本文链接

晓兵博客: https://logread.cn/post/study/summary/

腾讯云: https://cloud.tencent.com/developer/article/2355052

CSDN: https://blog.csdn.net/ssbandjl/article/details/131265171

术语

xilinx Zynq: 这个词是由锌创造的,锌是电池、太阳能屏幕、合金产品和药品中最常见的化学元素锌。 锌与其他金属的合金可实现增强功能,根据合金的不同用途呈现出不同的颜色

orthogonal frequency-division multiplexing (OFDM) : 正交频分多路复用

Shared Virtual Memory (SVM) 统一虚拟地址空间

SD: segment descriptor 段描述

Physical Buffer List Entry (PBLE)

Virtual Station Interfaces (VSIs) : 虚拟

virtio full emulation (VFE): 虚拟全仿真, vdpa

“other interrupt causes” (OIC)

功能级别重置 (FLR)

Page Table Entry(PTE): 页表条目

CFS: 完全公平调度

EDF: 最早截止期限有限调度算法

MSI: Message Signaled Interrupts (MSIs), 消息信号中断( MSI ) 是一种发出中断信号的方法,它使用特殊的带内消息来替代专用中断线路上的传统带外信号。虽然消息信号中断在设备中实现起来更为复杂,但与基于引脚的带外中断信号相比,它们具有一些显着的优势,例如改进的中断处理性能。这与传统中断机制(例如传统中断请求(IRQ)系统)形成对比

CMB(controller memory buffer): 控制器内存缓冲区

IO映射: 控制寄存器, 状态寄存器, 数据寄存器

BBR: Bottleneck Bandwidth and Round-trip propagation time 瓶颈带宽和往返传播时间

SMMUv3: System Memory Management Unit (SMMU) (otherwise known as IOMMU) , 系统内存管理单元, 通过添加存储在内存中的配置来扩展早期 SMMU 架构的架构概念,从而支持更大的 IO 系统。 SMMUv3 还为基于 PCI Express 的系统提供增强支持,包括 PCIe ATS 和 PRI 功能。

CSR, 控制/状态寄存器 CBL, 命令块列表 RFA, 接收帧区域

embedded switch (eSwitch) within the NIC subsystem

Network functions virtualization (NFV): 网络功能虚拟化

SDN Software-defined networking

Direct Access Transport (DAT)

KNEM stands for Kernel-Nemesis 内核克星/复仇者

NVIDIA Messaging Accelerator (VMA)

Accelerated IO SW library (XLIO) 系统镜像(System Image,SI)

Cache空间用作RAM,Cache As Ram, CAR

Quantum 量子

Advanced Programmable Interrupt Controller(APIC,高级可编程中断控制器)

中断服务例程(ISR)

GRO(Generic Receive Offloading), Large Receive Offloading (LRO) 是一个硬件优化,GRO 是 LRO 的一种软件实现

Linux 内核内存模型 ( LKMM )

bql: Byte Queue Limits Digital Phase Locked Loops (DPLL)

Interconnect Context Memory (ICM) : 互联上下文内存

Precision Time Measurement (PTM) : 精密时间测量

vDPA (virtio data path acceleration)

default timeouts register (DTOR)

Dynamic Device Personalization, DDP

mailbox Transmit Queue (ATQ)

RISC (reduced instruction set computer) and CISC (complex instruction set computer) : 分别对应精简指令集和复杂指令集

论文

中国大陆发第一篇计算机顶会的人都怎么样了: https://www.zhihu.com/question/627499717

云栖大会

聚合话题

https://yunqi.aliyun.com/2023/agenda

Nvidia

GDS(GPU直接存储): https://docs.nvidia.com/gpudirect-storage/design-guide/index.html

技术视频点播(精选播放列表): https://www.nvidia.com/en-us/on-demand/featured-playlist/

Nvidia网络文档: https://docs.nvidia.com/networking/software/index.html

学习目录: https://nvdam.widen.net/s/wlbgbqr7cj/nvidia-learning-training-course-catalog

DOCA纠删码编程指南

https://docs.nvidia.com/doca/sdk/erasure-coding-programming-guide/index.html

Nvidia在线研讨会(视频教程)

https://www.nvidia.cn/about-nvidia/webinar-portal

Nvidia深度学习学院

https://www.nvidia.com/en-us/training/online/

Nvidia深度学习培训中心

https://www.nvidia.cn/training/

Mellanox学院

https://academy.nvidia.com/

英伟达三大AI法宝:CUDA、Nvlink、InfiniBand

https://mp.weixin.qq.com/s/rVWoSl-QbwdwoSB0cNTXbQ

Nvidia技术博客: https://developer.nvidia.com/blog/

Nvidia文档: https://docs.nvidia.com/

Nvidia下载中心: https://developer.nvidia.com/downloads

Nvidia开发者会员专区: https://developer.nvidia.com/nvidia-developer-program

Spectrum(光谱)

使用 Spectrum 以太网交换机加速数据中心结构(2022/03): https://www.nvidia.com/en-us/on-demand/session/gtcspring22-s41648

DPU

DOCA文档汇总: https://docs.nvidia.com/doca/sdk/index.html

DOCA与UCX编程指导: https://docs.nvidia.com/doca/sdk/ucx-programming-guide/index.html

DPU_BSP手册: https://docs.nvidia.com/networking/display/bluefielddpubspv420

DOCA与RDMA编程手册: https://docs.nvidia.com/doca/sdk/rdma-programming-guide/index.html

DOCA核心编程指南: https://docs.nvidia.com/doca/sdk/doca-core-programming-guide/index.html

DOCA参考应用汇总(概览): https://docs.nvidia.com/doca/sdk/applications-overview/index.html

DOCA_PCC(可编程拥塞控制): https://docs.nvidia.com/doca/sdk/pcc/index.html

在 BlueField DPU 上使用 NVIDIA DOCA 开发应用程序简介: https://www.nvidia.com/en-us/on-demand/session/other2022-dc0511/

使用 Red Hat OpenShift 和 NVIDIA BlueField DPU 加速云就绪基础设施和 Kubernetes: https://developer.nvidia.com/zh-cn/blog/accelerating-cloud-ready-infrastructure-and-kubernetes-with-red-hat-openshift-and-bluefield-dpu/

BF驱动BSP术语: https://docs.nvidia.com/networking/display/bluefielddpuosv450

DPU演化: https://zhuanlan.zhihu.com/p/402706136

使用 NVIDIA BlueField DPU 和 DPDK 开发应用程序: https://developer.nvidia.com/blog/developing-applications-with-nvidia-bluefield-dpu-and-dpdk/

中文: https://zhuanlan.zhihu.com/p/468289438

DPU专栏(大空新一): https://blog.csdn.net/gongtxy/article/details/128740434

Corundum 是一款基于FPGA的开源高性能NIC和网络内计算平台-文档: https://docs.corundum.io/en/latest/

业界第一个真正意义上开源100 Gbps NIC Corundum介绍: https://mp.weixin.qq.com/s/Dc8rdlyR3Z5VohghTYEANQ

BF2快速入门指南(NVIDIA BlueField-2 DPU快速入门指南): https://devforum.nvidia.cn/forum.php?mod=viewthread&tid=12823&extra=page%3D1

openvswitch卸载: NVIDIA Docs Hub NVIDIA Networking Networking Software DOCA Documentation v2.5.0 OpenvSwitch Offload

网络

多主机Multihost用户手册: https://network.nvidia.com/sites/default/files/doc-2020/connectx-4-multi-host-evaluation-kit-user-manual.pdf

MultiHostSDK: https://network.nvidia.com/sites/default/files/doc-2020/pb-multi-host-evb-kit.pdf

DPU多主机功能简介: https://docs.nvidia.com/networking/display/bluefielddpuosv450/multi-host

多主机架构解决方案/优点: https://www.nvidia.com/en-us/networking/multi-host/

6DX白皮书: https://www.nvidia.com/content/dam/en-zz/Solutions/networking/ethernet-adapters/connectX-6-dx-datasheet.pdf

下一代AI网络白皮书: https://resources.nvidia.com/en-us-accelerated-networking-resource-library/next-generation-netw?ncid=so-face-852347-vt45&fbclid=IwAR1ONw41JYU_J1CnJbELrWgAg50rSbAHbrPKZwCmp7P3euKhD0Tal1F0JqQ

mlx5驱动源码分析: https://blog.csdn.net/leiyanjie8995/article/details/121341828 cx6dx_ovs流表卸载源码分析: https://blog.csdn.net/qq_20679687/article/details/131632198

驱动/适配器编程

Mellanox CX4/4Lx适配器-程序员参考手册 (PRM)-PRM-CX4/4Lx: https://network.nvidia.com/files/doc-2020/ethernet-adapters-programming-manual.pdf

GPU

GPU集群组网技术分析:NVLink、InfiniBand、ROCE、DDC: https://mp.weixin.qq.com/s/CJCm7nqACSjJWGHHd7cMFA

“AI时代,你需要了解的GPU互联技术:NVLink、IB、ROCE” https://mp.weixin.qq.com/s/u7ZOD0qLIpUXgcZtluoBmg

热门互联网络技术

第30届IEEE热互连研讨会: https://hoti.org/

技术视频: https://www.youtube.com/@hoti-hotinterconnectssympo5358/videos

议题: https://hoti.org/program/

Hot Interconnects 是最先进的硬件和软件架构的研究人员和开发人员的首要国际论坛,…

Hot Interconnects 是最先进的硬件和软件架构以及各种规模互连网络实现的研究人员和开发人员的首要国际论坛,范围从多核片上互连到系统、集群、数据中心内的互连 和云。 业界和学术界的领导人出席了这一年度会议。 这种氛围提供了大量与该领域前沿人士互动的机会。

主题包括跨领域问题,涵盖计算机系统、网络技术和高性能互连网络的通信协议。 本次会议特别针对这些领域的令人兴奋的新技术和产品创新。

欢迎来到第 30 届 IEEE 热互连研讨会。 HotI30 将以虚拟方式举行。 IEEE Hot Interconnects 是最先进的硬件和软件架构以及各种规模互连网络实现的研究人员和开发人员的首要国际论坛,范围从多核片上互连到系统集群内的互连和数据互连。 中心。 工业界和学术界的领导者出席会议,与该领域最前沿的人士进行互动。

我们的目标是解决数据中心网络和超级计算社区的问题。 我们希望您能加入我们,不仅从内容中受益,还从本次活动始终提供的绝佳交流机会中受益

加速器互连 几乎所有新兴的高价值应用程序,尤其是那些位于边缘的应用程序,都使用加速器而不是 CPU 来实现其基本功能。 此外,在任何规模的集群中,包括边缘集群,都会采用多个加速器来满足此类应用程序的动态需求。 这些加速器可以是同一类型(作为池),也可以是不同类型。 但与 CPU 的情况一样,系统的性能几乎总是受到 I/O 或互连的限制。 因此,我们想要探索加速器的互连如何不仅避免成为瓶颈,更重要的是使加速器协同工作以实现新的应用程序并彻底改变计算架构。 该小组将探讨最具创新性的加速器互连方法,以充分利用这些令人惊叹的加速器的功能。

性能

P99 CONF 大会(P99分位数, 极致)是面向工程师和由工程师举办的跨行业虚拟活动,以低延迟、高性能设计为中心, 突出低延迟、高性能分布式计算所需的工程挑战和创造性解决方案

P99大会: https://www.p99conf.io/on-demand/

HPC高性能计算

Mochi 项目是阿贡国家实验室、洛斯阿拉莫斯国家实验室、卡内基梅隆大学和 HDF 集团之间的合作。 该项目的目标是探索一种软件定义的存储方法,用于组合存储服务,为极端规模的科学应用程序提供新级别的功能、性能和可靠性

Mochi(麻薯): https://mochi.readthedocs.io/en/latest/

TOP500: https://www.top500.org/lists/top500/2023/11/

Linux

linux内核文档: https://www.kernel.org/doc/html/latest

内核子系统: https://docs.kernel.org/subsystem-apis.html#networking-interfaces

linux内存管理: https://www.kernel.org/doc/gorman/html/understand/

centos_内核编译: https://wiki.centos.org/HowTos

linux内核makefile和官方文档: https://www.kernel.org/doc/html/latest/kbuild/makefiles.html

https://docs.kernel.org/kbuild/makefiles.html

https://youjiali1995.github.io/categories/

https://www.scylladb.com/

https://www.scylladb.com/blog/

linux内核: http://www.kerneltravel.net/

linux内核博客: https://blog.csdn.net/hu1610552336?type=blog

linux内核社群: https://www.sohu.com/a/395455440_467784

linux文件系统: https://www.kernel.org/doc/Documentation/filesystems/vfs.txt

linux内核CLK(中国Linux内核开发者大会): http://ckernel.org/

linux内核文档: https://linux-kernel-labs.github.io/refs/pull/183/merge/labs/device_model.html

systemtap教程汇总: https://github.com/lichuang/awesome-systemtap-cn

中文档: https://spacewander.gitbooks.io/systemtapbeginnersguide_zh/content/4_3_UserSpaceStackBacktraces.html

探测点: https://linux.die.net/man/3/stapprobes

官方示例: https://sourceware.org/systemtap/examples/

bpftrace官方教程,指南: https://github.com/iovisor/bpftrace/blob/master/docs/reference_guide.md

动态调试: https://www.kernel.org/doc/html/v4.14/admin-guide/dynamic-debug-howto.html

中国Unix操作系统: http://blog.chinaunix.net/os.html

函数调用栈: https://ivanzz1001.github.io/records/post/linux/2017/10/18/linux-function_callstack_part1

linux内核: https://ivanzz1001.github.io/records/post/kernel/2020/03/31/kernel_intro_part1

linux内核: https://www.zhihu.com/people/ning-meng-meng-meng-47-83

linux内核: https://hackmd.io/@RinHizakura

高级内存管理(弹性缓冲区): https://www.kernel.org/doc/gorman/html/understand/understand012.html

linux内核,内存管理: https://mp.weixin.qq.com/s/Torf_AHepkgNrznG5G2NJw

再论软硬件融合(软件定义硬件SDX) https://mp.weixin.qq.com/s/jn9-vpQl3SNV14QGfC8wXg

Linux天气预报: https://www.linux.com/news/linux-weather-forecast

Linux变更-新功能: https://kernelnewbies.org/LinuxChanges

Linux2023十大革新功能: https://mp.weixin.qq.com/s/kVdKcQUqm7vAYuljMtemuw

2023内核开发者大会: https://github.com/ChinaLinuxKernel/CLK2023/tree/main/%E4%B8%BB%E8%AE%BA%E5%9D%9B

Linux内核源码分析-强烈推荐收藏: https://mp.weixin.qq.com/s/9q_n6xnWyjRk0Ph8TG32dg

深入理解epoll:高效I/O多路复用的核心技术: https://mp.weixin.qq.com/s/QFqcn5ck7vNBP254tTrUQQ

深入理解Linux内核I/O机制:探索文件系统与设备驱动: https://mp.weixin.qq.com/s/kCdZUZrYMMaN3L526liN4g

Block multi-queue 架构解析(一)数据结构: https://blog.csdn.net/qq_32740107/article/details/106302376

Linux内核学习笔记: https://void-star.icu/archives/153

内核笔记: https://zhuanlan.zhihu.com/p/635315467

Linux内核视频教学(腾讯课堂/零声): https://ke.qq.com/course/4032547

2022 Linux 内核十大技术革新功能 | 年终盘点: https://cloud.tencent.com/developer/article/2257733

蜗窝科技: http://www.wowotech.net/

驱动

如何写Linux PCI驱动: https://docs.kernel.org/translations/zh_CN/PCI/pci.html

Linux设备驱动第三版: https://lwn.net/Kernel/LDD3/

Linux内核在线书籍: https://0xax.gitbooks.io/linux-insides/content/

用户态驱动VFIO

使用VFIO进行用户态驱动开发: https://zhuanlan.zhihu.com/p/532927980

VFIO简介: https://blog.csdn.net/wentyoon/article/details/60144824

内存管理

内存管理1_为什么需要虚拟内存: https://zhuanlan.zhihu.com/p/404813126

深入浅出Linux内核内存管理基础: https://zhuanlan.zhihu.com/p/597993158

一步一图带你深入理解 Linux 物理内存管理

【概述】-Linux内核三驾马车之-内存管理: https://blog.acean.vip/post/linux-kernel/gai-shu-linuxnei-he-san-jia-ma-che-zhi-nei-cun-guan-li#toc_9

数据结构

ring_buf环形缓冲区

ring buffer,一篇文章讲透它?: https://zhuanlan.zhihu.com/p/534098236 内存管理专题: https://www.cnblogs.com/arnoldlu/p/8051674.html

BPF

Linux超能力BPF技术介绍及学习分享(技术创作101训练营): https://cloud.tencent.com/developer/article/1698426

driver

内核驱动调试方法: https://wiki.phytec.com/pages/viewpage.action?pageId=132776352 利用GDB、KGDB调试应用程序及内核驱动模块: https://blog.nsfocus.net/gdb-kgdb-debug-application/

log

内核如何打印日志及动态调试(dev_info/dbg…): https://cloud.tencent.com/developer/article/1819284

虚拟化

虚拟化(六六哥): https://blog.csdn.net/leoufung/category_5845399.html

基于VIRTIO的虚拟化方案

https://mp.weixin.qq.com/s/RNpR4Chv6Az2HS34Ov1vmQ

qemu

系列专栏: https://phmatthaus.blog.csdn.net/category_12072874_2.html

QEMU虚拟机(一)、源码分析概论(相关参考链接): https://blog.csdn.net/weixin_38387929/article/details/120121636

QEMU源码全解析 —— PCI设备模拟(7)(MMIO设备读写/触发中断/DMA等): https://phmatthaus.blog.csdn.net/article/details/135547308

使用GDB: GDB usage: https://qemu-project.gitlab.io/qemu/system/gdb.html

https://xilinx-wiki.atlassian.net/wiki/spaces/A/pages/821624963/Debugging+Guest+Applications+with+QEMU+and+GDB

使用qemu和gdb调试内核: https://imkira.com/QEMU-GDB-Linux-Kernel/

存储

华为存储: https://e.huawei.com/cn/material/MaterialList?id={A246BE8A-D1A6-4AA6-852F-16AD3C4BC7AA}

存储技术最前线: nvme, ceph, spdk, 虚拟化汇总

IO子系统全流程

SSD存储汇总:

存储随笔: https://mp.weixin.qq.com/template/article/1688888165/index.html

nvme: https://mp.weixin.qq.com/mp/homepage?__biz=MzIwNTUxNDgwNg==&hid=1&sn=b8edd088ef636f736518b201f697cc9d&devicetype=iOS16.3.1&version=18002038&lang=zh_CN&nettype=WIFI&ascene=7&session_us=gh_b79997f0dcfa&fontScale=94&scene=1

存储随笔《NVMe专题》大合集: https://memory-logger.blog.csdn.net/article/details/76136138

闪存和存储: https://www.techtarget.com/searchstorage/resources/Flash-memory-and-storage

存储之道: https://blog.51cto.com/alanwu/category4

Nvmeof

NvidiaNvmeof配置视频: https://academy.nvidia.com/en/videos/

nvmeof协议规范(spec): http://47.92.214.21:8888/nvme/NVMe-over-Fabrics-1.1-2019.10.22-Ratified.pdf

nvmeof内核驱动: https://lincolnli.blog.csdn.net/article/details/132489838

收藏:NVMe协议基础原理介绍: https://cloud.tencent.com/developer/article/2192563

nvme设备驱动(专栏): https://www.cnblogs.com/vlhn/tag/NVMe

IB

协议规范: [http://47.92.214.21:8888/rdma/IB%20Specification%20Vol%201-Release-1.4-2020-04-07_ib_spec_vol1.pdf](http://47.92.214.21:8888/rdma/IB Specification Vol 1-Release-1.4-2020-04-07_ib_spec_vol1.pdf)

IB协议中文翻译版本链接(汇总):https://ngdcn.com/tags-28.html 第3章(架构预览):https://ngdcn.com/post/94.html 第5章(数据报文格式):https://ngdcn.com/post/97.html 第9章(传输层):https://ngdcn.com/post/107.html 第10章(软件传输接口):https://ngdcn.com/post/148.html

数据库

redis_rocksdb_leveldb_scylladb_raft: https://youjiali1995.github.io/categories/

rocksdb

rocksdb博客: https://rocksdb.org/blog/

ceph

ceph博客: https://ceph.io/en/news/blog

ceph专栏(黑桃): https://ivanzz1001.github.io/records/categories/#ceph

ceph专栏(): https://www.dovefi.com/post/

ceph专栏(知乎): https://www.zhihu.com/column/c_1661407683897901056

ceph管理手册(红帽): https://access.redhat.com/documentation/en-us/red_hat_ceph_storage/1.3/html/administration_guide/index

ceph原理_rocksdb_bluestor_两阶段2PC_事务_WAL等: https://www.zhihu.com/people/allincache/posts

ceph内部原理(读写详细流程): https://bean-li.github.io/tags/#ceph

Ceph学习——Librbd块存储库与RBD读写流程源码分析(RBD): https://blog.csdn.net/CSND_PAN/article/details/78728743

Ceph分布式存储学习笔记: https://zhuanlan.zhihu.com/p/643942435

ceph延迟分析: https://noahdesu.github.io/2014/06/08/rados-request-latency.html

ceph 的数据存储之路 (5) —–osd 数据处理, https://my.oschina.net/u/2460844/blog/534390

osd落盘, 主副本和从副本同步: https://my.oschina.net/u/2460844/blog/534390

daos

DAOS在后傲腾时代的发展策略: https://www.bilibili.com/video/BV1Qw411377s

项目计划: https://daosio.atlassian.net/wiki/spaces/DC/pages/4836661105/Roadmap

设计文档: https://daosio.atlassian.net/wiki/spaces/DC/pages/4806836263/Resources

晓兵技术杂谈(系列)

https://cloud.tencent.com/developer/user/5060293/video

欢迎对DAOS, SPDK, RDMA等高性能技术感兴趣的朋友加我WX(ssbandjl)进入DAOS技术交流(群)

主题 时间 链接 备注
晓兵技术杂谈1-intel_daos项目简介高性能分布式存储RDMA_SPDK_NVME_HPC对象存储源码分析_c语言 2023-09-02 00:22:15 https://cloud.tencent.com/developer/video/78657 https://www.bilibili.com/video/BV1mz4y1M7GN 分布式异步对象存储 (DAOS) 是一种开源对象存储,专为大规模分布式非易失性内存 (NVM) 而设计。DAOS 利用下一代 NVM 技术,如 Intel © Optane ™ Persistent Memory 和 NVM express (NVMe),同时在商用硬件之上提供键值存储接口,提供事务性非阻塞 I/O 等功能,具有自我修复、端到端数据完整性、细粒度数据控制和弹性存储的高级数据保护,以优化性能和成本, 参考: https://github.com/ssbandjl/daos
晓兵技术杂谈2-daos用户态文件系统io路径dfuse_io全路径io栈分布式存储c语言 2023-09-03 17:21:31 https://cloud.tencent.com/developer/video/78647 https://www.bilibili.com/video/BV1gu411P71a 分布式异步对象存储(DAOS,Distributed Asynchronous Object Storage)是一个开源的可扩展存储系统,从根本上设计用于在用户空间支持SCM和NVMe存储。DAOS在IO500基准测试中展现出领先的性能 DAOS从头开始设计,以利用现代化存储硬件(SCM、NVMe和CXL SSD)。其先进的低级键值API使其具有比传统基于POSIX的并行文件系统更高的IOPS和可扩展性,DAOS的IO500结果 2 也证实了这一点。需要注意的是,虽然DAOS提供了POSIX抽象层,但它也可以直接与自定义I/O中间件(如MPI-IO、HDF和几个AI/分析框架)集成,以提供比POSIX更多的功能。本研究仅在DAOS POSIX容器之上使用DAOS DFS API 参考: https://logread.cn/post/daos/daos_dfuse_fs_io_path/ https://blog.csdn.net/ssbandjl/article/details/132651510 https://cloud.tencent.com/developer/article/2322705
晓兵技术杂谈3-分布式存储系统-daos的rdma通信框架_verbs_rpc_cart_mercury_libfabric_c语言 2023-09-08 23:16:22 https://cloud.tencent.com/developer/video/78658 https://www.bilibili.com/video/BV1fp4y1L7VZ 晓兵技术杂谈3-分布式存储系统-daos与rdma通信框架_verbs_rpc_cart_mercury_libfabric_c语言 CaRT(集体和 RPC 传输)与Mercury CaRT 是用于大数据和 百亿级 HPC 的开源 RPC 传输层。 它支持传统的 P2P RPC 和集体 RPC,后者通过可扩展的基于树的消息传播在一组目标服务器上调用 RPC。Cart将Mercury封装,对应用程序提供初始化上下文,创建请求,发送请求,请求回调等数据通道接口,典型的上层应用如心跳swim,rank管理, 持久内存和Nvme读写IO, 底层通过libfabric或ucx封装RDMA的发送SEND/接收RECV和READ读/WRITE写接口 参考: https://blog.csdn.net/ssbandjl/article/details/130448137 htthttps://blog.csdn.net/ssbandjl/article/details/130655867 https://github.com/ssbandjl/libfabric/blob/main/ofi_guide.md https://github.com/ssbandjl/mercury.git https://github.com/ssbandjl/daos/tree/master/src/cart
晓兵技术杂谈4-分布式存储系统-daos的异步任务调度引擎框架TSE_task_schedule_engine_c语言编程 2023-09-16 22:01:48 https://cloud.tencent.com/developer/video/78659 https://www.bilibili.com/video/BV18u411c7Mg 晓兵技术杂谈4-分布式存储系统-daos的异步任务调度引擎框架TSE_task_schedule_engine_c语言编程 任务调度引擎: https://blog.csdn.net/ssbandjl/article/details/132178633 事件和事件队列,网络上下文即EV/EQ/CartContext https://blog.csdn.net/ssbandjl/article/details/132385609
晓兵技术杂谈5-分布式存储系统-daos底层通信rdma接口libfabric教程gpu数据dma_socket大块数据bulk传输原理rdma单边读写c语言编程.mp4 2023-09-24 23:41:08 https://cloud.tencent.com/developer/video/78660 https://www.bilibili.com/video/BV1nr4y1f7kE 晓兵技术杂谈5-分布式存储系统-daos底层通信rdma接口libfabric教程gpu数据dma_socket大块数据bulk传输原理rdma单边读写c语言编程ai_intel_panda熊猫博士 https://logread.cn/post/ofa/libfabric_tutorial_rdma_gpu_intel_dma_video_panda/ https://blog.csdn.net/ssbandjl/article/details/133248887
晓兵技术杂谈6-分布式存储系统-daos底层通信rdma接口分层调用详解libfabric_mercury高性能网络通信c语言编程云计算存储.mp4 2023-10-13 21:24:52 https://cloud.tencent.com/developer/video/78733 https://www.bilibili.com/video/BV1iN411b7qS 晓兵技术杂谈6分布式存储系统DAOS Mercury(HG) Libfabric(OFI) RDMA 分层verbs接口调用详解c语言代码走读通信_rpc, DAOS Mercury(HG) Libfabric(OFI) RDMA 分层verbs接口调用详解 , https://cloud.tencent.com/developer/article/2335208 htthttps://logread.cn/post/daos/daos_mercury_libfabric_rxm_rdma_verbs_rpc_bulk_api/ 常用学习链接(随时更新): https://logread.cn/post/study/summary/
晓兵技术杂谈7_DAOS分布式存储引擎是如何收到客户端RPC请求并处理的_c语言_rpc_cart_回调_mercury回调_rdma回调 2023-10-21 23:51:23 https://cloud.tencent.com/developer/video/78791 https://www.bilibili.com/video/BV11M41197yb 晓兵技术杂谈7_DAOS分布式存储引擎是如何收到客户端RPC请求并处理的_c语言_rpc_cart_回调_mercury回调_rdma回调 Intel高性能IO500分布式存储系统DAOS资源汇总-包含RDMA 网络 SPDK NVME TSE 任务调度 异步 事件队列等 https://cloud.tencent.com/developer/article/2344030 DAOS引擎是如何收到客户端RPC并处理的? https://cloud.tencent.com/developer/article/2348703
晓兵技术杂谈8_UCX统一通信框架_UCT传输层简介_rdma_零拷贝_rpc_高性能网络_c语言_常用通信语义_内存注册_云计算_GPU通信等 2023-11-04 22:41:48 https://cloud.tencent.com/developer/video/78930 https://www.bilibili.com/video/BV1Yg4y1X7aq 晓兵技术杂谈8_UCX统一通信框架_UCT传输层简介_rdma_零拷贝_rpc_高性能网络_c语言_常用通信语义_内存注册_云计算_GPU通信等_博客链接: https://cloud.tencent.com/developer/article/2349389 UCT设计: https://github.com/openucx/ucx/wiki/UCT-Design UCT文档: https://openucx.readthedocs.io/en/master/ucx_features.html
晓兵技术杂谈9_UCX统一通信框架_UCT传输层深入_服务端与客户端_连接建立主流程_rdma_零拷贝_rpc_高性能网络_c语言_常用通信语义_内存注册_云计算_GPU通信等 2023-11-18 21:16:47 https://cloud.tencent.com/developer/video/79142 https://www.bilibili.com/video/BV1pw411H78x UCX-UCT统一通信传输层2-深入: https://cloud.tencent.com/developer/article/2351953 UCX-UCT统一通信传输层1-简介: https://cloud.tencent.com/developer/article/2349389 UCX-UCT统一通信传输层3-服务端和客户端调用栈详解(及相关)_源码解读: https://cloud.tencent.com/developer/article/2353126
晓兵技术杂谈10_UCX统一通信框架_UCT传输层深入_服务端与客户端_连接建立主流程_rdma_零拷贝_rpc_高性能网络_c语言_常用通信语义_内存注册_云计算_GPU通信等 2023-11-19 21:08:10 https://cloud.tencent.com/developer/video/79146 https://www.bilibili.com/video/BV1s94y1H7jV 晓兵技术杂谈10_UCX统一通信框架_UCT传输层深入_服务端与客户端_连接建立主流程_rdma_零拷贝_rpc_高性能网络_c语言_常用通信语义_内存注册_云计算_GPU通信等 参考: UCX-UCT统一通信传输层2-深入: https://cloud.tencent.com/developer/article/2351953 UCX-UCT统一通信传输层1-简介: https://cloud.tencent.com/developer/article/2349389 UCX-UCT统一通信传输层3-服务端和客户端调用栈详解(及相关)_源码解读: https://cloud.tencent.com/developer/article/2353126

spdk

spdk技术文章: https://spdk.io/cn/articles/

吴斌_spdk_ceph_计算机: https://rootw.github.io/archive/

dpdk

dpdk官方博客: https://www.dpdk.org/blog

dpdk入门: https://www.zhihu.com/question/313210254

Dpdk/网络协议栈/vpp/OvS/DDos/NFV/虚拟化/高性能专家: https://ke.qq.com/course/5066203

源码分析: https://zzqcn.github.io/opensource/dpdk/hf-dpdk/index.html

DPDK源码分析系列: https://blog.csdn.net/apelife/category_9036462.html

DPDK环形/无锁/多消费/多生产者队列: https://dpdk-docs.readthedocs.io/en/latest/prog_guide/ring_lib.html

网络

rdma

mojo-rdma代码优化技巧(提升性能): https://www.rdmamojo.com/2013/06/08/tips-and-tricks-to-optimize-your-rdma-code/

rdma杂谈_技术原理: https://www.zhihu.com/column/c_1231181516811390976

nvidia_mlx_RDMA编程指导: https://docs.nvidia.com/networking/display/RDMAAwareProgrammingUM/Glossary

https://docs.nvidia.com/networking/display/RDMAAwareProgrammingUM/Programming+Examples+Using+IBV+Verbs

RoCE无损和有损以及Mellanox网卡新功能: https://cloud.tencent.com/developer/article/2304247

郭传雄: https://sysnetome.com/publication.html

测试杂谈(建连/抓包/性能测试等): https://mp.weixin.qq.com/s/2vyu08bsAi882YULEm_6FQ

学习可编程交换芯片原理,洞悉头部云厂商研发可编程交换机的战略布局(P4): https://mp.weixin.qq.com/s/JeEA_cfZpX5HksLmJ9IFXw

实测-基于Corundum(刚玉)架构的100G RDMA网卡: https://mp.weixin.qq.com/s/vNXfnYsDgZX3ryHjpxNFoQ

![image-20240109112747957](/Users/xb/Library/Application Support/typora-user-images/image-20240109112747957.png)

阿里RDMA通信库X-RDMA论文精读: https://mp.weixin.qq.com/s/IyFLlJ6O8B1Z99Z65Mui8w

网络的智能应该放在哪里:网卡、交换机还是 xPU - 李博杰的文章 - 知乎: https://zhuanlan.zhihu.com/p/657498193

携程技术(Linux 网络栈接收数据(RX):原理及内核实现(2022)): https://arthurchiao.art/blog/linux-net-stack-implementation-rx-zh/

mlx5以太网驱动程序: https://doc.dpdk.org/guides/nics/mlx5.html

网络性能调优(红帽): https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/6/html/performance_tuning_guide/main-network#doc-wrapper

RDMA知识点概要: https://lincolnli.blog.csdn.net/article/details/131329488, 汇总: https://lincolnli.blog.csdn.net/category_12356430.html

intel irdma参考文档: https://blog.csdn.net/bandaoyu/article/details/116203690

RDMA联盟/spec/doc等: http://www.rdmaconsortium.org/

RDMA高级/MR: https://zhuanlan.zhihu.com/p/567720023

pkey

kernel ib doc: https://docs.kernel.org/infiniband/user_mad.html

mojo pkey: https://www.rdmamojo.com/2014/04/30/partition-key-p_key/

other pkey: https://blog.csdn.net/qq_42890800/article/details/133176579

刚玉/corundum

实测RDMA(【实测】基于Corundum架构的100G RDMA网卡): https://cloud.tencent.com/developer/article/2351007

FPGA100G网卡: https://blog.csdn.net/qq_41667729/article/details/131987565

corundum文档: https://docs.corundum.io/en/latest/gettingstarted.html irdma自述文件(inter rdma): https://downloadmirror.intel.com/738730/README_irdma.txt

ofed rdma driver download: https://network.nvidia.com/products/infiniband-drivers/linux/mlnx_ofed/

FPGA/以太网卡/verilog

25G网卡工程及项目: https://blog.csdn.net/qq_41667729/article/details/131624649

ucx

UCX-UCT统一通信传输层3-服务端和客户端调用栈详解(及相关)_源码解读: https://cloud.tencent.com/developer/article/2353126

vxlan

vxlan与ovs: https://zhuanlan.zhihu.com/p/540245277

编码

google c++编码规范

https://google.github.io/styleguide/cppguide.html

计算机

计算机经典书籍一千本: https://gitee.com/mphyatyh/CS-Booksf

龙蜥社区: https://openanolis.cn/

计算机/操作系统/自学/麻省理工: https://csdiy.wiki/%E6%93%8D%E4%BD%9C%E7%B3%BB%E7%BB%9F/MIT6.S081/

PCI/PCIe

PCIE进化史: https://cloud.tencent.com/developer/article/2192556

bar(基地址寄存器): https://stackoverflow.com/questions/30190050/what-is-the-base-address-register-bar-in-pcie

配置空间: https://en.wikipedia.org/wiki/PCI_configuration_space

e1000网络驱动分析: https://developer.aliyun.com/article/559567

e1000e网卡驱动源码分析: https://dachunlv.com/2021/04/03/linux/%E4%BB%A3%E7%A0%81%E5%88%86%E6%9E%90/e1000e/

前兆网卡实现机制: https://blog.csdn.net/fengying765/article/details/6776394

ioremap函数讲解: http://blog.chinaunix.net/uid-26384989-id-5756930.html

独立于总线的设备访问: https://zhuanlan.zhihu.com/p/548249146

开发者分享 | 使用 lspci 和 setpci 调试 PCIe 问题: https://xilinx.eetrend.com/content/2020/100059215.html

使用setpci配置pci设备: https://blog.csdn.net/choumin/article/details/129864512

读写寄存器: https://blog.csdn.net/Frank_sample/article/details/118385535

PCIe5 spec: https://picture.iczhiku.com/resource/eetop/SYkDTqhOLhpUTnMx.pdf

如何为Linux写一个PCI驱动:

https://docs.kernel.org/PCI/pci.html

https://elinux.org/images/3/3a/PCI_Endpoint_drivers_in_Linux_kernel_and_How_to_write_one_.pdf

https://olegkutkov.me/2021/01/07/writing-a-pci-device-driver-for-linux/

https://zhuanlan.zhihu.com/p/372708521

sysfs-bus-pci(系统文件/总线/PCI): https://www.kernel.org/doc/Documentation/ABI/testing/sysfs-bus-pci

PCIe总线/地址空间/DMA/配置空间/协议解析: https://www.zhihu.com/question/550129239/answer/3378732759

中断

中断原理及触发: https://phmatthaus.blog.csdn.net/article/details/135621181

pci_change_irq_level, map_irq, set_irq

PCIE中断-MSI/MSI-x及代码分析: https://blog.csdn.net/yhb1047818384/article/details/106676560

使用MSI驱动: https://www.infradead.org/~mchehab/rst_conversion/PCI/msi-howto.html

DMA

DMA原理: https://zhuanlan.zhihu.com/p/547764502

kernel_how_to_use_dma: https://www.kernel.org/doc/Documentation/DMA-API-HOWTO.txt

IOMMU/DMA/虚拟地址/物理地址: https://zhuanlan.zhihu.com/p/413978652

DMA实现机制: https://cloud.tencent.com/developer/article/1628161

PCIe5 spec: https://picture.iczhiku.com/resource/eetop/SYkDTqhOLhpUTnMx.pdf pcie学习总结: https://www.zhihu.com/column/c_1633240317909524480

深入PCI与PCIe之二:软件篇: https://zhuanlan.zhihu.com/p/26244141

UEFI和BIOS探秘

老狼知乎专栏: https://www.zhihu.com/column/UEFIBlog

UEFI和BIOS区别: https://www.zhihu.com/question/21672895/answer/774538058

FPGA

赛灵思Xilinx

赛灵思软硬件生态/WIKI: https://xilinx-wiki.atlassian.net/wiki/spaces/A/overview

其他

范老师博客汇总-云计算-虚拟化-容器

腾讯云沙龙活动: https://cloud.tencent.com/developer/salon/activities

腾讯tvp(Techo TVP开发者峰会—— 数据的冰与火之歌),2021-04-24 周六 09:30-17:40: https://cloud.tencent.com/developer/salon/salon-1406

tvp文章: https://cloud.tencent.com/developer/article/1820349

rdma mellanox ofed驱动编译安装部署: https://github.com/opencurve/curve-meetup-slides/blob/9d17e0eb42df2bee0ec8c7a6c6aaf95a340028f5/PrePaper/2023/CurveBS%20RDMA%26SPDK%20%E9%83%A8%E7%BD%B2%E6%8C%87%E5%8D%97.md

智东西公开课: https://apposcmf8kb5033.pc.xiaoe-tech.com/

分布式数据库: https://www.zhihu.com/column/c_1165316856901668864

单元测试框架对比: http://www.throwtheswitch.org/comparison-of-unit-test-frameworks

分布式存储DAOS系统技术详解系列: https://cloud.tencent.com/developer/article/2344030

分布式存储DAOS相关技术视频: https://cloud.tencent.com/developer/user/5060293/video

RDMA无损和有损原理: https://cloud.tencent.com/developer/article/2304247

统一通信UCX_RDMA(系列): https://cloud.tencent.com/developer/article/2351953

nvmeof内核rdma源码分析: https://cloud.tencent.com/developer/article/2302404

内核多路径: https://cloud.tencent.com/developer/article/2299112

linux io 过程自顶向下分析 https://my.oschina.net/fileoptions/blog/3058792 linux 各种模式下 io 路径跟踪 https://my.oschina.net/fileoptions/blog/3061822

dpdk doc: http://core.dpdk.org/doc/ linux guide: http://doc.dpdk.org/guides/linux_gsg/ 程序员手册 http://doc.dpdk.org/guides/prog_guide/ 概述: http://doc.dpdk.org/guides/prog_guide/overview.html 手册: https://tonydeng.github.io/sdn-handbook/dpdk/ 示例程序: http://doc.dpdk.org/guides/sample_app_ug/intro.html dpdk技术原理及架构: https://mp.weixin.qq.com/s/YBqAmnFiNYF724JFg1cCdw

Cache的基本原理: https://zhuanlan.zhihu.com/p/102293437

ceph ceph rbd块存储: https://rootw.github.io/2018/01/RBD-all/

两万字长文,史上最全 C++ 年度总结 https://mp.weixin.qq.com/s/5w4xVcRuOVD-dT1WLlWkQA

dpu rdma https://mp.weixin.qq.com/s/56mI6M-r63IX2-AJogwIEw

vscode 调试技巧|程序不是写出来的?是调出来的!https://mp.weixin.qq.com/s/6JKIkTJn28cG66Slwv_TIA

gc https://wiki.brewlin.com/wiki/algorithm/gc-learning/GC%E7%AE%97%E6%B3%95%E5%88%86%E6%9E%90%E4%B8%8E%E5%AE%9E%E7%8E%B0/ 13 常见的 GC 算法(GC 的背景与原理): https://learn.lianglianglee.com/%E4%B8%93%E6%A0%8F/JVM%20%E6%A0%B8%E5%BF%83%E6%8A%80%E6%9C%AF%2032%20%E8%AE%B2%EF%BC%88%E5%AE%8C%EF%BC%89/13%20%E5%B8%B8%E8%A7%81%E7%9A%84%20GC%20%E7%AE%97%E6%B3%95%EF%BC%88GC%20%E7%9A%84%E8%83%8C%E6%99%AF%E4%B8%8E%E5%8E%9F%E7%90%86%EF%BC%89.md

glibc提供的malloc()的调试工具: https://www.cnblogs.com/arnoldlu/p/10827884.html

c++ c++引用计数计数及智能指针的简单实现: https://www.cnblogs.com/QG-whz/p/4777312.html

golang go语言原本: https://golang.design/under-the-hood/zh-cn/part1basic/ch01basic/go/ 内存分配: https://books.studygolang.com/GoExpertProgramming/chapter04/4.1-memory_alloc.html, https://segmentfault.com/a/1190000023866459 https://www.ququ123.top/2022/04/golang_memory_allocator/ https://blog.csdn.net/u010853261/article/details/102945046 gc: Go 垃圾回收器指南 https://colobu.com/2022/07/16/A-Guide-to-the-Go-Garbage-Collector/

RocksDB 笔记: https://alexstocks.github.io/html/rocksdb.html

makefile: https://mp.weixin.qq.com/s/nWnpZ7Rcnm-Hq4NC7ZpGfQ

软硬件角度看内存: https://mp.weixin.qq.com/s/s7t4XIDs5YvjZlc4Gt9BPQ

RPC框架汇总: https://awesomeopensource.com/projects/rpc-framework 元数据MD设计思想: https://blog.csdn.net/winux/article/details/113408049 PureFlash存储服务pfs: https://github.com/cocalele/PureFlash  PureFlash控制服务pfconductor: https://github.com/cocalele/jconductor  fio with pfbd: https://gitee.com/cocalele/fio  qemu with pfbd: https://gitee.com/cocalele/qemu  将PureFlash volume导出成iSCSI target的tcmu: https://gitee.com/cocalele/tcmu-runner  基于PureFlash的NAS文件服务: https://github.com/cocalele/ViveNAS  PureFlash k8s operator & CSI: https://github.com/cocalele/pureflash-csi

fs: ViveNAS - 一个基于LSM tree的文件存储实现 (一): https://blog.csdn.net/winux/article/details/130415715 meson: https://soaringleefighting.github.io/2021/08/12/%E3%80%90%E5%B7%A5%E7%A8%8B%E5%8C%96%E7%B3%BB%E5%88%97%E3%80%91Meson%E7%BC%96%E8%AF%91%E7%B3%BB%E7%BB%9F%E4%BD%BF%E7%94%A8%E6%96%B9%E6%B3%95/ 高级内存管理(弹性缓冲区): https://www.kernel.org/doc/gorman/html/understand/understand012.html linux内核,内存管理: https://mp.weixin.qq.com/s/Torf_AHepkgNrznG5G2NJw

滴滴Ceph分布式存储系统优化之锁优化: https://mp.weixin.qq.com/s/oWujGOLLGItu1Bv5AuO0-A C++11 并发指南五(std::condition_variable 详解): https://blog.csdn.net/Watson2016/article/details/52861094

数据结构体,算法: https://www.hello-algo.com/chapter_array_and_linkedlist/summary/#1

linux内核: https://hackmd.io/@RinHizakura nvmeof内核驱动: https://lincolnli.blog.csdn.net/article/details/132489838 无锁(lock-free): https://www.cnblogs.com/gaochundong/p/lock_free_programming.html

操作系统OS 龙蜥: https://yunqi.aliyun.com/2023/subforum/YQ-Club-0036

章夏Sean

国际热钱去哪儿了? https://www.bilibili.com/video/BV1zi4y1e7M1 什么样的企业能盈利?https://www.bilibili.com/video/BV1ep4y1R7hu

英伟达为啥这么牛?https://www.bilibili.com/video/BV1tC4y1q7GX

学习

编程面试大学: https://github.com/jwasham/coding-interview-university/blob/main/translations/README-cn.md

晓兵

https://logread.cn/post/study/summary/

博客:

https://cloud.tencent.com/developer/user/5060293/articles

https://cloud.tencent.com/developer/user/5060293/video

https://logread.cn

https://blog.csdn.net/ssbandjl

weixin: ssbandjl

公众号: 云原生云

云原生云