Граф коммитов

2561 Коммитов

Автор SHA1 Сообщение Дата
Yifan Xiong 2fc4b568b7 Update README for more examples
Add a subsection in job-tutorial/README.md for more examples and list
all job exampels.
2017-11-24 14:29:18 +08:00
Yifan Xiong 573f50b982 Fix yarn container entrypoint for nodes without gpu
Support nodes without gpu.
2017-11-24 13:25:49 +08:00
Yifan Xiong b5805b2dbc Add job examples
Add TensorFlow ImageNet trainning and TensorBoard visualization
examples.
2017-11-24 13:10:00 +08:00
fan yang 3c438e3c99
Merge pull request #45 from Microsoft/rename
rename all aii/AII to pai/PAI
2017-11-24 13:06:21 +08:00
fanyangCS a4cb98ec3b rename all aii/AII to pai/PAI 2017-11-24 12:15:03 +08:00
fan yang d0bda220b3
Merge pull request #44 from Microsoft/readme
minor fix
2017-11-24 12:09:36 +08:00
YundongYe 72abaca6fe
Merge pull request #43 from Microsoft/zk_config_change_for_zkcluster
Zk config change for zkcluster
2017-11-24 11:58:38 +08:00
fanyangCS c5e89a7844 minor fix 2017-11-24 11:53:48 +08:00
yuye@microsoft.com fe28bb3184 add comments on clusterconfig about zk. 2017-11-24 11:38:44 +08:00
Yifan Xiong 55734973e5
Merge pull request #42 from Microsoft/xiongyf/deploy-docker-non-cred
Handle non auth docker registry
2017-11-24 11:11:10 +08:00
yuye@microsoft.com ecffbeb40b remove unused env in zookeeper.yaml. 2017-11-24 11:04:20 +08:00
yuye@microsoft.com 24ebf7b48e Fix a bug. 2017-11-24 10:56:09 +08:00
yuye@microsoft.com 6a15a977ac Fix a bug. 2017-11-24 10:53:07 +08:00
yuye@microsoft.com fc5f862ab0 Add zookeeper HA. 2017-11-24 10:43:47 +08:00
Yifan Xiong 7b78b6d740 Handle non auth docker registry
Support docker registry which doesn't require authentication in deployment.
2017-11-23 20:45:01 +08:00
fan yang 48c9bfe38e
Merge pull request #41 from Microsoft/readme
update readme, rename examples to job-tutorial
2017-11-23 20:28:03 +08:00
fanyangCS 4c0e002b3b put contributing to the end 2017-11-23 20:25:41 +08:00
fanyangCS ded33c2ce8 fix merge 2017-11-23 20:18:43 +08:00
Yifan Xiong d21bdb91ef
Merge pull request #40 from Microsoft/xiongyf/update-doc
Update docs according to changes in #37
2017-11-23 19:57:45 +08:00
fanyangCS e5f868a33d update readme, rename examples to job-tutorial 2017-11-23 19:49:41 +08:00
Yifan Xiong f8632b9ce7 Update docs according to changes in #37
Update readme and examples according to changes in #37.
2017-11-23 19:40:28 +08:00
YundongYe 1ea7614fec
Merge pull request #38 from Microsoft/config_map_docker_cred
Config map docker cred
2017-11-23 18:07:47 +08:00
yuye@microsoft.com 060a848ce9 remove the comment to avoid confusing user. 2017-11-23 18:06:40 +08:00
fan yang e70b1589f9
Merge pull request #37 from Microsoft/example-readme
refine readme
2017-11-23 17:57:52 +08:00
fanyangCS facb163a74 refine readme 2017-11-23 16:05:34 +08:00
yuye@microsoft.com 07c75483a2 Update the main function of deploy.py and add config.json.template. 2017-11-23 15:54:54 +08:00
yuye@microsoft.com 4b4b5284c5 Change the way to get config.json. Because of the permission problem. 2017-11-23 15:52:53 +08:00
yuye@microsoft.com 59f5eaf949 Fix issue. Add function in main. 2017-11-23 15:38:23 +08:00
yuye@microsoft.com 3d20f97122 Update hadoop-nm yaml file. Add docker cred mount point. 2017-11-23 15:30:19 +08:00
yuye@microsoft.com 1b79cd91fb add copy feature to deploy.py. And update service.yaml and configmap-create.sh 2017-11-23 15:24:35 +08:00
fan yang 4e85af3910
Merge pull request #36 from Microsoft/service-deployment
remove the assumption on eth0
2017-11-23 11:58:38 +08:00
fanyangCS b7e63c1588 remove the assumption on eth0 2017-11-23 11:55:53 +08:00
Yifan Xiong 01a3fddaea
Merge pull request #34 from Microsoft/xiongyf/deploy-set-hosts
[Deployment] Edit /etc/hosts in k8s Deployment
2017-11-23 10:55:54 +08:00
YundongYe 0bcaa169de
Merge pull request #35 from Microsoft/update_service_yaml
Remove uncorrect comments
2017-11-23 09:56:36 +08:00
yuye@microsoft.com b5b089ae3b Remove uncorrect comments 2017-11-23 09:55:42 +08:00
fan yang ffc0cb06c1
Merge pull request #32 from Microsoft/webportal
web portal readme update
2017-11-22 21:36:22 +08:00
fan yang 15c2655761
Merge pull request #33 from Microsoft/rest-readme
rest server readme
2017-11-22 21:35:53 +08:00
Yifan Xiong 5ccb9b58e7
Merge pull request #29 from Microsoft/xiongyf/docker-runtime-env
[Examples] Fix Hadoop ENV Issue in Base Image
2017-11-22 20:26:33 +08:00
Yifan Xiong 6d0c239a82 Edit /etc/hosts in k8s deployment
According to [Hadoop
Wiki](https://wiki.apache.org/hadoop/ConnectionRefused) and #25, edit /etc/hosts during k8s deployment:
- Change `127.0.0.1` line to `127.0.0.1 localhost`, append if it doesn't
exist
- Comment `127.0.1.1` line
- Change host ip line to `hostip hostname`, append if it doesn't exist
2017-11-22 20:16:45 +08:00
fanyangCS ca9e955c34 rest server readme 2017-11-22 20:06:54 +08:00
YundongYe 59fe82eeee
Merge pull request #31 from Microsoft/yuye/host_config_feature
Yuye/host config feature
2017-11-22 19:24:18 +08:00
fanyangCS 711a2adc5c web portal readme update 2017-11-22 19:01:19 +08:00
yuye@microsoft.com ad7a0b2c10 Fix bug in script 2017-11-22 17:44:05 +08:00
yuye@microsoft.com 9093f409a2 Make the script executable 2017-11-22 03:59:04 -05:00
yuye@microsoft.com 81ebd40697 Add the script header to the script. 2017-11-22 16:54:35 +08:00
YundongYe e4df423fe9
Merge pull request #30 from Microsoft/hadoop_image_driver_env_fix
Fix the issue of nvidia env problem.
2017-11-22 16:47:59 +08:00
yuye@microsoft.com c6d4cce0ae Fix the issue of nvidia env problem. 2017-11-22 16:43:21 +08:00
yuye@microsoft.com 344bc0fe1e Add script in the base-image. Add change the generate script of nm and datanode. 2017-11-22 16:34:04 +08:00
yuye@microsoft.com 80837c4767 Add the script to update host-configuration 2017-11-22 16:24:13 +08:00
yuye@microsoft.com d36263f170 Update datanode's hdfs-site.xml and nodemanager's yarn-site.xml. And add the mount point of host config in the yaml of them. 2017-11-22 16:10:32 +08:00