.tech/tags/h100.md>), [intermediate](<https://devfeed.tech/tags/intermediate.md>), [kernel](<https://devfeed.tech/tags/kernel.md>), [libraries](<https://devfeed.tech/tags/libraries.md>), [lifetimes](<https://devfeed.tech/tags/lifetimes.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [memory](<https://devfeed.tech/tags/memory.md>), [nccl](<https://devfeed.tech/tags/nccl.md>), [networking](<https://devfeed.tech/tags/networking.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [nvshmem](<https://devfeed.tech/tags/nvshmem.md>), [ordering](<https://devfeed.tech/tags/ordering.md>), [qwen3](<https://devfeed.tech/tags/qwen3.md>)

### AI overview

Obaleth integrates GPU-initiated communication into a kernel's intermediate representation so a compiler can co-optimize communication with computation and generate specialized transport code. It schedules completions around buffer lifetimes and optimizes memory ordering, synchronization, and batching across NVIDIA and AMD GPUs. The abstract reports that on H100s it sends small messages up to 3.3x faster; on 256 GPUs its all-to-all runs 1.25x and 2.3x faster than NVSHMEM and NCCL GIN, respectively, with lower memory and register use. In vLLM, serving Qwen3-235B on 16 B200 GPUs, it answers requests up to 1.29x faster than DeepEP.

### Source excerpt

arXiv:2611.18145v1 Announce Type: new Abstract: Machine learning systems specialize computation to the workload and hardware, but communication remains in generic libraries whose device code and transport decisions are not visible to high-level compilers. We present Obaleth, a compiler that makes GPU-initiated communication part of a kernel's intermediate representation, co-optimizes it alongside computation, and lowers it to specialized transport code. Obaleth plans completions around buffer lifetimes and optimizes memory ordering, synchronization, and batching across NVIDIA and AMD GPUs. On H100 GPUs, Obaleth sends small messages up to 3.3$\times$ faster, and on 256 GPUs runs all-to-all 1.25$\times$ and 2.3$\times$ faster than NVSHMEM and NCCL GIN, with much smaller memory and register use. In vLLM, it serves Qwen3-235B on 16 B200 GPUs and answers requests up to 1.29$\times$ faster than with DeepEP.

## CuEVM: Scaling Smart Contract Execution on GPUs

DevFeed: [CuEVM: Scaling Smart Contract Execution on GPUs](<https://devfeed.tech/articles/cuevm-scaling-smart-contract-execution-on-gpus-131526.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.14916>)

Author: Nhut-Minh Ho, Tien Tuan Anh Dinh, Quang Trung Ta, Fredrik Svantes, Xiaokui Xiao, Beng Chin Ooi

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>), [arXiv: Databases](<https://devfeed.tech/sources/arxiv-databases.md>), [arXiv: Distributed Computing](<https://devfeed.tech/sources/arxiv-distributed-computing.md>)

Topics: [smart contract](<https://devfeed.tech/topics/smart-contract.md>), [Ethereum](<https://devfeed.tech/topics/ethereum.md>), [GPU optimization](<https://devfeed.tech/topics/gpu-optimization.md>), [GPU](<https://devfeed.tech/topics/gpu.md>)

Tags: [accelerate](<https://devfeed.tech/tags/accelerate.md>), [blockchain](<https://devfeed.tech/tags/blockchain.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [cs-db](<https://devfeed.tech/tags/cs-db.md>), [cs-dc](<https://devfeed.tech/tags/cs-dc.md>), [data-processing](<https://devfeed.tech/tags/data-processing.md>), [data-structures](<https://devfeed.tech/tags/data-structures.md>), [engine](<https://devfeed.tech/tags/engine.md>), [ethereum](<https://devfeed.tech/tags/ethereum.md>), [evm](<https://devfeed.tech/tags/evm.md>), [execution](<https://devfeed.tech/tags/execution.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [gpus](<https://devfeed.tech/tags/gpus.md>), [parallel](<https://devfeed.tech/tags/parallel.md>), [performance](<https://devfeed.tech/tags/performance.md>), [scaling](<https://devfeed.tech/tags/scaling.md>), [security-testing](<https://devfeed.tech/tags/security-testing.md>), [semantics](<https://devfeed.tech/tags/semantics.md>), [smart-contract](<https://devfeed.tech/tags/smart-contract.md>), [transaction](<https://devfeed.tech/tags/transaction.md>)

### AI overview

CuEVM treats EVM transaction execution as massively parallel GPU data processing, using GPU-aware scheduling and novel data structures to support EVM semantics natively rather than relying on host CPU operations for many tasks. The abstract reports 100% pass rates on Ethereum VMTests and 96.2% on the full execution-test suite. On one commodity GPU it executes over 8 million ERC-20 transfers per second, up to two orders of magnitude faster than existing baselines. Its CuEVM-fuzz tool achieves over a million end-to-end executions per second, reported as up to 23x faster than CPU-based and 4.8x faster than GPU-based baselines. These results are abstract-level claims; the supplied excerpt does not provide benchmark setup or limitations.

### Source excerpt

arXiv:2611.14916v1 Announce Type: new Abstract: Smart contracts on Ethereum blockchain process transactions worth billions of US dollars daily. They run on top of Ethereum Virtual Machine (EVM) which executes transactions sequentially. This execution model creates a performance bottleneck, especially for new blockchain workloads with large numbers of transactions, such as security testing. We present CuEVM, a GPU-native and semantically complete EVM execution engine that treats transaction execution as a massively parallel data processing problem. It employs novel data structures and GPU-aware scheduling to accelerate execution. Unlike prior works that transpile the contract and rely on a host CPU for many operations, CuEVM supports complete EVM semantics natively on the GPU. It passes 100% of the Ethereum VMTests and 96.2% of the full execution-test suite. On a single commodity GPU, CuEVM executes over 8 million ERC-20 transfers per second, outperforming existing baselines by up to two orders of magnitude. We build a smart contract fuzzer, called CuEVM-fuzz, on top of CuEVM. The fuzzer achieves over a million end-to-end executions per second, up to 23x and 4.8x higher than the state-of-the-art CPU-based and GPU-based baselines, respectively.

## PACER: GPU Excursion Management for ML Workloads

DevFeed: [PACER: GPU Excursion Management for ML Workloads](<https://devfeed.tech/articles/pacer-gpu-excursion-management-for-ml-workloads-131555.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.17206>)

Author: Rutwik Jain, Nathaniel Morris, Yiwei Jiang, Shidhartha Das, Gary Kao, Indrani Paul, Shivaram Venkataraman, Srilatha Manne, Matthew D. Sinclair

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Distributed Computing](<https://devfeed.tech/sources/arxiv-distributed-computing.md>)

Topics: [GPU](<https://devfeed.tech/topics/gpu.md>), [Electrical engineering and computer science (EECS)](<https://devfeed.tech/topics/electrical-engineering-and-computer-science-eecs.md>)

Tags: [cs-dc](<https://devfeed.tech/tags/cs-dc.md>), [firmware](<https://devfeed.tech/tags/firmware.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [kernels](<https://devfeed.tech/tags/kernels.md>), [large-language-models-llms](<https://devfeed.tech/tags/large-language-models-llms.md>), [ml](<https://devfeed.tech/tags/ml.md>), [performance](<https://devfeed.tech/tags/performance.md>), [power-management](<https://devfeed.tech/tags/power-management.md>), [process-variation](<https://devfeed.tech/tags/process-variation.md>), [techniques](<https://devfeed.tech/tags/techniques.md>)

### AI overview

PACER targets intra-GPU performance variability in ML workloads, where short kernels and rapidly changing compute demands can trigger power excursions above rated limits. Sustained peaks lead to throttling, while coarse-grained power management responds poorly to these transient demands. The authors report characterizing the magnitude and frequency of these excursions and using hardware and firmware mechanisms to make power management more consistent. Across popular workloads, their approach improves performance by 3-12% over state-of-the-art GPU power management and comes within 3% of an oracular solution. The supplied text is an abstract and does not detail the mechanisms or evaluation setup.

### Source excerpt

arXiv:2611.17206v1 Announce Type: new Abstract: Variability is an inherent characteristic of modern GPU systems, arising from a combination of process variation, dynamic environmental conditions, and power management decisions. Prior work has characterized inter-GPU variability and found consistent performance differences across GPUs. However, we observe that modern workloads, particularly large language models (LLMs), exhibit significant intra-GPU variability. Their short-running kernels interact poorly with the coarse granularity of dynamic power management mechanisms, leading to unpredictable performance. A key source of this intra-GPU transient variability is power excursions, also referred to as electrical design power peaks (EDP$_{p}$), where power temporarily exceeds the GPU's rated power limits. While power management mechanisms are designed to ensure reliable operation despite the presence of EDP$_{p}$ sustained power peaks cause throttling that detrimentally impacts overall application performance. To the best of our knowledge, our work is the first to characterize the magnitude and frequency of EDP$_{p}$, showing that frequent, significant power overshoots are especially common in ML workloads, which combine high power draw, short kernels, and rapidly changing compute demands. To address this, we design and leverage hardware and firmware mechanisms to provide consistent power management behavior that reduces intra-GPU variability due to EDP$_{p}$ and improves overall performance. We evaluate our techniques across a range of popular workloads. Our approach improves performance by 3-12% over state-of-the-art GPU power management and comes within 3% of an oracular solution.

## KernelAudit: Contract-Aware Auditing of Validators for LLM-Generated GPU Kernels

DevFeed: [KernelAudit: Contract-Aware Auditing of Validators for LLM-Generated GPU Kernels](<https://devfeed.tech/articles/kernelaudit-contract-aware-auditing-of-validators-for-llm-generated-gpu-kernels-131714.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.15506>)

Author: Ziyuan Li, Zihe Song, Xiaofang Qi, Wei Yang, Tao Xie

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Software Engineering](<https://devfeed.tech/sources/arxiv-software-engineering.md>)

Topics: [audit](<https://devfeed.tech/topics/audit.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [GPU](<https://devfeed.tech/topics/gpu.md>)

Tags: [auditing](<https://devfeed.tech/tags/auditing.md>), [contracts](<https://devfeed.tech/tags/contracts.md>), [cs-se](<https://devfeed.tech/tags/cs-se.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [kernels](<https://devfeed.tech/tags/kernels.md>), [llm](<https://devfeed.tech/tags/llm.md>), [measurement](<https://devfeed.tech/tags/measurement.md>), [semantic](<https://devfeed.tech/tags/semantic.md>), [traces](<https://devfeed.tech/tags/traces.md>), [validation](<https://devfeed.tech/tags/validation.md>)

### AI overview

KernelAudit independently checks whether validators judge LLM-generated GPU kernels against task contracts. Its audit oracle labels candidates PASS or FAIL across semantic correctness, implementation integrity, and measurement integrity: reference-calibrated stress tests check outputs on valid inputs, execution traces identify library use, and timing checks test whether reported runtime includes the full computation. The authors evaluated six validators against 716 candidates from seven corpora, including constructed controls, and report nontrivial false-reject rates among the validators and high agreement between oracle labels and human judgments.

### Source excerpt

arXiv:2611.15506v1 Announce Type: new Abstract: The effectiveness of generating GPU kernels driven by a large language model (LLM) is highly dependent on the quality of the used validator (used to accept or reject a generated kernel candidate against the contracts of the given kernel-generation task), whose results inform reward assignment, candidate selection, and speedup claims during kernel generation. However, an existing validator suffers from various limitations, e.g., wrongly accepting a candidate due to missing its forbidden library use or wrongly rejecting a candidate due to determining its permitted library use as forbidden (constraints on library uses are stated in the task contracts). To assess the quality of a validator against the given task contracts, we propose an independent-audit approach for the validator by (1) construction of audit oracle: labeling a candidate as PASS or FAIL based on three separate dimensions in our proposed contract model (i.e., semantic correctness, implementation integrity, and measurement integrity) and (2) auditing of the validator against the audit oracle: comparing the labels of the candidate against the validation results produced by the validator. For semantic correctness, reference-calibrated stress tests check output correctness on valid inputs for a candidate. For implementation integrity, reference and candidate execution traces are used to determine specific library uses. For measurement integrity, timing checks assess whether the reported runtime reflects the full time needed to complete the candidate's computation. We apply our approach to assess the quality of six validators and 716 candidates from seven corpora, including constructed controls. The evaluation results show that our approach helps assess the existing six validators to have nontrivial false-reject rates. There is high-degree agreement between the candidate labels produced by our audit oracle and human judgments.

## A Parallel Greedy Algorithm for Maximal b-Matching in Hypergraphs

DevFeed: [A Parallel Greedy Algorithm for Maximal b-Matching in Hypergraphs](<https://devfeed.tech/articles/a-parallel-greedy-algorithm-for-maximal-b-matching-in-hypergraphs-131534.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.15615>)

Author: Saeed Bakhshan, Daniel Grosu, Henrik Reinst\\"adtler, Christian Schulz

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Distributed Computing](<https://devfeed.tech/sources/arxiv-distributed-computing.md>)

Topics: [Combinatorial optimization](<https://devfeed.tech/topics/combinatorial-optimization.md>), [Algorithms](<https://devfeed.tech/topics/algorithms.md>), [High-Performance Computing](<https://devfeed.tech/topics/high-performance-computing.md>), [GPU](<https://devfeed.tech/topics/gpu.md>), [GPU optimization](<https://devfeed.tech/topics/gpu-optimization.md>)

Tags: [algorithm](<https://devfeed.tech/tags/algorithm.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [cs-dc](<https://devfeed.tech/tags/cs-dc.md>), [cs-ds](<https://devfeed.tech/tags/cs-ds.md>), [cuda](<https://devfeed.tech/tags/cuda.md>), [greedy](<https://devfeed.tech/tags/greedy.md>), [greedy-algorithm](<https://devfeed.tech/tags/greedy-algorithm.md>), [matching](<https://devfeed.tech/tags/matching.md>), [parallel](<https://devfeed.tech/tags/parallel.md>)

### AI overview

The proposed \parallelBatchBMatch algorithm computes the lexicographically first maximal b-matching, matching sequential greedy output for a fixed edge order. With uniformly random priorities, it has O(m') expected work, where m' is the number of vertex-edge incidences. It uses at most m rounds for any priority order, and has O(log² m') depth per round with high probability, yielding O(m log² m') total depth with high probability. A CUDA implementation was evaluated on real-world hypergraphs at capacities 1, 3, and 5; for b = 5, it was up to 19.07 times faster than the existing state-of-the-art greedy algorithm.

### Source excerpt

arXiv:2611.15615v1 Announce Type: cross Abstract: We present \parallelBatchBMatch, a parallel batch-vote algorithm for maximal $b$-matching in hypergraphs. The algorithm is deterministic for a given ordering of edges and produces the same output as the sequential greedy algorithm, i.e., the lexicographically first maximal $b$-matching. Under a uniformly random priority order it performs $O(m')$ expected work in the binary fork--join model, where $m'$ is the number of pins, the total number of vertex--edge incidences. The algorithm performs at most $m$ rounds for every priority order, where $m$ is the number of edges. Every round of the algorithm has $O(\log^2 m')$ depth with high probability, and thus the total depth of the algorithm is $O(m \log^2 m')$ with high probability. We implement the parallel algorithm in CUDA and evaluate it on real-world hypergraphs with capacities $b \in \{1, 3, 5\}$. The parallel algorithm computes the same matchings as the greedy algorithms and for the case of $b = 5$ it is up to $19.07$ times faster than the existing state-of-the art greedy algorithm.

## RAIN: Region-Aware Inversion Network for Semantic Watermark Extraction

DevFeed: [RAIN: Region-Aware Inversion Network for Semantic Watermark Extraction](<https://devfeed.tech/articles/rain-region-aware-inversion-network-for-semantic-watermark-extraction-131967.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2609.14856>)

Author: Zilai Li

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [diffusion-transformers](<https://devfeed.tech/topics/diffusion-transformers.md>), [GPU](<https://devfeed.tech/topics/gpu.md>)

Tags: [conditional](<https://devfeed.tech/tags/conditional.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [diffusion](<https://devfeed.tech/tags/diffusion.md>), [diffusion-models](<https://devfeed.tech/tags/diffusion-models.md>), [flow-matching](<https://devfeed.tech/tags/flow-matching.md>), [gaussian](<https://devfeed.tech/tags/gaussian.md>), [github](<https://devfeed.tech/tags/github.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [lightweight](<https://devfeed.tech/tags/lightweight.md>), [noise](<https://devfeed.tech/tags/noise.md>), [one](<https://devfeed.tech/tags/one.md>), [parallel](<https://devfeed.tech/tags/parallel.md>), [pipeline](<https://devfeed.tech/tags/pipeline.md>), [prompt](<https://devfeed.tech/tags/prompt.md>), [semantic](<https://devfeed.tech/tags/semantic.md>), [watermark](<https://devfeed.tech/tags/watermark.md>)

### AI overview

RAIN is a prompt-free, one-step method for extracting Gaussian-Shading semantic watermarks without iterative diffusion inversion or repeated evaluation of a diffusion-scale U-Net. It estimates a useful noise statistic near the high-SNR image endpoint, where the method argues that Gaussian Shading only requires the recovered latent to fall within the correct watermark decision region. RAIN splits endpoint recovery into an image-like anchor and a noise-oriented residual to make greater use of GPU parallelism. The abstract reports lower noise-extraction cost than OSI and FARI, but provides no quantitative comparison in the supplied excerpt.

### Source excerpt

arXiv:2609.14856v3 Announce Type: replace-cross Abstract: Semantic watermarks for diffusion models embed ownership information into the generative process while preserving perceptual quality, but Gaussian-Shading extraction conventionally requires multi-step diffusion inversion to recover the initial noise. Recent one-step methods show that this cost can be reduced substantially. We study this problem through extended flow matching and conditional regression. The key observation is that, near the high-SNR image endpoint, recovering a useful noise statistic given by the first-step output of the extended flow matching in the high-SNR regime is much simpler than reconstructing the full inverse trajectory, and Gaussian Shading only requires the recovered latent to remain in the correct watermark decision region. Based on this observation, we propose a lightweight, prompt-free extractor that decomposes endpoint recovery into an image-like anchor and a noise-oriented residual, which increases the capability of the model to utilize GPU parallel computation. The resulting method avoids iterative inversion and repeated evaluation of a diffusion-scale U-Net, providing an efficient one-step extraction pipeline with a concise theoretical interpretation. The computational cost of extracting noise is lower than that of both OSI and FARI. The github repo is there: https://github.com/TheLovesOfLadyPurple/RAIN-lightweight-NN-for-one-step-semantic-watermark-extraction

## Scaling VLA Serving Under Robotic Execution Constraints

DevFeed: [Scaling VLA Serving Under Robotic Execution Constraints](<https://devfeed.tech/articles/scaling-vla-serving-under-robotic-execution-constraints-131528.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.14969>)

Author: Yue Zhang, Zicong Ye, Lufang Chen, Dance Yang, Mingyang Li, Zhenwei Duan, Michael Zhao, Ziye Deng, Starrick Liu, Ruizhi Chengze, YongHua Chen, SongAo Han, Yu Sun, Ran Yu, Xiaofan Li, Ruyi Gan, Hao Wan

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Distributed Computing](<https://devfeed.tech/sources/arxiv-distributed-computing.md>)

Topics: [Robot Manipulation](<https://devfeed.tech/topics/robot-manipulation.md>), [Physical AI](<https://devfeed.tech/topics/physical-ai.md>), [LLM optimization](<https://devfeed.tech/topics/llm-optimization.md>), [GPU](<https://devfeed.tech/topics/gpu.md>), [real-time](<https://devfeed.tech/topics/real-time.md>)

Tags: [capacity](<https://devfeed.tech/tags/capacity.md>), [cost](<https://devfeed.tech/tags/cost.md>), [critical](<https://devfeed.tech/tags/critical.md>), [cs-dc](<https://devfeed.tech/tags/cs-dc.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [deployments](<https://devfeed.tech/tags/deployments.md>), [efficiency](<https://devfeed.tech/tags/efficiency.md>), [execution](<https://devfeed.tech/tags/execution.md>), [fleet](<https://devfeed.tech/tags/fleet.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [gpu-efficiency](<https://devfeed.tech/tags/gpu-efficiency.md>), [inference](<https://devfeed.tech/tags/inference.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [language](<https://devfeed.tech/tags/language.md>), [large](<https://devfeed.tech/tags/large.md>), [manipulation](<https://devfeed.tech/tags/manipulation.md>), [models](<https://devfeed.tech/tags/models.md>), [parallelism](<https://devfeed.tech/tags/parallelism.md>), [production](<https://devfeed.tech/tags/production.md>), [robotic](<https://devfeed.tech/tags/robotic.md>), [serving](<https://devfeed.tech/tags/serving.md>), [vla](<https://devfeed.tech/tags/vla.md>)

### AI overview

XInfer is a production-deployed serving system intended to lower inference infrastructure costs for fleets of VLA-powered robots. It coordinates heterogeneous VLA computation with changing response requirements during robotic execution, using parallelism while managing interference to improve GPU efficiency and protect time-critical feedback. Evaluated on three VLA models with production-derived workloads under physical-execution constraints, it achieved 1.26-3.75x the effective robot capacity of state-of-the-art systems. The supplied excerpt does not specify the compared systems or evaluation details.

### Source excerpt

arXiv:2611.14969v1 Announce Type: cross Abstract: Vision-language-action (VLA) models have emerged as a promising paradigm for general-purpose robotic manipulation. In our production setting, hundreds of VLA-powered robots are already operating in real-world environments, creating an immediate need for efficient fleet serving. We present XInfer, a production-deployed VLA serving system designed to reduce the inference infrastructure cost of large-scale robotic deployments. XInfer jointly coordinates heterogeneous VLA computation with the evolving response requirements of robotic execution. By exploiting computational parallelism while managing interference, XInfer improves GPU efficiency while protecting time-critical robotic feedback. We evaluate XInfer across three VLA models using production-derived workloads, achieving 1.26-3.75 times the effective robot capacity (ERC) of state-of-the-art systems under physical-execution constraints. Our implementation is publicly available at https://github.com/X-Square-Robot/xinfer.

## Read Once, Answer Many: Parallel Prompting for Shared-Context LLM Inference

DevFeed: [Read Once, Answer Many: Parallel Prompting for Shared-Context LLM Inference](<https://devfeed.tech/articles/read-once-answer-many-parallel-prompting-for-shared-context-llm-inference-131531.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.15130>)

Author: Zekun Zhao, Jeffrey Flanigan

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Distributed Computing](<https://devfeed.tech/sources/arxiv-distributed-computing.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [LLM optimization](<https://devfeed.tech/topics/llm-optimization.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [GPU optimization](<https://devfeed.tech/topics/gpu-optimization.md>), [GPU](<https://devfeed.tech/topics/gpu.md>), [Cache](<https://devfeed.tech/topics/cache.md>)

Tags: [8k](<https://devfeed.tech/tags/8k.md>), [a100](<https://devfeed.tech/tags/a100.md>), [attention](<https://devfeed.tech/tags/attention.md>), [backend](<https://devfeed.tech/tags/backend.md>), [bandwidth](<https://devfeed.tech/tags/bandwidth.md>), [cache](<https://devfeed.tech/tags/cache.md>), [caching](<https://devfeed.tech/tags/caching.md>), [context](<https://devfeed.tech/tags/context.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [cs-dc](<https://devfeed.tech/tags/cs-dc.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cuda](<https://devfeed.tech/tags/cuda.md>), [data-labeling](<https://devfeed.tech/tags/data-labeling.md>), [decode](<https://devfeed.tech/tags/decode.md>), [decoding](<https://devfeed.tech/tags/decoding.md>), [engine](<https://devfeed.tech/tags/engine.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [flashinfer](<https://devfeed.tech/tags/flashinfer.md>), [inference](<https://devfeed.tech/tags/inference.md>), [llm](<https://devfeed.tech/tags/llm.md>), [llm-inference](<https://devfeed.tech/tags/llm-inference.md>), [parallel](<https://devfeed.tech/tags/parallel.md>), [prompting](<https://devfeed.tech/tags/prompting.md>)

### AI overview

Parallel Prompting reduces repeated KV-cache reads when answering many independent questions about one long document: it decodes the questions as separate streams in a single request, reading the shared document once per step. A stream-ID mask inside the attention kernel keeps each answer conditioned on its own question and the document, without a dense mask tensor. The SGLang implementation adds about 80 lines across three existing files plus a small engine package; fixed-shape decode steps can be captured as a CUDA graph. With CodeLlama-7B, 8k-16k-token documents, and 64-256 questions, it ran 3.7x faster than stock SGLang with FlashInfer on an A100 and 5.9-6.0x faster on an RTX A6000 across two SGLang releases. It was faster in all 144 reported runs; the evidence is limited to these tested configurations.

### Source excerpt

arXiv:2611.15130v1 Announce Type: cross Abstract: LLM workloads sometimes ask many independent questions about the same long document, as in document QA, data labeling and LLM-based evaluation. Prefix caching removes the redundant prefill shared by these queries, but each decode step still reads the document's key-value (KV) cache once per question, so decoding is bound by memory bandwidth and scales with the number of questions. We present Parallel Prompting, which decodes P questions as P independent streams of a single request all at once, fixing this issue. In our method, every step reads the shared document once for all P streams, and a stream-id attention mask, evaluated inside the attention kernel, keeps each answer conditioned only on the document and its own question. We implement Parallel Prompting in SGLang by adding about 80 lines to three existing files and a small engine package; the mask needs no dense tensor, and every decode step has a fixed shape that is captured as a CUDA graph. With CodeLlama-7B, 8k-16k-token documents and 64-256 questions per document, Parallel Prompting is 3.7x faster than stock SGLang with its FlashInfer backend on an A100 and 5.9-6.0x faster on an RTX A6000, on two SGLang releases, including the most recent release. Timed on identical inputs, it is faster than stock SGLang in all 144 runs (2 GPUs x 2 releases x 6 workloads x 6 input sets). Code: https://github.com/jlab-nlp/parallel-prompting

## Parallel Computing Systems for the Optimal Selection of Process Equipment

DevFeed: [Parallel Computing Systems for the Optimal Selection of Process Equipment](<https://devfeed.tech/articles/parallel-computing-systems-for-the-optimal-selection-of-process-equipment-131545.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.16290>)

Author: Andrey B. Borisenko

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Distributed Computing](<https://devfeed.tech/sources/arxiv-distributed-computing.md>)

Topics: [High-Performance Computing](<https://devfeed.tech/topics/high-performance-computing.md>), [Combinatorial optimization](<https://devfeed.tech/topics/combinatorial-optimization.md>), [GPU](<https://devfeed.tech/topics/gpu.md>), [GPU optimization](<https://devfeed.tech/topics/gpu-optimization.md>)

Tags: [algorithm](<https://devfeed.tech/tags/algorithm.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [batch](<https://devfeed.tech/tags/batch.md>), [branch](<https://devfeed.tech/tags/branch.md>), [capital](<https://devfeed.tech/tags/capital.md>), [chemical-engineering](<https://devfeed.tech/tags/chemical-engineering.md>), [clusters](<https://devfeed.tech/tags/clusters.md>), [computing](<https://devfeed.tech/tags/computing.md>), [core](<https://devfeed.tech/tags/core.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cs-dc](<https://devfeed.tech/tags/cs-dc.md>), [cuda](<https://devfeed.tech/tags/cuda.md>), [develop](<https://devfeed.tech/tags/develop.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [equipment](<https://devfeed.tech/tags/equipment.md>), [parallel](<https://devfeed.tech/tags/parallel.md>), [parallel-computing](<https://devfeed.tech/tags/parallel-computing.md>), [process](<https://devfeed.tech/tags/process.md>), [selection](<https://devfeed.tech/tags/selection.md>), [systems](<https://devfeed.tech/tags/systems.md>)

### AI overview

The monograph tackles NP-hard equipment selection for multiproduct batch plants, minimizing capital cost while meeting operability and production-horizon requirements. It develops sequential branch-and-bound and parallel implementations for OpenMP shared-memory systems, MPI clusters, and CUDA GPUs, with tree-traversal and hardware-specific optimizations. To reduce runtime, a GPU hybrid metaheuristic uses Ant Colony Optimization to find a feasible solution and Simulated Annealing to improve it; several statistical approaches for tuning ant-colony parameters are also studied. The abstract reports that this method found solutions within 0.01% of the branch-and-bound optimum in tens of seconds, compared with tens of hours. These results are reported for experiments across multicore processors, clusters, and multiple GPU generations.

### Source excerpt

arXiv:2611.16290v1 Announce Type: new Abstract: This monograph addresses the optimal selection of equipment for chemical-engineering systems of multiproduct batch plants, a large-scale NP-hard discrete optimization problem in which, for each processing stage, a unit of standard size must be chosen so that the total capital cost of the equipment is minimized while the system remains operable and produces the required products within a given time horizon. We present the mathematical model of system operation and analyze its computational properties. We then develop a sequential branch-and-bound algorithm and its parallel versions for shared-memory systems (OpenMP), distributed-memory clusters (MPI, master-worker), and graphics processing units (CUDA), including recursive and iterative tree traversal, shared-memory optimization, and branch-divergence reduction. To overcome the exponential growth of the run time, we propose a hybrid metaheuristic in which Ant Colony Optimization finds a feasible initial solution and Simulated Annealing improves it, and we implement it on GPUs with thousands of independent threads. Finally, we propose random, constant, and multi-constant approaches to tuning the parameters of the ant colony algorithm based on statistical analysis of many parallel runs, and we study the program on GPUs of three generations and on nodes with several GPUs. Computational experiments on multi-core processors, clusters, and GPUs show that the parallel hybrid algorithm finds solutions within 0.01% of the branch-and-bound optimum in tens of seconds instead of tens of hours.

[Next page](<https://devfeed.tech/topics/gpu.md?cursor=WyIyMDI2LTExLTEzVDA1OjAwOjAwKzAwOjAwIiwgIjJlNGU3MDQ0LTg2NmMtNDk2ZS04Y2Q5LTQyZTYwY2Q2NmMwMCJd>)