# rdma

Published articles for rdma.

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

## Нейро сети для самых маленьких. Часть первая (которая после нулевой). Удобство в прокрустовом ложе оптимизации

DevFeed: [Нейро сети для самых маленьких. Часть первая (которая после нулевой). Удобство в прокрустовом ложе оптимизации](<https://devfeed.tech/articles/article-24859.md>)

Original publisher: [Read original article](<https://habr.com/ru/companies/yandex/articles/1047072/>)

Author: eucariot (Яндекс, Yandex Cloud & Yandex Infrastructure)

Published: 2026-07-01T07:00:06Z

Content type: article

Language: ru

Sources: [Яндекс - Как мы делаем Яндекс / Статьи](<https://devfeed.tech/sources/source.md>)

Topics: [High-Performance Computing](<https://devfeed.tech/topics/high-performance-computing.md>), [GPU](<https://devfeed.tech/topics/gpu.md>), [InfiniBand](<https://devfeed.tech/topics/infiniband.md>), [NVLink](<https://devfeed.tech/topics/nvlink.md>), [Linux](<https://devfeed.tech/topics/linux.md>)

Tags: [ethernet](<https://devfeed.tech/tags/ethernet.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [gpudirect-rdma](<https://devfeed.tech/tags/gpudirect-rdma.md>), [high-performance-computing](<https://devfeed.tech/tags/high-performance-computing.md>), [infiniband](<https://devfeed.tech/tags/infiniband.md>), [linux](<https://devfeed.tech/tags/linux.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [nvlink](<https://devfeed.tech/tags/nvlink.md>), [performance](<https://devfeed.tech/tags/performance.md>), [rdma](<https://devfeed.tech/tags/rdma.md>), [roce](<https://devfeed.tech/tags/roce.md>), [tcp](<https://devfeed.tech/tags/tcp.md>), [zero-copy](<https://devfeed.tech/tags/zero-copy.md>)

### AI overview

This introductory article in a series explains the infrastructure used to train and run neural networks and for high-performance computing. It surveys specialized technologies including GPUs and TPUs, RDMA, kernel bypass, NVLink, InfiniBand, and RoCE, arguing that specialized solutions can outperform and cost less than a generic Linux and Ethernet/IP stack at scale.

### Source excerpt

Это первая (после нулевой) статья из серии Нейро сети для самых маленьких, в которой мы разбираем инфраструктуру для запуска нейронных сетей. Для обучения и инференса нейросетей и для любых видов High Performance Computing используются специализированные технологии: GPU/TPU, RDMA, Kernel bypass, NVLink, InfiniBand, RoCE и другие. Про некоторые из них большинство только что-то слышали, но сталкиваться с ними не приходилось. Нельзя просто взять ванильный стек Linux, воткнуть в него 400Gb Ethernet+IP и получить рабочее решение. Почему? Потому что общее решение на масштабе в большинстве случаев проигрывает специализированным как в скорости, так и в стоимости. Как бы странно последнее ни звучало. Читать далее

## QEMU version 10.1.0 released

DevFeed: [QEMU version 10.1.0 released](<https://devfeed.tech/articles/qemu-version-10-1-0-released-32653.md>)

Original publisher: [Read original article](<https://www.qemu.org/2025/08/26/qemu-10-1-0/>)

Published: 2025-08-26T23:25:00Z

Content type: release

Language: en

Sources: [QEMU](<https://devfeed.tech/sources/qemu.md>)

Topics: [qemu](<https://devfeed.tech/topics/qemu.md>), [version](<https://devfeed.tech/topics/version.md>), [virtualization](<https://devfeed.tech/topics/virtualization.md>), [virtual machines](<https://devfeed.tech/topics/virtual-machines.md>), [RISC-V](<https://devfeed.tech/topics/riscv.md>), [cpu](<https://devfeed.tech/topics/cpu.md>), [Kernel](<https://devfeed.tech/topics/kernel.md>), [Arm](<https://devfeed.tech/topics/arm.md>), [InfiniBand](<https://devfeed.tech/topics/infiniband.md>), [intel](<https://devfeed.tech/topics/intel.md>), [GB200](<https://devfeed.tech/topics/gb200.md>)

Tags: [arm](<https://devfeed.tech/tags/arm.md>), [cxl](<https://devfeed.tech/tags/cxl.md>), [gb200](<https://devfeed.tech/tags/gb200.md>), [intel](<https://devfeed.tech/tags/intel.md>), [ipv6](<https://devfeed.tech/tags/ipv6.md>), [kernel](<https://devfeed.tech/tags/kernel.md>), [kvm](<https://devfeed.tech/tags/kvm.md>), [migration](<https://devfeed.tech/tags/migration.md>), [qemu](<https://devfeed.tech/tags/qemu.md>), [qemu-10-1](<https://devfeed.tech/tags/qemu-10-1.md>), [rdma](<https://devfeed.tech/tags/rdma.md>), [release](<https://devfeed.tech/tags/release.md>), [releases](<https://devfeed.tech/tags/releases.md>), [risc-v](<https://devfeed.tech/tags/risc-v.md>), [version](<https://devfeed.tech/tags/version.md>), [virtual-machines](<https://devfeed.tech/tags/virtual-machines.md>), [virtualization](<https://devfeed.tech/tags/virtualization.md>), [x86](<https://devfeed.tech/tags/x86.md>)

### AI overview

QEMU 10.1.0 has been released with more than 2,700 commits from 226 authors. Highlights include confidential-guest support, live-migration improvements, Windows guest load querying, new ARM and RISC-V capabilities, and expanded KVM and x86 virtualization support.

### Source excerpt

We'd like to announce the availability of the QEMU 10.1.0 release. This release contains 2700+ commits from 226 authors.

## A Rare, Nondeterministic RDMA Bug in Large Distributed Training Jobs

DevFeed: [A Rare, Nondeterministic RDMA Bug in Large Distributed Training Jobs](<https://devfeed.tech/articles/a-cursed-bug-21940.md>)

Original publisher: [Read original article](<https://blog.nelhage.com/post/a-cursed-bug/>)

Author: Nelson Elhage

Published: 2022-02-23T03:03:48Z

Content type: opinion

Language: en

Sources: [Nelson Elhage](<https://devfeed.tech/sources/nelson-elhage.md>)

Topics: [bug](<https://devfeed.tech/topics/bug.md>), [distributed-systems](<https://devfeed.tech/topics/distributed-systems.md>), [distributed-training](<https://devfeed.tech/topics/distributed-training.md>), [Processes](<https://devfeed.tech/topics/processes.md>), [coding](<https://devfeed.tech/topics/coding.md>)

Tags: [bug](<https://devfeed.tech/tags/bug.md>), [code](<https://devfeed.tech/tags/code.md>), [debugging](<https://devfeed.tech/tags/debugging.md>), [distributed-systems](<https://devfeed.tech/tags/distributed-systems.md>), [distributed-training](<https://devfeed.tech/tags/distributed-training.md>), [process](<https://devfeed.tech/tags/process.md>), [processes](<https://devfeed.tech/tags/processes.md>), [python](<https://devfeed.tech/tags/python.md>), [rdma](<https://devfeed.tech/tags/rdma.md>)

### AI overview

An Anthropic engineer describes a rare, nondeterministic bug in very large distributed training jobs. The bug originated in the RDMA software stack and caused subprocess launches to fail on a small fraction of nodes, sometimes producing Python exceptions or segmentation faults.

### Source excerpt

In my day job at Anthropic, we run relatively large distributed systems to train large language models. One of the joys of using a lot of computing resources, especially on somewhat niche software stacks, is that you spend a lot of time running into the long-tail of bugs which only happen rarely or in very unusual configurations, which you happen to be the first to encounter. These bugs are frustrating, but I also often enjoy them.