measurementlab/msak Docker 镜像 - 轩辕镜像
measurementlab/msakmeasurementlab
0 次下载
Measurements Swiss Army Knife
 describes the throughput1 and latency1 protocols.
msak-server- is the MSAK servermsak-client- is a full reference client for the throughput1 protocol
Additional reference clients are also available:
minimal-download- is a minimal download-only, reference cleint for the throughput1 protocolmsak-latency- is a reference client for the latency1 protocol
Server
To build the server and run locally without TLS certificates:
sh$ go install github.com/m-lab/msak/cmd/msak-server@latest ... $ msak-server 2024/01/04 17:41:01 INFO <msak-server/server.go:117> About to listen for ws tests endpoint=:8080 2024/01/04 17:41:01 INFO <latency1/latency1.go:286> Accepting UDP packets...
Clients
To build the client and target the local server:
sh$ go install github.com/m-lab/msak/cmd/msak-client@latest ... $ msak-client -duration=2s -streams=1 -server localhost:8080 -scheme ws Starting download stream (server: localhost:8080) Connected to ws://localhost:8080/throughput/v1/download?bytes=0&... Elapsed: 0.10s, Goodput: 0.000000 Mb/s, MinRTT: 0 Elapsed: 0.20s, Goodput: 0.000000 Mb/s, MinRTT: 0 Elapsed: 0.30s, Goodput: 0.000000 Mb/s, MinRTT: 0 Elapsed: 0.40s, Goodput: 0.000000 Mb/s, MinRTT: 0 Elapsed: 0.50s, Goodput: ***.574715 Mb/s, MinRTT: 0 Elapsed: 0.60s, Goodput: ***.580735 Mb/s, MinRTT: 0 Elapsed: 0.70s, Goodput: ***.483102 Mb/s, MinRTT: 0 Elapsed: 0.80s, Goodput: ***.*** Mb/s, MinRTT: 0 Elapsed: 0.90s, Goodput: 21180.762969 Mb/s, MinRTT: 0 Elapsed: 1.00s, Goodput: 21419.807987 Mb/s, MinRTT: 0 Stream 0 complete (server localhost:8080) Starting upload stream (server: localhost:8080) Connected to ws://localhost:8080/throughput/v1/upload?bytes=0&... Elapsed: 0.10s, Goodput: 0.000000 Mb/s, MinRTT: 0 Elapsed: 0.20s, Goodput: ***.*** Mb/s, MinRTT: 0 Elapsed: 0.30s, Goodput: ***.665589 Mb/s, MinRTT: 0 Elapsed: 0.40s, Goodput: ***.343408 Mb/s, MinRTT: 0 Elapsed: 0.50s, Goodput: ***.*** Mb/s, MinRTT: 0 Elapsed: 0.60s, Goodput: 23205.208525 Mb/s, MinRTT: 0 Elapsed: 0.70s, Goodput: ***.442053 Mb/s, MinRTT: 0 Elapsed: 0.80s, Goodput: 21784.415044 Mb/s, MinRTT: 0 Elapsed: 0.90s, Goodput: ***.955232 Mb/s, MinRTT: 0 Elapsed: 1.00s, Goodput: ***.*** Mb/s, MinRTT: 0 Stream 0 complete (server localhost:8080)
To build the minimal client and target a local or remote server:
sh$ go install github.com/m-lab/msak/cmd/minimal-download@latest ... # Local $ minimal-download -duration 1s -server.url ws://localhost:8080/throughput/v1/download minimal-download -duration 1s -server.url ws://localhost:8080/throughput/v1/download Download server #1 - rate 35260.80 Mbps, rtt 0.03ms, elapsed 0.2076s, application r/w: 0/918552576, network r/w: 0/917513214 kernel* r/w: 538/914903145 Download server #1 - rate 34191.58 Mbps, rtt 0.03ms, elapsed 0.5168s, application r/w: 0/2213545117, network r/w: 0/2212518109 kernel* r/w: 538/2208912708 Download server #1 - rate 33703.03 Mbps, rtt 0.05ms, elapsed 0.9170s, application r/w: 0/3868199075, network r/w: 0/3867187851 kernel* r/w: 538/3863293895 Download server #1 - rate 33591.42 Mbps, rtt 0.03ms, elapsed 1.0005s, application r/w: 0/4203744426, network r/w: 0/4203784992 kernel* r/w: 538/4200858760 Download client #1 - Avg 33552.56 Mbps, MinRTT 0.00ms, elapsed 1.0023s, application r/w: 0/4203745461 # Remote with time limit. $ minimal-download -duration 1s Download server #1 - rate 239.68 Mbps, rtt 13.96ms, elapsed 0.1014s, application r/w: 0/6815744, network r/w: 0/6400516 kernel* r/w: 1304/3039466 Download server #1 - rate 375.56 Mbps, rtt 15.13ms, elapsed 0.2024s, application r/w: 0/***, network r/w: 0/*** kernel* r/w: 1304/9503338 Download server #1 - rate 429.03 Mbps, rtt 19.15ms, elapsed 0.3034s, application r/w: 0/***, network r/w: 0/*** kernel* r/w: 1304/*** Download server #1 - rate 473.54 Mbps, rtt 15.88ms, elapsed 0.5237s, application r/w: 0/35654810, network r/w: 0/34737910 kernel* r/w: 1304/30997450 Download server #1 - rate 487.79 Mbps, rtt 15.34ms, elapsed 0.6464s, application r/w: 0/42995877, network r/w: 0/42564857 kernel* r/w: 1304/39414674 Download server #1 - rate 499.34 Mbps, rtt 17.36ms, elapsed 1.0154s, application r/w: 0/66065584, network r/w: 0/66158482 kernel* r/w: 1304/63380522 Download client #1 - Avg 502.43 Mbps, MinRTT 4.11ms, elapsed 1.0520s, application r/w: 0/66066624 # Remote with bytes limit. $ minimal-download -bytes=*** Download server #1 - rate 8.24 Mbps, rtt 12.17ms, elapsed 0.0128s, application r/w: 0/***, network r/w: 0/*** kernel* r/w: 1309/*** Download client #1 - Avg 30.51 Mbps, MinRTT 10.99ms, elapsed 0.0433s, application r/w: 0/***
Every TCP connection has performance metrics accessible from the server end and
the client end. The minimal-download client reports metrics generated by the
server side and concludes with client side average performance. Client side
performance is comparable to what a user (or user application) would see.
Measurements
The application, network, and kernel metrics may differ to the degree that some data is buffered, or includes added headers, or is traversing the physical network itself. For example, after a websocket write and before TLS/WebSocket headers are added (application), after TLS/WebSocket headers are added before being sent to the Linux kernel (network), or after the Linux kernel sends over the physical network and before the remote client acknowledges the bytes received (kernel).
The maximum difference between the application and network sent sizes should be
equal to the spec.MaxScaledMessageSize + WebSocket/TLS headers size, which
should typically be below 1MB, and the maximum difference between the network
and kernel sent sizes should equal the Linux kernel buffers plus the network's
bandwidth delay product. Typically values range between 64k and 4MB or more.
measurementlab/prometheus-bigquery-exporter
measurementlab
将BigQuery查询结果转换为Prometheus指标的导出器。
100K+ pulls
上次更新：8 个月前
measurementlab/uuid
measurementlab
暂无描述
50K+ pulls
上次更新：1 年前
measurementlab/traceroute-caller
measurementlab
暂无描述
50K+ pulls
上次更新：1 年前
measurementlab/pusher
measurementlab
Archive experiment data from M-Lab experiments into tarfiles uploaded to Google Cloud Storage.
50K+ pulls
上次更新：1 年前
measurementlab/tcp-info
measurementlab
Provides tcp instrumentation as a service.
150K+ pulls
上次更新：1 年前
measurementlab/uuid-annotator
measurementlab
暂无描述
50K+ pulls
上次更新：1 年前
轩辕镜像配置手册
探索更多轩辕镜像的使用方法，找到最适合您系统的配置方式
登录仓库拉取
通过 Docker 登录认证访问私有仓库
Linux
在 Linux 系统配置镜像服务
Windows/Mac
在 Docker Desktop 配置镜像
Docker Compose
Docker Compose 项目配置
K8s Containerd
Kubernetes 集群配置 Containerd
K3s
K3s 轻量级 Kubernetes 镜像加速
Dev Containers
VS Code Dev Containers 配置
MacOS OrbStack
MacOS OrbStack 容器配置
宝塔面板
在宝塔面板一键配置镜像
群晖
Synology 群晖 NAS 配置
飞牛
飞牛 fnOS 系统配置镜像
极空间
极空间 NAS 系统配置服务
爱快路由
爱快 iKuai 路由系统配置
绿联
绿联 NAS 系统配置镜像
威联通
QNAP 威联通 NAS 配置
Podman
Podman 容器引擎配置
Singularity/Apptainer
HPC 科学计算容器配置
其他仓库配置
ghcr、Quay、nvcr 等镜像仓库
专属域名拉取
无需登录使用专属域名
需要其他帮助？请查看我们的 常见问题Docker 镜像访问常见问题解答 或 提交工单
镜像拉取常见问题
轩辕镜像免费版与专业版有什么区别？
免费版仅支持 Docker Hub 访问，不承诺可用性和速度；专业版支持更多镜像源，保证可用性和稳定速度，提供优先客服响应。
轩辕镜像支持哪些镜像仓库？
专业版支持 docker.io、gcr.io、ghcr.io、registry.k8s.io、nvcr.io、quay.io、mcr.microsoft.com、docker.elastic.co 等；免费版仅支持 docker.io。
流量耗尽错误提示
当返回 402 Payment Required 错误时，表示流量已耗尽，需要充值流量包以恢复服务。
410 错误问题
通常由 Docker 版本过低导致，需要升级到 20.x 或更高版本以支持 V2 协议。
manifest unknown 错误
先检查 Docker 版本，版本过低则升级；版本正常则验证镜像信息是否正确。
镜像拉取成功后，如何去掉轩辕镜像域名前缀？
使用 docker tag 命令为镜像打上新标签，去掉域名前缀，使镜像名称更简洁。
用户好评
来自真实用户的反馈，见证轩辕镜像的优质服务
oldzhang
运维工程师
Linux服务器
5
"Docker访问体验非常流畅，大镜像也能快速完成下载。"
咨询镜像拉取问题请 提交工单，官方技术交流群：1072982923
轩辕镜像面向开发者与科研用户，提供开源镜像的搜索和访问支持。所有镜像均来源于原始仓库，本站不存储、不修改、不传播任何镜像内容。