# gpudirect rdma

GPUDirect RDMA is an NVIDIA technology and API that enables direct data exchange between NVIDIA GPU memory and peer PCIe devices, including RDMA-capable network adapters, without copying through host memory.

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

## When One GPU Isn't Enough: Serving GLM-5.3 Across Two NVIDIA DGX Stations with vLLM

DevFeed: [When One GPU Isn't Enough: Serving GLM-5.3 Across Two NVIDIA DGX Stations with vLLM](<https://devfeed.tech/articles/when-one-gpu-isn-t-enough-serving-glm-5-3-across-two-nvidia-dgx-stations-with-vllm-132447.md>)

Original publisher: [Read original article](<https://developer.nvidia.com/blog/when-one-gpu-isnt-enough-serving-glm-5-3-across-two-nvidia-dgx-stations-with-vllm/>)

Author: Fatima Fall

Published: 2026-11-13T17:00:00Z

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: [vllm](<https://devfeed.tech/topics/vllm.md>), [DGX Station](<https://devfeed.tech/topics/dgx-station.md>), [gpudirect rdma](<https://devfeed.tech/topics/gpudirect-rdma.md>), [NCCL](<https://devfeed.tech/topics/nccl.md>), [Benchmark](<https://devfeed.tech/topics/benchmark.md>), [Multi-GPU](<https://devfeed.tech/topics/multi-gpu.md>), [quantization](<https://devfeed.tech/topics/quantization.md>), [NVFP4](<https://devfeed.tech/topics/nvfp4.md>)

Tags: [agentic-ai-generative-ai](<https://devfeed.tech/tags/agentic-ai-generative-ai.md>), [agents](<https://devfeed.tech/tags/agents.md>), [api](<https://devfeed.tech/tags/api.md>), [benchmarking](<https://devfeed.tech/tags/benchmarking.md>), [dgx](<https://devfeed.tech/tags/dgx.md>), [dgx-station](<https://devfeed.tech/tags/dgx-station.md>), [edge-computing](<https://devfeed.tech/tags/edge-computing.md>), [glm-5-3](<https://devfeed.tech/tags/glm-5-3.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [gpu-memory](<https://devfeed.tech/tags/gpu-memory.md>), [inference-performance](<https://devfeed.tech/tags/inference-performance.md>), [llm-benchmarking](<https://devfeed.tech/tags/llm-benchmarking.md>), [llms](<https://devfeed.tech/tags/llms.md>), [local-ai](<https://devfeed.tech/tags/local-ai.md>), [networking-communications](<https://devfeed.tech/tags/networking-communications.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [serving](<https://devfeed.tech/tags/serving.md>), [vllm](<https://devfeed.tech/tags/vllm.md>)

### AI overview

The tutorial runs the 753-billion-parameter GLM-5.3 NVFP4 model across two GB300 DGX Stations because its roughly 464 GB checkpoint exceeds one GPU's 252 GB of memory. vLLM tensor parallelism splits weights and computation across both GPUs; ConnectX-8 RDMA carries rank-to-rank traffic, and Station A exposes a single OpenAI-compatible API. The workflow covers validating the link, downloading and syncing identical checkpoint copies, launching one vLLM rank per station, checking RDMA and API health, and measuring load with AIPerf on 80 SPEED-Bench coding conversations. Prerequisites include substantial disk space on each station, SSH access, Hugging Face access to the gated dataset, and acceptance of the model and dataset terms. Reported throughput peaks at 607.8 tokens/s at concurrency 16, then drops to 385.0 tokens/s at concurrency 32, where mean latency is 453.5 seconds; mean per-user throughput declines as concurrency rises. The excerpt notes that longer prompts or more concurrent users need more KV cache, and suggests reducing context length or request concurrency if memory runs out.

### Source excerpt

Running a frontier-scale model locally means keeping all its weights in GPU memory; the largest models no longer fit on one machine. Two NVIDIA DGX Stations...

## Modal becomes a gVisor maintainer

DevFeed: [Modal becomes a gVisor maintainer](<https://devfeed.tech/articles/modal-becomes-a-gvisor-maintainer-129819.md>)

Original publisher: [Read original article](<https://modal.com/blog/modal-gvisor-maintainer>)

Author: Luis Capelo

Published: 2026-11-11T12:00:00Z

Content type: article

Language: en

Sources: [Modal](<https://devfeed.tech/sources/modal.md>)

Topics: [virtualization](<https://devfeed.tech/topics/virtualization.md>), [gpudirect rdma](<https://devfeed.tech/topics/gpudirect-rdma.md>), [rdma](<https://devfeed.tech/topics/rdma.md>), [cgroups](<https://devfeed.tech/topics/cgroups.md>), [GPU rendering](<https://devfeed.tech/topics/gpu-rendering.md>), [Cloud Gaming](<https://devfeed.tech/topics/cloud-gaming.md>)

Tags: [agentic](<https://devfeed.tech/tags/agentic.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-coding-agents](<https://devfeed.tech/tags/ai-coding-agents.md>), [amazon-s3](<https://devfeed.tech/tags/amazon-s3.md>), [aws](<https://devfeed.tech/tags/aws.md>), [bare-metal](<https://devfeed.tech/tags/bare-metal.md>), [blackwell](<https://devfeed.tech/tags/blackwell.md>), [blackwell-gpu](<https://devfeed.tech/tags/blackwell-gpu.md>), [clickhouse](<https://devfeed.tech/tags/clickhouse.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [cloud-gaming](<https://devfeed.tech/tags/cloud-gaming.md>), [cloud-native-computing-foundation](<https://devfeed.tech/tags/cloud-native-computing-foundation.md>), [clusters](<https://devfeed.tech/tags/clusters.md>), [containers](<https://devfeed.tech/tags/containers.md>), [gvisor](<https://devfeed.tech/tags/gvisor.md>), [modal](<https://devfeed.tech/tags/modal.md>)

### AI overview

Modal is joining gVisor's maintainers as the project moves to CNCF governance, building on its use of gVisor for more than 40 million containers a day. gVisor places a Go-based application kernel (the Sentry) between container workloads and the host kernel, then layers Linux isolation controls around it; unlike hardware virtualization, it can run on Linux machines without nested virtualization. Modal cites multi-tenant isolation, GPU access through nvproxy, roughly 25 ms startup with 12 MB memory overhead, checkpoint and restore, and portability across clouds and bare-metal systems as reasons for choosing it. Its upstream work includes GPU driver and workload support, an RDMA proxy for multi-node training, snapshots, and filesystem and shared-memory performance improvements. The article reports about 99% of native bus bandwidth in NCCL all-reduce benchmarks for RDMA and describes multi-GPU snapshots and GPUDirect Storage as ongoing work. Modal's vision is broader accelerator support, portable state, and faster I/O under neutral, multi-vendor governance.

### Source excerpt

Modal runs upwards of 40 million gVisor containers every day. Here's why we bet on it, what we've built into it, and where we want to take it next.

## Release Sooner, Expose Less: Offloading MoE Dispatch in NCCL-EP under Dual-Batch Overlap

DevFeed: [Release Sooner, Expose Less: Offloading MoE Dispatch in NCCL-EP under Dual-Batch Overlap](<https://devfeed.tech/articles/release-sooner-expose-less-offloading-moe-dispatch-in-nccl-ep-under-dual-batch-overlap-126744.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.12362>)

Author: Hsin-Chun Yin, Khaled Hamidouche, Ian Ziemba, Arjun Kashyap, Pak Markthub, Subhadeep Bhattacharya, Ke Wen, Assaf Ravid, Manjunath Gorentla Venkata, Gil Bloch

Published: 2026-11-10T05:00:00Z

Content type: article

Language: en

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

Topics: [Mixture of Experts (MoE)](<https://devfeed.tech/topics/mixture-of-experts-moe.md>), [gpudirect rdma](<https://devfeed.tech/topics/gpudirect-rdma.md>), [rdma](<https://devfeed.tech/topics/rdma.md>), [vllm](<https://devfeed.tech/topics/vllm.md>)

Tags: [abstract](<https://devfeed.tech/tags/abstract.md>), [accessible](<https://devfeed.tech/tags/accessible.md>), [after](<https://devfeed.tech/tags/after.md>), [all](<https://devfeed.tech/tags/all.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [batch](<https://devfeed.tech/tags/batch.md>), [but](<https://devfeed.tech/tags/but.md>), [can](<https://devfeed.tech/tags/can.md>), [communication](<https://devfeed.tech/tags/communication.md>), [configuration](<https://devfeed.tech/tags/configuration.md>), [cs-dc](<https://devfeed.tech/tags/cs-dc.md>), [deepep](<https://devfeed.tech/tags/deepep.md>), [deferred](<https://devfeed.tech/tags/deferred.md>), [dispatch](<https://devfeed.tech/tags/dispatch.md>), [less](<https://devfeed.tech/tags/less.md>), [moe](<https://devfeed.tech/tags/moe.md>), [nccl](<https://devfeed.tech/tags/nccl.md>), [offloading](<https://devfeed.tech/tags/offloading.md>), [overlap](<https://devfeed.tech/tags/overlap.md>), [release](<https://devfeed.tech/tags/release.md>), [under](<https://devfeed.tech/tags/under.md>)

### AI overview

Dual-batch overlap hides MoE network progress only after dispatch send releases the GPU, so send time remains exposed. On a GB300 NVL72, three NCCL-EP configurations reduce dispatch-send time versus default GDAKI across most evaluated operating points: GPI has the GPU build GPU-Friendly Descriptors (GFDs) for NIC-accessible transport; GPI-MC represents token fan-out in one GFD; and GDAKI Free Flow advances sends on a host thread. Offload lengthens the in-flight interval, and 32-512 microseconds of independent work between send and receive can hide deferred progress, leaving less exposed time than default GDAKI, depending on configuration. In disaggregated vLLM serving with DBO, median TPOT falls 15-25% versus default GDAKI and 16-42% versus DeepEP v1. These are results for the reported system and evaluated sweep.

### Source excerpt

arXiv:2611.12362v1 Announce Type: new Abstract: Dual-batch overlap (DBO) hides mixture-of-experts (MoE) communication behind another microbatch's computation, but only after dispatch send releases the GPU. The time dispatch send holds the GPU therefore stays exposed, while network progress after release can be hidden. We evaluate three research configurations built on NCCL-EP that target this exposed communication time: GPI, in which the GPU constructs GPU-Friendly Descriptors (GFDs) and submits them to NIC-accessible memory for transport; GPI-MC, which describes MoE token fan-out in one GFD; and GDAKI Free Flow, which moves send progression to a host thread. On a GB300 NVL72 system, all three approaches reduce dispatch-send time relative to default GDAKI over most operating points in the evaluated parameter sweep. An experiment that inserts a GPU delay between dispatch send and receive shows that offload increases the in-flight interval. Depending on the configuration, 32 to 512 microseconds of computation or other independent work between send and receive hides the deferred progress, leaving less exposed time than default GDAKI. In disaggregated vLLM serving with DBO, these approaches lower median TPOT by 15-25% relative to NCCL-EP default GDAKI and by 16-42% relative to DeepEP v1.

## Convergent Science Advances Engineering Simulation on CoreWeave

DevFeed: [Convergent Science Advances Engineering Simulation on CoreWeave](<https://devfeed.tech/articles/convergent-science-advances-engineering-simulation-on-coreweave-114287.md>)

Original publisher: [Read original article](<https://wf.coreweave.com/blog/convergent-science-advances-engineering-simulation-on-coreweave>)

Author: CoreWeave

Published: 2026-11-03T15:33:33Z

Content type: article

Language: en

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

Topics: [Computational Fluid Dynamics (CFD)](<https://devfeed.tech/topics/cfd.md>), [Simulation](<https://devfeed.tech/topics/simulation.md>), [gpudirect rdma](<https://devfeed.tech/topics/gpudirect-rdma.md>), [NVLink](<https://devfeed.tech/topics/nvlink.md>)

Tags: [blog](<https://devfeed.tech/tags/blog.md>), [cfd](<https://devfeed.tech/tags/cfd.md>), [coreweave](<https://devfeed.tech/tags/coreweave.md>), [dns](<https://devfeed.tech/tags/dns.md>), [engineering-simulation](<https://devfeed.tech/tags/engineering-simulation.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [scale](<https://devfeed.tech/tags/scale.md>), [simulations](<https://devfeed.tech/tags/simulations.md>), [size](<https://devfeed.tech/tags/size.md>)

### AI overview

Convergent Science ran a 536-million-cell CONVERGE CFD direct numerical simulation to completion in 3.62 hours on two NVIDIA HGX B200 instances (16 Blackwell GPUs). In its scaling study, it also compared NVIDIA RTX PRO 6000 Blackwell Server Edition and HGX B200 using a validated turbulent-flow case. The excerpt distinguishes weak scaling (larger meshes as GPU resources grow) from strong scaling (same mesh completed faster), and explains that memory determines whether a job fits while communication affects speed as GPUs are added. RTX PRO systems target jobs within one server, using PCIe between GPUs; HGX B200 supports larger, multi-node cases with more memory per GPU, NVLink within nodes, and 400 Gb/s InfiniBand between them. Results depend on workload and topology. The tested DNS case was lean; production simulations with additional physics may require more memory at the same cell count, so its figures may underestimate those needs.

### Source excerpt

How far can GPU CFD scale? Convergent Science ran 536 million cells on two NVIDIA HGX B200 instances in 3.62 hours. Here is how to size and scale your own simulations.

## How DRA can improve the performance of distributed LLM inference workloads with RDMA and SR-IOV

DevFeed: [How DRA can improve the performance of distributed LLM inference workloads with RDMA and SR-IOV](<https://devfeed.tech/articles/how-dra-can-improve-the-performance-of-distributed-llm-inference-workloads-with-rdma-and-sr-iov-105451.md>)

Original publisher: [Read original article](<https://developers.redhat.com/articles/2026/10/28/how-dra-can-improve-performance-of-distributed-llm-inference-workloads-with-rdma-and-sr-iov>)

Author: David Whyte-Gray, William Zhao, Raj Joshi, Sebastian Scheinkman

Published: 2026-10-28T13:16:22Z

Content type: article

Language: en

Sources: [Red Hat](<https://devfeed.tech/sources/red-hat.md>), [Red Hat Developer](<https://devfeed.tech/sources/red-hat-developer.md>)

Topics: [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [gpudirect rdma](<https://devfeed.tech/topics/gpudirect-rdma.md>), [rdma](<https://devfeed.tech/topics/rdma.md>), [performance-engineering](<https://devfeed.tech/topics/performance-engineering.md>), [vllm](<https://devfeed.tech/topics/vllm.md>), [LLM optimization](<https://devfeed.tech/topics/llm-optimization.md>), [High-Performance Computing](<https://devfeed.tech/topics/high-performance-computing.md>)

Tags: [ai-workloads](<https://devfeed.tech/tags/ai-workloads.md>), [alignment](<https://devfeed.tech/tags/alignment.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [bandwidth](<https://devfeed.tech/tags/bandwidth.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [dra](<https://devfeed.tech/tags/dra.md>), [llm-inference](<https://devfeed.tech/tags/llm-inference.md>), [performance](<https://devfeed.tech/tags/performance.md>), [rdma](<https://devfeed.tech/tags/rdma.md>), [vllm](<https://devfeed.tech/tags/vllm.md>)

### AI overview

GPU and RDMA NIC placement can sharply affect distributed inference performance, and Kubernetes DRA can require the two devices to share a PCIe root. Traditional device plugins allocate GPU and SR-IOV VF independently, without expressing that topology relationship. The described setup uses the NVIDIA GPU DRA driver and `dra-driver-sriov`; both publish `resource.kubernetes.io/pcieRoot`, and a ResourceClaim applies `matchAttribute` so a claim remains pending if no aligned pair is available. SR-IOV policies must select and advertise VFs, and the cluster needs DRA enabled and the relevant drivers and operator installed. In tests on an A100 platform, aligned placement delivered 194.6 Gbit/s RDMA bandwidth versus 127.1 Gbit/s for same-NUMA but different-root placement and 110.0 Gbit/s across NUMA nodes. For the tested llm-d Qwen prefill/decode workload, alignment raised KV-transfer throughput from 9,612.6 to 20,709.6 MB/s and cut mean latency from 122.32 to 56.78 ms. Results depend on the tested model, GPU, NIC, and server topology.

### Source excerpt

Modern LLM inference is rapidly shifting from a pure compute challenge to a distributed network bottleneck. To maintain high throughput and handle the massive context windows demanded by agentic workloads, production engines like vLLM and distributed LLM inference frameworks like llm-d are moving toward disaggregated, highly parallelized architectures. The post How DRA can improve the performance of distributed LLM inference workloads with RDMA and SR-IOV appeared first on Red Hat Developer.

## Best Enterprise GPUs 2026: PCIe AI Accelerators, Lab-Tested and Ranked

DevFeed: [Best Enterprise GPUs 2026: PCIe AI Accelerators, Lab-Tested and Ranked](<https://devfeed.tech/articles/best-enterprise-gpus-2026-pcie-ai-accelerators-lab-tested-and-ranked-87055.md>)

Original publisher: [Read original article](<https://www.storagereview.com/best/enterprise-gpus>)

Author: Brian Beeler

Published: 2026-10-14T21:19:15Z

Content type: comparison

Language: en

Sources: [StorageReview.com](<https://devfeed.tech/sources/storagereview-com.md>)

Topics: [gpudirect rdma](<https://devfeed.tech/topics/gpudirect-rdma.md>), [vllm](<https://devfeed.tech/topics/vllm.md>), [NVIDIA DGX](<https://devfeed.tech/topics/nvidia-dgx.md>), [Multi-GPU](<https://devfeed.tech/topics/multi-gpu.md>)

Tags: [accelerators](<https://devfeed.tech/tags/accelerators.md>), [ai](<https://devfeed.tech/tags/ai.md>), [comparison](<https://devfeed.tech/tags/comparison.md>), [enterprise](<https://devfeed.tech/tags/enterprise.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [gpus](<https://devfeed.tech/tags/gpus.md>), [vllm](<https://devfeed.tech/tags/vllm.md>)

### AI overview

The article compares PCIe GPUs for enterprise inference using vLLM tests across multiple models and concurrency levels. It identifies the H200 NVL as the fastest measured card, while explaining tradeoffs involving memory capacity, power limits, chassis fit, and workloads. The RTX PRO 6000, RTX PRO 4500, L40S, and L4 are also evaluated, with the H100 NVL included for comparison.

### Source excerpt

Published October 14, 2026. Five PCIe accelerators ranked on one vLLM container and one model set, with the H100 NVL in Also Tested; the H200 NVL carries a power note explained below. Every accelerator ranked on this page has been through the StorageReview lab as a single PCIe card. The enterprise GPU market is split The post Best Enterprise GPUs 2026: PCIe AI Accelerators, Lab-Tested and Ranked appeared first on StorageReview.com.

## Push-Based KV Transfer to Improve TTFT and TPOT for Disaggregated Inference in vLLM

DevFeed: [Push-Based KV Transfer to Improve TTFT and TPOT for Disaggregated Inference in vLLM](<https://devfeed.tech/articles/push-based-kv-transfer-to-improve-ttft-and-tpot-for-disaggregated-inference-in-vllm-87122.md>)

Original publisher: [Read original article](<https://vllm.ai/blog/2026-10-14-nixl-push-connector>)

Author: Sunita Nadampalli (AWS), Nicolò Lucchesi (Mistral)

Published: 2026-10-14T00:00:00Z

Content type: article

Language: en

Sources: [vLLM Blog](<https://devfeed.tech/sources/vllm-blog.md>)

Topics: [vllm](<https://devfeed.tech/topics/vllm.md>), [rdma](<https://devfeed.tech/topics/rdma.md>), [gpudirect rdma](<https://devfeed.tech/topics/gpudirect-rdma.md>), [AWS Trainium](<https://devfeed.tech/topics/aws-trainium.md>), [AWS Inferentia](<https://devfeed.tech/topics/aws-inferentia.md>), [quantization](<https://devfeed.tech/topics/quantization.md>)

Tags: [ai-engineering](<https://devfeed.tech/tags/ai-engineering.md>), [amazon-ec2](<https://devfeed.tech/tags/amazon-ec2.md>), [aws-ec2](<https://devfeed.tech/tags/aws-ec2.md>), [aws-trainium](<https://devfeed.tech/tags/aws-trainium.md>), [disaggregation](<https://devfeed.tech/tags/disaggregation.md>), [inference](<https://devfeed.tech/tags/inference.md>), [kv-cache](<https://devfeed.tech/tags/kv-cache.md>), [llm-inference](<https://devfeed.tech/tags/llm-inference.md>), [model-serving](<https://devfeed.tech/tags/model-serving.md>), [nixl](<https://devfeed.tech/tags/nixl.md>), [open-source-ai](<https://devfeed.tech/tags/open-source-ai.md>), [performance](<https://devfeed.tech/tags/performance.md>), [rdma](<https://devfeed.tech/tags/rdma.md>), [vllm](<https://devfeed.tech/tags/vllm.md>), [vllm-blog](<https://devfeed.tech/tags/vllm-blog.md>)

### AI overview

The article introduces a push-based NIXL connector for vLLM disaggregated inference. It overlaps KV-cache transfer setup with prefill, moving work off the decode instance's latency-critical path. In a Qwen3-32B benchmark on a one-prefill, one-decode deployment using Amazon EC2 P5en instances with EFA, push mode achieved up to 3x better time to first token and up to 30% better time per output token than pull mode; results vary by workload and configuration.

### Source excerpt

A push-based NIXL connector for vLLM disaggregated serving: the prefill instance writes KV directly into decode memory over RDMA, taking the router relay, decode-side allocation, and transfer setup off the TTFT critical path.

## KIOXIA CM9-DLC Brings Direct Liquid Cooling to Dual-Port E3.S SSDs up to 30.72TB

DevFeed: [KIOXIA CM9-DLC Brings Direct Liquid Cooling to Dual-Port E3.S SSDs up to 30.72TB](<https://devfeed.tech/articles/kioxia-cm9-dlc-brings-direct-liquid-cooling-to-dual-port-e3-s-ssds-up-to-30-72tb-85443.md>)

Original publisher: [Read original article](<https://www.storagereview.com/news/kioxia-cm9-dlc-direct-liquid-cooling-e3s-ssd-30-72tb>)

Author: Brian Beeler

Published: 2026-10-12T15:01:33Z

Content type: news

Language: en

Sources: [StorageReview.com](<https://devfeed.tech/sources/storagereview-com.md>)

Topics: [gpudirect rdma](<https://devfeed.tech/topics/gpudirect-rdma.md>), [GPUDirect](<https://devfeed.tech/topics/gpudirect.md>), [Infrastructure as a Service (IaaS)](<https://devfeed.tech/topics/infrastructure-as-a-service-iaas.md>)

Tags: [e3-s](<https://devfeed.tech/tags/e3-s.md>), [enterprise](<https://devfeed.tech/tags/enterprise.md>), [gpus](<https://devfeed.tech/tags/gpus.md>), [kioxia](<https://devfeed.tech/tags/kioxia.md>), [kv-cache](<https://devfeed.tech/tags/kv-cache.md>), [pcie-5-0](<https://devfeed.tech/tags/pcie-5-0.md>), [servers](<https://devfeed.tech/tags/servers.md>), [ssd](<https://devfeed.tech/tags/ssd.md>)

### AI overview

KIOXIA announced CM9-DLC, a direct-liquid-cooled option in its PCIe 5.0 CM9 enterprise SSD line. The E3.S drives support liquid-cooled and air-cooled servers, with capacities up to 30.72TB for read-intensive use and 25.6TB for mixed use. The article also describes the drives' AI server context-caching use case and notes that pricing and general availability dates have not been announced.

### Source excerpt

KIOXIA America announced the CM9-DLC today, a set of direct liquid cooling (DLC) models in its CM9 Series of PCIe 5.0 enterprise SSDs. The E3.S drives mount adjacent to cold plates in liquid-cooled servers and also run just fine in air-cooled systems. Capacities reach 30.72TB in the read-intensive (1 DWPD) version and 25.6TB in the The post KIOXIA CM9-DLC Brings Direct Liquid Cooling to Dual-Port E3.S SSDs up to 30.72TB appeared first on StorageReview.com.

## Scality AI Inference Factory Serves KV Cache From Object Storage Over RDMA, 14x Faster Than Recompute

DevFeed: [Scality AI Inference Factory Serves KV Cache From Object Storage Over RDMA, 14x Faster Than Recompute](<https://devfeed.tech/articles/scality-ai-inference-factory-serves-kv-cache-from-object-storage-over-rdma-14x-faster-than-recompute-77324.md>)

Original publisher: [Read original article](<https://www.storagereview.com/news/scality-ai-inference-factory-kv-cache-object-storage-rdma>)

Author: Harold Fritts

Published: 2026-10-08T16:32:44Z

Content type: news

Language: en

Sources: [StorageReview.com](<https://devfeed.tech/sources/storagereview-com.md>)

Topics: [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [vllm](<https://devfeed.tech/topics/vllm.md>), [rdma](<https://devfeed.tech/topics/rdma.md>), [gpudirect rdma](<https://devfeed.tech/topics/gpudirect-rdma.md>), [Dynamo-Triton](<https://devfeed.tech/topics/dynamo-triton.md>), [Amazon S3](<https://devfeed.tech/topics/amazon-s3.md>), [versioning](<https://devfeed.tech/topics/versioning.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-data-infrastructure](<https://devfeed.tech/tags/ai-data-infrastructure.md>), [ai-inference](<https://devfeed.tech/tags/ai-inference.md>), [bandwidth](<https://devfeed.tech/tags/bandwidth.md>), [code](<https://devfeed.tech/tags/code.md>), [concurrent](<https://devfeed.tech/tags/concurrent.md>), [enterprise](<https://devfeed.tech/tags/enterprise.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [object-storage](<https://devfeed.tech/tags/object-storage.md>), [stack](<https://devfeed.tech/tags/stack.md>)

### AI overview

Scality launched AI Inference Factory, an open-code stack for running open-weight models on customer-owned infrastructure. It uses ADI object storage as a shared KV cache, with vLLM and Dynamo separating prefill and decode, and an RDMA path for moving cache data to GPUs. In a small lab, Scality reported that restoring a 14K-token context took 166 ms, compared with 2.3 seconds to recompute it; it also reported serving 8 TB of KV cache with first-token latency between 355 and 373 ms across the tested cache sizes.

### Source excerpt

Scality today launched Scality AI Inference Factory, an open-code software stack for running open-weight models on infrastructure the customer owns. Scality validates, ships, and maintains the whole stack, which places its AI Data Infrastructure (ADI) object storage under the inference layer as a shared key-value (KV) cache, and it is aimed at enterprises, government agencies, The post Scality AI Inference Factory Serves KV Cache From Object Storage Over RDMA, 14x Faster Than Recompute appeared first on StorageReview.com.

## How DOCA GPUNetIO Unifies GPU-Initiated Networking Across the NVIDIA Software Stack

DevFeed: [How DOCA GPUNetIO Unifies GPU-Initiated Networking Across the NVIDIA Software Stack](<https://devfeed.tech/articles/how-doca-gpunetio-unifies-gpu-initiated-networking-across-the-nvidia-software-stack-65883.md>)

Original publisher: [Read original article](<https://developer.nvidia.com/blog/doca-gpunetio-gda-ki-unified-gpu-networking/>)

Author: Elizabeth Goodman

Published: 2026-10-06T19:07:02Z

Content type: article

Language: en

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

Topics: [gpudirect rdma](<https://devfeed.tech/topics/gpudirect-rdma.md>), [DOCA](<https://devfeed.tech/topics/doca.md>), [NCCL](<https://devfeed.tech/topics/nccl.md>), [Multi-GPU](<https://devfeed.tech/topics/multi-gpu.md>), [GPUDirect](<https://devfeed.tech/topics/gpudirect.md>), [scalable-architecture](<https://devfeed.tech/topics/scalable-architecture.md>)

Tags: [architecture](<https://devfeed.tech/tags/architecture.md>), [data-center-cloud](<https://devfeed.tech/tags/data-center-cloud.md>), [doca](<https://devfeed.tech/tags/doca.md>), [edge-computing](<https://devfeed.tech/tags/edge-computing.md>), [featured](<https://devfeed.tech/tags/featured.md>), [nccl](<https://devfeed.tech/tags/nccl.md>), [networking](<https://devfeed.tech/tags/networking.md>), [networking-communications](<https://devfeed.tech/tags/networking-communications.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [nvlink](<https://devfeed.tech/tags/nvlink.md>), [nvshmem](<https://devfeed.tech/tags/nvshmem.md>), [software](<https://devfeed.tech/tags/software.md>)

### AI overview

NVIDIA's DOCA GPUNetIO provides a shared foundation for GPU-initiated networking, allowing CUDA kernels to drive Ethernet and RDMA operations while reducing CPU involvement in the application data path. The article describes its open-source and DOCA SDK implementations, programming model, and integrations with libraries including NCCL and NVSHMEM. It reports that enabling GDA-KI in an NVSHMEM benchmark improved scaling across CTAs and QPs for small messages by removing a CPU proxy bottleneck.

### Source excerpt

GPU applications increasingly need networking and data movement to behave like first-class GPU-controlled operations rather than host-driven services. When the...

## Gigabyte W775-V10-L01 Hands-on Bringing NVIDIA GB300 Deskside

DevFeed: [Gigabyte W775-V10-L01 Hands-on Bringing NVIDIA GB300 Deskside](<https://devfeed.tech/articles/gigabyte-w775-v10-l01-hands-on-bringing-nvidia-gb300-deskside-65692.md>)

Original publisher: [Read original article](<https://www.servethehome.com/gigabyte-w775-v10-l01-hands-on-bringing-nvidia-gb300-deskside/>)

Author: Patrick Kennedy

Published: 2026-10-06T17:00:41Z

Content type: comparison

Language: en

Sources: [ServeTheHome](<https://devfeed.tech/sources/servethehome.md>)

Topics: [DGX Station](<https://devfeed.tech/topics/dgx-station.md>), [GB300 NVL72](<https://devfeed.tech/topics/gb300-nvl72.md>), [gpudirect rdma](<https://devfeed.tech/topics/gpudirect-rdma.md>), [Nvidia](<https://devfeed.tech/topics/nvidia.md>)

Tags: [10gbe](<https://devfeed.tech/tags/10gbe.md>), [ai](<https://devfeed.tech/tags/ai.md>), [blackwell](<https://devfeed.tech/tags/blackwell.md>), [connectx](<https://devfeed.tech/tags/connectx.md>), [data-center](<https://devfeed.tech/tags/data-center.md>), [dgx-station](<https://devfeed.tech/tags/dgx-station.md>), [gb300](<https://devfeed.tech/tags/gb300.md>), [gigabyte](<https://devfeed.tech/tags/gigabyte.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [grace-cpu](<https://devfeed.tech/tags/grace-cpu.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [workstation](<https://devfeed.tech/tags/workstation.md>)

### AI overview

The article reviews Gigabyte's W775-V10-L01, a desk-side NVIDIA DGX Station built around the GB300 platform. It describes the system's liquid cooling, power requirements, networking, and component layout, and reports testing Qwen3.6 inference performance, ConnectX-8 RDMA networking, and the Grace CPU-to-GPU C2C link.

### Source excerpt

We test the Gigabyte W775-V10-L01, including getting over 2.2B tokens/ day on the Blackwell Ultra GPU, 800Gbps from the ConnectX-8 SuperNIC, and testing the NVIDIA Grace CPU with the C2C link between it and the GPU The post Gigabyte W775-V10-L01 Hands-on Bringing NVIDIA GB300 Deskside appeared first on ServeTheHome.