Code

New provisioner for RunPod#2829

Code

Merged

Michaelvll merged 88 commits intomasterskypilot-org/skypilot:masterfrom runpod-new-provisionerskypilot-org/skypilot:runpod-new-provisionerCopy head branch name to clipboardJan 13, 2024

+965-23

Commits

Commits on Nov 16, 2023

• init

  suquark

  committed
  fda9c83

  Copy full SHA for fda9c83

• remove ray

  suquark

  committed
  a972b83

  Copy full SHA for a972b83

• update config

  suquark

  committed
  95f470e

  Copy full SHA for 95f470e

• update

  suquark

  committed
  8dd39e7

  Copy full SHA for 8dd39e7

• update

  suquark

  committed
  45ef17a

  Copy full SHA for 45ef17a

• update

  suquark

  committed
  5d07647

  Copy full SHA for 5d07647

• complete bootstrapping

  suquark

  committed
  b45c09f

  Copy full SHA for b45c09f

• add start instance

  suquark

  committed
  ee7a924

  Copy full SHA for ee7a924

• fix

  suquark

  committed
  70ede43

  Copy full SHA for 70ede43

• fix

  suquark

  committed
  f8fd06d

  Copy full SHA for f8fd06d

• fix

  suquark

  committed
  83880d2

  Copy full SHA for 83880d2

• update

  suquark

  committed
  75b23ac

  Copy full SHA for 75b23ac

• wait stopping instances

  suquark

  committed
  26149b7

  Copy full SHA for 26149b7

• support normal gcp tpus first

  suquark

  committed
  03ff947

  Copy full SHA for 03ff947

• fix gcp

  suquark

  committed
  aea9d1a

  Copy full SHA for aea9d1a

• support get cluster info

  suquark

  committed
  0135eea

  Copy full SHA for 0135eea

• fix

  suquark

  committed
  a5b7537

  Copy full SHA for a5b7537

• update

  suquark

  committed
  2bb7438

  Copy full SHA for 2bb7438

• wait for instance starting

  suquark

  committed
  e525f06

  Copy full SHA for e525f06

• rename

  suquark

  committed
  0ae66fc

  Copy full SHA for 0ae66fc

• hide gcp package import

  suquark

  committed
  084170b

  Copy full SHA for 084170b

• fix

  suquark

  committed
  6940640

  Copy full SHA for 6940640

• fix

  suquark

  committed
  45f00c1

  Copy full SHA for 45f00c1

• update constants

  suquark

  committed
  9b5428f

  Copy full SHA for 9b5428f

• fix comments

  suquark

  committed
  c4bed46

  Copy full SHA for c4bed46

• remove unused methods

  suquark

  committed
  4f7fb16

  Copy full SHA for 4f7fb16

• fix comments

  suquark

  committed
  0647233

  Copy full SHA for 0647233

• sync 'config' & 'constants' with upstream, Nov 16

  suquark

  committed
  a4dbcb0

  Copy full SHA for a4dbcb0

• sync 'instace_utils' with the upstream, Nov 16

  suquark

  committed
  43bf2e3

  Copy full SHA for 43bf2e3

Commits on Nov 17, 2023

• fix typing

  suquark

  committed
  0a867ee

  Copy full SHA for 0a867ee

• parallelize provisioning

  suquark

  committed
  9ee76de

  Copy full SHA for 9ee76de

• Fix TPU node

  Michaelvll

  committed
  8e584d4

  Copy full SHA for 8e584d4

• Fix TPU NAME env for tpu node

  Michaelvll

  committed
  d713618

  Copy full SHA for d713618

Commits on Nov 18, 2023

• implement bulk provision

  suquark

  committed
  aebb197

  Copy full SHA for aebb197

Commits on Nov 20, 2023

• refactor selflink

  Michaelvll

  committed
  b5b0246

  Copy full SHA for b5b0246

• format

  Michaelvll

  committed
  b02e70d

  Copy full SHA for b02e70d

Commits on Nov 21, 2023

• reduce the sleep time for autostop

  Michaelvll

  committed
  01ba45f

  Copy full SHA for 01ba45f

Commits on Nov 26, 2023

• provisioner version refactoring

  Michaelvll

  committed
  d90c22b

  Copy full SHA for d90c22b

Commits on Nov 27, 2023

• refactor

  Michaelvll

  committed
  2df5d39

  Copy full SHA for 2df5d39

• Add logging

  Michaelvll

  committed
  06fbad6

  Copy full SHA for 06fbad6

• avoid saving the provisioner version

  Michaelvll

  committed
  aee06dd

  Copy full SHA for aee06dd

• format

  Michaelvll

  committed
  918a3c6

  Copy full SHA for 918a3c6

• format

  Michaelvll

  committed
  61bec6a

  Copy full SHA for 61bec6a

• Fix scheduling field in config

  Michaelvll

  committed
  0f07964

  Copy full SHA for 0f07964

• format

  Michaelvll

  committed
  279b301

  Copy full SHA for 279b301

• fix public key content

  Michaelvll

  committed
  625c510

  Copy full SHA for 625c510

• Fix provisioner version for azure

  Michaelvll

  committed
  ebe92f5

  Copy full SHA for ebe92f5

• Use ray port from head node for workers

  Michaelvll

  committed
  54a87ef

  Copy full SHA for 54a87ef

• format

  Michaelvll

  committed
  b1896ac

  Copy full SHA for b1896ac

• fix ray_port

  Michaelvll

  committed
  15e193d

  Copy full SHA for 15e193d

• fix smoke tests

  Michaelvll

  committed
  504b7e7

  Copy full SHA for 504b7e7

• shorter sleep time

  Michaelvll

  committed
  b6ba235

  Copy full SHA for b6ba235

• refactor status refresh version

  Michaelvll

  committed
  497c438

  Copy full SHA for 497c438

Commits on Dec 1, 2023

• Merge branch 'master' of github.com:skypilot-org/skypilot into new_provisioner_gcp

  Michaelvll

  committed
  061e924

  Copy full SHA for 061e924

• Use new provisioner to launch runpod to avoid issue with ray autoscaler on head

  Show description for 21a3692
  Michaelvll

  committed
  21a3692

  Copy full SHA for 21a3692

• Add wait for the instances to be ready

  Michaelvll

  committed
  533afb0

  Copy full SHA for 533afb0

• fix setup

  Michaelvll

  committed
  389bd21

  Copy full SHA for 389bd21

• Retry and give for getting internal IP

  Michaelvll

  committed
  b797e3b

  Copy full SHA for b797e3b

• comment

  Michaelvll

  committed
  4ef8a59

  Copy full SHA for 4ef8a59

• Remove internal IP

  Michaelvll

  committed
  184c1de

  Copy full SHA for 184c1de

• use external IP

  Show description for 8162e84
  Michaelvll

  committed
  8162e84

  Copy full SHA for 8162e84

• fix ssh port

  Michaelvll

  committed
  eba6787

  Copy full SHA for eba6787

• Unsupported feature

  Michaelvll

  committed
  9bbf3c2

  Copy full SHA for 9bbf3c2

• typo

  Michaelvll

  committed
  472f1f6

  Copy full SHA for 472f1f6

Commits on Jan 1, 2024

• Merge branch 'master' of github.com:skypilot-org/skypilot into runpod-new-provisioner

  Michaelvll

  committed
  9f29cbf

  Copy full SHA for 9f29cbf

• fix ssh ports

  Michaelvll

  committed
  0b582ae

  Copy full SHA for 0b582ae

• rename var

  Michaelvll

  committed
  106eefa

  Copy full SHA for 106eefa

Commits on Jan 2, 2024

• format

  Michaelvll

  committed
  8e8501f

  Copy full SHA for 8e8501f

• Fix cloud unsupported resources

  Michaelvll

  committed
  1e92cfd

  Copy full SHA for 1e92cfd

Commits on Jan 5, 2024

• Runpod update name mapping (#2945)

  landscapepainter

  authored
  fa07c72

  Copy full SHA for fa07c72

Commits on Jan 7, 2024

• Avoid using GpuInfo

  Michaelvll

  committed
  3489df5

  Copy full SHA for 3489df5

• Merge branch 'master' of github.com:skypilot-org/skypilot into runpod-new-provisioner

  Michaelvll

  committed
  5bbfc3d

  Copy full SHA for 5bbfc3d

• fix all_regions

  Michaelvll

  committed
  1859101

  Copy full SHA for 1859101

• Fix runpod list accelerators

  Michaelvll

  committed
  ed955df

  Copy full SHA for ed955df

• format

  Michaelvll

  committed
  045bab6

  Copy full SHA for 045bab6

Commits on Jan 8, 2024

• revert to GpuInfo

  Michaelvll

  committed
  9630832

  Copy full SHA for 9630832

• Fix get_feasible_launchable_resources

  Michaelvll

  committed
  97165a4

  Copy full SHA for 97165a4

• Add error

  Michaelvll

  committed
  f527545

  Copy full SHA for f527545

• Fix optimizer random_dag for feature check

  Michaelvll

  committed
  680beca

  Copy full SHA for 680beca

Commits on Jan 12, 2024

• Merge branch 'master' of github.com:skypilot-org/skypilot into runpod-new-provisioner

  Michaelvll

  committed
  497039c

  Copy full SHA for 497039c

Commits on Jan 13, 2024

• address comments

  Michaelvll

  committed
  8193931

  Copy full SHA for 8193931

• remove test code

  Michaelvll

  committed
  e5631f3

  Copy full SHA for e5631f3

• format

  Michaelvll

  committed
  7399d53

  Copy full SHA for 7399d53

• Add type hints

  Michaelvll

  committed
  9342d20

  Copy full SHA for 9342d20

• format

  Michaelvll

  committed
  1955380

  Copy full SHA for 1955380

• format

  Michaelvll

  committed
  8b48f32

  Copy full SHA for 8b48f32

• fix keyerror

  Michaelvll

  committed
  07498d9

  Copy full SHA for 07498d9

• Address comments

  Michaelvll

  committed
  b11446e

  Copy full SHA for b11446e