Граф коммитов

104 Коммитов

Автор SHA1 Сообщение Дата
Fred Park ace0dde416 Tag for 2.5.2 release 2017-02-23 08:06:27 -08:00
Fred Park 7dd02b3c27 Automatic path sub for Gluster/AzStorage xfer
- Resolves #37
2017-02-23 07:51:49 -08:00
Fred Park af1e03cfa8 Add troubleshooting guide
- Minor convoy.batch fixes
2017-02-22 20:34:20 -08:00
Fred Park edacb92590 Add Chainer recipes 2017-02-22 19:37:03 -08:00
Fred Park 23ebd58c0d Update recipes for TensorFlow 1.0.0 2017-02-17 13:54:18 -08:00
Fred Park 4eea944bb3 Tag for 2.5.1 release 2017-02-01 11:06:26 -08:00
Fred Park 78fad1c3e3 Add support for task retention time
- Resolves #30
2017-01-31 09:40:16 -08:00
Fred Park 25dcc983ef Fix unencrypted task file mover delimiter issue
- Resolves #29
2017-01-30 15:06:32 -08:00
Fred Park 4ce2689aca Install all intel mpi rpms on SLES-HPC 2017-01-26 10:44:43 -08:00
Fred Park 0fe858c14e Add FAQ and fix autogen task id rollover
- Resolves #27
2017-01-26 08:30:53 -08:00
Fred Park 270ef0c7b1 Fix Docker tmpdir
- Fix typo with ev secret id ref to keyvault
- Add travis py36 env
2017-01-24 14:43:44 -08:00
Fred Park fa4e1f847c Add env var secret id support
- Tag for 2.5.0 release
- Resolves #12
- Partially resolves #15
2017-01-19 10:16:42 -08:00
Fred Park 040a068265 Various fixes
- This resolves #13 and resolves #16
2017-01-19 10:16:42 -08:00
Fred Park 9b6dbef19f Add task dependency id range support 2017-01-12 09:30:41 -08:00
Fred Park c95520eaea Tag for 2.4.0 release
- Update KeyVault docs with Azure CLI 2.0 commands
- Resolves #10
2017-01-11 09:22:45 -08:00
Fred Park fc583e89da Update CHANGELOG links for 2.3.1 2017-01-03 10:04:40 -08:00
Fred Park ae7e5df410 Tag for 2.3.1 release
- Update some docs
2017-01-03 08:51:19 -08:00
Fred Park d9bf6c92da Add nvidia-docker support to ssh tunnel 2016-12-15 11:05:52 -08:00
Fred Park 57b47b353f Add pool ssh command, resolves #9
- Make the ssh docker tunnel script much easier to use
- Add an ssh guide to docs
2016-12-15 07:39:04 -08:00
Fred Park f00f222877 Infiniband settings changes
- Add CNTK ib recipe
- Update READMEs to remove GPU preview notes
- Tag for 2.2.0 release
2016-12-09 11:34:21 -08:00
Fred Park 5baf61d8f4 Fix SAS key and KeyError masking in data movement 2016-11-30 14:41:39 -08:00
Fred Park 8f0fa2f446 Tag for 2.1.0 release
- Pass version to nodeprep and pull backend docker images by version
2016-11-30 08:27:46 -08:00
Fred Park 28732f2aea Add listskus subcommand
- Update docs for envvars
2016-11-29 15:31:23 -08:00
Fred Park 6548dc3508 Fix cascade run exit code not propagating 2016-11-28 14:15:53 -08:00
Fred Park 0d7814a58f Add envvar support for certain config options
- Some refactoring in install.sh script
- Update docs
2016-11-28 11:02:09 -08:00
Fred Park 8577232349 Tag for 2.0.0 release 2016-11-23 09:06:37 -08:00
Fred Park fe0403de9a Update MXNet GPU docker image 2016-11-22 14:27:33 -08:00
Fred Park c047b522c3 Update CNTK docker images to 2.0beta4
- Fix termtasks for multi-instance tasks with named containers
2016-11-22 00:16:16 -08:00
Fred Park 3d3fd99b3d Update Caffe recipes and images 2016-11-21 13:30:03 -08:00
Fred Park 44080c123a Prepend job id to Docker container names
- Update TensorFlow to 0.11.0 and custom compile to add compute/sm 3.7
2016-11-20 14:57:45 -08:00
Fred Park 8cb2ba9583 Allow GPU property to be optional for NC VMs
- Update all GPU compute recipes to omit gpu driver
2016-11-20 09:00:19 -08:00
Fred Park 453ae98a65 Terminate cascade on thread failures 2016-11-19 10:39:58 -08:00
Fred Park c7744f95bf Support for internet accessible private registries 2016-11-19 09:00:01 -08:00
Fred Park 4399dbf4db Tag for 2.0.0rc3 release
- Fix flake8 issues
2016-11-14 11:10:59 -08:00
Fred Park 0ae05f2d84 Add CUDA_CACHE_ vars for GPU tasks 2016-11-13 11:42:56 -08:00
Fred Park ff43949d0e Add Keras+Theano recipes 2016-11-13 01:43:01 -08:00
Fred Park b8fcdede8f Add --tail option for jobs add
- Simplify quickstart with --tail
2016-11-12 22:35:36 -08:00
Fred Park e6593281eb Refactor direct config access out of data/storage 2016-11-12 12:35:56 -08:00
Fred Park 03ced70c38 Continue settings refactor
- Credentials
- Some of global config
2016-11-11 21:08:58 -08:00
Fred Park e700ee05b7 Add docker login prior to image update
- Move docker hub creds to credentials json
- Begin refactor of configuration settings retrieval
2016-11-11 09:30:14 -08:00
Fred Park da573524de Preliminary steps for ACR support
- Fix update docker images with private registry
- Automatically clean dangling image refs on update
- Remove private registry file/image id support
- Refactor fleet initialization steps to one entry point
- Simplify shipyard context init
2016-11-10 09:48:00 -08:00
Fred Park f8ac2ccc40 Add support for single node direct ingress
- Add missing support for relative_destination_path in single node
transfers
2016-11-09 15:38:19 -08:00
Fred Park ec60383723 Add pool udi command and listtasks job scoping
- Update docs
2016-11-09 11:59:01 -08:00
Fred Park bbd948fdab Add jobid/taskid scoping to data listfiles
- Update some docs
2016-11-08 15:06:40 -08:00
Fred Park 47f12b0bcb Fix pool resize down with wait 2016-11-07 18:34:35 -08:00
Fred Park 9db24df307 Add relative destination path
- Check for vm_count for glusterfs setup
2016-11-07 10:23:43 -08:00
Fred Park 4742b77b17 Minor updates 2016-11-04 07:25:55 -07:00
Fred Park 162219f1b8 Add HPCG benchmark recipe
- Prepare for 2.0.0rc2 release
2016-11-02 08:47:39 -07:00
Fred Park 9c604d2572 More comprehensive pool resize logic with SSH user 2016-11-01 17:10:03 -07:00
Fred Park 155fabfb8d Add HPL benchmark recipe
- Add SSH users on pool resize
- More improvements to install doc
2016-11-01 14:14:22 -07:00