skypilot-nightly 1.0.0.dev20241217__tar.gz → 1.0.0.dev20241219__tar.gz

{skypilot_nightly-1.0.0.dev20241217/skypilot_nightly.egg-info → skypilot_nightly-1.0.0.dev20241219}/PKG-INFO

@@ -1,13 +1,13 @@
 Metadata-Version: 2.1
 Name: skypilot-nightly
-Version: 1.0.0.dev20241217
+Version: 1.0.0.dev20241219
 Summary: SkyPilot: An intercloud broker for the clouds
 Author: SkyPilot Team
 License: Apache 2.0
 Project-URL: Homepage, https://github.com/skypilot-org/skypilot
 Project-URL: Issues, https://github.com/skypilot-org/skypilot/issues
 Project-URL: Discussion, https://github.com/skypilot-org/skypilot/discussions
-Project-URL: Documentation, https://skypilot.readthedocs.io/en/latest/
+Project-URL: Documentation, https://docs.skypilot.co/
 Classifier: Programming Language :: Python :: 3.7
 Classifier: Programming Language :: Python :: 3.8
 Classifier: Programming Language :: Python :: 3.9
@@ -145,7 +145,7 @@ Requires-Dist: pyvmomi==8.0.1.0.2; extra == "all"
 </p>
 
 <p align="center">
-  <a href="https://skypilot.readthedocs.io/en/latest/">
+  <a href="https://docs.skypilot.co/">
     <img alt="Documentation" src="https://readthedocs.org/projects/skypilot/badge/?version=latest">
   </a>
 
@@ -182,7 +182,7 @@ Requires-Dist: pyvmomi==8.0.1.0.2; extra == "all"
   <summary>Archived</summary>
 
 - [Jul 2024] [**Finetune**](./llm/llama-3_1-finetuning/) and [**serve**](./llm/llama-3_1/) **Llama 3.1** on your infra
-- [Apr 2024] Serve and finetune [**Llama 3**](https://skypilot.readthedocs.io/en/latest/gallery/llms/llama-3.html) on any cloud or Kubernetes: [**example**](./llm/llama-3/)
+- [Apr 2024] Serve and finetune [**Llama 3**](https://docs.skypilot.co/en/latest/gallery/llms/llama-3.html) on any cloud or Kubernetes: [**example**](./llm/llama-3/)
 - [Mar 2024] Serve and deploy [**Databricks DBRX**](https://www.databricks.com/blog/introducing-dbrx-new-state-art-open-llm) on your infra: [**example**](./llm/dbrx/)
 - [Feb 2024] Speed up your LLM deployments with [**SGLang**](https://github.com/sgl-project/sglang) for 5x throughput on SkyServe: [**example**](./llm/sglang/)
 - [Dec 2023] Using [**LoRAX**](https://github.com/predibase/lorax) to serve 1000s of finetuned LLMs on a single instance in the cloud: [**example**](./llm/lorax/)
@@ -199,17 +199,17 @@ Requires-Dist: pyvmomi==8.0.1.0.2; extra == "all"
 SkyPilot is a framework for running AI and batch workloads on any infra, offering unified execution, high cost savings, and high GPU availability.
 
 SkyPilot **abstracts away infra burdens**:
-- Launch [dev clusters](https://skypilot.readthedocs.io/en/latest/examples/interactive-development.html), [jobs](https://skypilot.readthedocs.io/en/latest/examples/managed-jobs.html), and [serving](https://skypilot.readthedocs.io/en/latest/serving/sky-serve.html) on any infra
+- Launch [dev clusters](https://docs.skypilot.co/en/latest/examples/interactive-development.html), [jobs](https://docs.skypilot.co/en/latest/examples/managed-jobs.html), and [serving](https://docs.skypilot.co/en/latest/serving/sky-serve.html) on any infra
 - Easy job management: queue, run, and auto-recover many jobs
 
 SkyPilot **supports multiple clusters, clouds, and hardware** ([the Sky](https://arxiv.org/abs/2205.07147)):
 - Bring your reserved GPUs, Kubernetes clusters, or 12+ clouds
-- [Flexible provisioning](https://skypilot.readthedocs.io/en/latest/examples/auto-failover.html) of GPUs, TPUs, CPUs, with auto-retry
+- [Flexible provisioning](https://docs.skypilot.co/en/latest/examples/auto-failover.html) of GPUs, TPUs, CPUs, with auto-retry
 
 SkyPilot **cuts your cloud costs & maximizes GPU availability**:
-* [Autostop](https://skypilot.readthedocs.io/en/latest/reference/auto-stop.html): automatic cleanup of idle resources
-* [Managed Spot](https://skypilot.readthedocs.io/en/latest/examples/managed-jobs.html): 3-6x cost savings using spot instances, with preemption auto-recovery
-* [Optimizer](https://skypilot.readthedocs.io/en/latest/examples/auto-failover.html): 2x cost savings by auto-picking the cheapest & most available infra
+* [Autostop](https://docs.skypilot.co/en/latest/reference/auto-stop.html): automatic cleanup of idle resources
+* [Managed Spot](https://docs.skypilot.co/en/latest/examples/managed-jobs.html): 3-6x cost savings using spot instances, with preemption auto-recovery
+* [Optimizer](https://docs.skypilot.co/en/latest/examples/auto-failover.html): 2x cost savings by auto-picking the cheapest & most available infra
 
 SkyPilot supports your existing GPU, TPU, and CPU workloads, with no code changes.
 
@@ -218,29 +218,29 @@ Install with pip:
 # Choose your clouds:
 pip install -U "skypilot[kubernetes,aws,gcp,azure,oci,lambda,runpod,fluidstack,paperspace,cudo,ibm,scp]"
 ```
-To get the latest features and fixes, use the nightly build or [install from source](https://skypilot.readthedocs.io/en/latest/getting-started/installation.html):
+To get the latest features and fixes, use the nightly build or [install from source](https://docs.skypilot.co/en/latest/getting-started/installation.html):
 ```bash
 # Choose your clouds:
 pip install "skypilot-nightly[kubernetes,aws,gcp,azure,oci,lambda,runpod,fluidstack,paperspace,cudo,ibm,scp]"
 ```
 
-[Current supported infra](https://skypilot.readthedocs.io/en/latest/getting-started/installation.html) (Kubernetes; AWS, GCP, Azure, OCI, Lambda Cloud, Fluidstack, RunPod, Cudo, Paperspace, Cloudflare, Samsung, IBM, VMware vSphere):
+[Current supported infra](https://docs.skypilot.co/en/latest/getting-started/installation.html) (Kubernetes; AWS, GCP, Azure, OCI, Lambda Cloud, Fluidstack, RunPod, Cudo, Paperspace, Cloudflare, Samsung, IBM, VMware vSphere):
 <p align="center">
   <img alt="SkyPilot" src="https://raw.githubusercontent.com/skypilot-org/skypilot/master/docs/source/images/cloud-logos-light.png" width=85%>
 </p>
 
 
 ## Getting Started
-You can find our documentation [here](https://skypilot.readthedocs.io/en/latest/).
-- [Installation](https://skypilot.readthedocs.io/en/latest/getting-started/installation.html)
-- [Quickstart](https://skypilot.readthedocs.io/en/latest/getting-started/quickstart.html)
-- [CLI reference](https://skypilot.readthedocs.io/en/latest/reference/cli.html)
+You can find our documentation [here](https://docs.skypilot.co/).
+- [Installation](https://docs.skypilot.co/en/latest/getting-started/installation.html)
+- [Quickstart](https://docs.skypilot.co/en/latest/getting-started/quickstart.html)
+- [CLI reference](https://docs.skypilot.co/en/latest/reference/cli.html)
 
 ## SkyPilot in 1 Minute
 
 A SkyPilot task specifies: resource requirements, data to be synced, setup commands, and the task commands.
 
-Once written in this [**unified interface**](https://skypilot.readthedocs.io/en/latest/reference/yaml-spec.html) (YAML or Python API), the task can be launched on any available cloud.  This avoids vendor lock-in, and allows easily moving jobs to a different provider.
+Once written in this [**unified interface**](https://docs.skypilot.co/en/latest/reference/yaml-spec.html) (YAML or Python API), the task can be launched on any available cloud.  This avoids vendor lock-in, and allows easily moving jobs to a different provider.
 
 Paste the following into a file `my_task.yaml`:
 
@@ -271,7 +271,7 @@ Prepare the workdir by cloning:
 git clone https://github.com/pytorch/examples.git ~/torch_examples
 ```
 
-Launch with `sky launch` (note: [access to GPU instances](https://skypilot.readthedocs.io/en/latest/cloud-setup/quota.html) is needed for this example):
+Launch with `sky launch` (note: [access to GPU instances](https://docs.skypilot.co/en/latest/cloud-setup/quota.html) is needed for this example):
 ```bash
 sky launch my_task.yaml
 ```
@@ -288,10 +288,10 @@ SkyPilot then performs the heavy-lifting for you, including:
 </p>
 
 
-Refer to [Quickstart](https://skypilot.readthedocs.io/en/latest/getting-started/quickstart.html) to get started with SkyPilot.
+Refer to [Quickstart](https://docs.skypilot.co/en/latest/getting-started/quickstart.html) to get started with SkyPilot.
 
 ## More Information
-To learn more, see [Concept: Sky Computing](https://docs.skypilot.co/en/latest/sky-computing.html), [SkyPilot docs](https://skypilot.readthedocs.io/en/latest/), and [SkyPilot blog](https://blog.skypilot.co/).
+To learn more, see [Concept: Sky Computing](https://docs.skypilot.co/en/latest/sky-computing.html), [SkyPilot docs](https://docs.skypilot.co/en/latest/), and [SkyPilot blog](https://blog.skypilot.co/).
 
 <!-- Keep this section in sync with index.rst in SkyPilot Docs -->
 Runnable examples:

{skypilot_nightly-1.0.0.dev20241217 → skypilot_nightly-1.0.0.dev20241219}/README.md

@@ -6,7 +6,7 @@
 </p>
 
 <p align="center">
-  <a href="https://skypilot.readthedocs.io/en/latest/">
+  <a href="https://docs.skypilot.co/">
     <img alt="Documentation" src="https://readthedocs.org/projects/skypilot/badge/?version=latest">
   </a>
 
@@ -43,7 +43,7 @@
   <summary>Archived</summary>
 
 - [Jul 2024] [**Finetune**](./llm/llama-3_1-finetuning/) and [**serve**](./llm/llama-3_1/) **Llama 3.1** on your infra
-- [Apr 2024] Serve and finetune [**Llama 3**](https://skypilot.readthedocs.io/en/latest/gallery/llms/llama-3.html) on any cloud or Kubernetes: [**example**](./llm/llama-3/)
+- [Apr 2024] Serve and finetune [**Llama 3**](https://docs.skypilot.co/en/latest/gallery/llms/llama-3.html) on any cloud or Kubernetes: [**example**](./llm/llama-3/)
 - [Mar 2024] Serve and deploy [**Databricks DBRX**](https://www.databricks.com/blog/introducing-dbrx-new-state-art-open-llm) on your infra: [**example**](./llm/dbrx/)
 - [Feb 2024] Speed up your LLM deployments with [**SGLang**](https://github.com/sgl-project/sglang) for 5x throughput on SkyServe: [**example**](./llm/sglang/)
 - [Dec 2023] Using [**LoRAX**](https://github.com/predibase/lorax) to serve 1000s of finetuned LLMs on a single instance in the cloud: [**example**](./llm/lorax/)
@@ -60,17 +60,17 @@
 SkyPilot is a framework for running AI and batch workloads on any infra, offering unified execution, high cost savings, and high GPU availability.
 
 SkyPilot **abstracts away infra burdens**:
-- Launch [dev clusters](https://skypilot.readthedocs.io/en/latest/examples/interactive-development.html), [jobs](https://skypilot.readthedocs.io/en/latest/examples/managed-jobs.html), and [serving](https://skypilot.readthedocs.io/en/latest/serving/sky-serve.html) on any infra
+- Launch [dev clusters](https://docs.skypilot.co/en/latest/examples/interactive-development.html), [jobs](https://docs.skypilot.co/en/latest/examples/managed-jobs.html), and [serving](https://docs.skypilot.co/en/latest/serving/sky-serve.html) on any infra
 - Easy job management: queue, run, and auto-recover many jobs
 
 SkyPilot **supports multiple clusters, clouds, and hardware** ([the Sky](https://arxiv.org/abs/2205.07147)):
 - Bring your reserved GPUs, Kubernetes clusters, or 12+ clouds
-- [Flexible provisioning](https://skypilot.readthedocs.io/en/latest/examples/auto-failover.html) of GPUs, TPUs, CPUs, with auto-retry
+- [Flexible provisioning](https://docs.skypilot.co/en/latest/examples/auto-failover.html) of GPUs, TPUs, CPUs, with auto-retry
 
 SkyPilot **cuts your cloud costs & maximizes GPU availability**:
-* [Autostop](https://skypilot.readthedocs.io/en/latest/reference/auto-stop.html): automatic cleanup of idle resources
-* [Managed Spot](https://skypilot.readthedocs.io/en/latest/examples/managed-jobs.html): 3-6x cost savings using spot instances, with preemption auto-recovery
-* [Optimizer](https://skypilot.readthedocs.io/en/latest/examples/auto-failover.html): 2x cost savings by auto-picking the cheapest & most available infra
+* [Autostop](https://docs.skypilot.co/en/latest/reference/auto-stop.html): automatic cleanup of idle resources
+* [Managed Spot](https://docs.skypilot.co/en/latest/examples/managed-jobs.html): 3-6x cost savings using spot instances, with preemption auto-recovery
+* [Optimizer](https://docs.skypilot.co/en/latest/examples/auto-failover.html): 2x cost savings by auto-picking the cheapest & most available infra
 
 SkyPilot supports your existing GPU, TPU, and CPU workloads, with no code changes.
 
@@ -79,13 +79,13 @@ Install with pip:
 # Choose your clouds:
 pip install -U "skypilot[kubernetes,aws,gcp,azure,oci,lambda,runpod,fluidstack,paperspace,cudo,ibm,scp]"
 ```
-To get the latest features and fixes, use the nightly build or [install from source](https://skypilot.readthedocs.io/en/latest/getting-started/installation.html):
+To get the latest features and fixes, use the nightly build or [install from source](https://docs.skypilot.co/en/latest/getting-started/installation.html):
 ```bash
 # Choose your clouds:
 pip install "skypilot-nightly[kubernetes,aws,gcp,azure,oci,lambda,runpod,fluidstack,paperspace,cudo,ibm,scp]"
 ```
 
-[Current supported infra](https://skypilot.readthedocs.io/en/latest/getting-started/installation.html) (Kubernetes; AWS, GCP, Azure, OCI, Lambda Cloud, Fluidstack, RunPod, Cudo, Paperspace, Cloudflare, Samsung, IBM, VMware vSphere):
+[Current supported infra](https://docs.skypilot.co/en/latest/getting-started/installation.html) (Kubernetes; AWS, GCP, Azure, OCI, Lambda Cloud, Fluidstack, RunPod, Cudo, Paperspace, Cloudflare, Samsung, IBM, VMware vSphere):
 <p align="center">
   <picture>
     <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/skypilot-org/skypilot/master/docs/source/images/cloud-logos-dark.png">
@@ -95,16 +95,16 @@ pip install "skypilot-nightly[kubernetes,aws,gcp,azure,oci,lambda,runpod,fluidst
 
 
 ## Getting Started
-You can find our documentation [here](https://skypilot.readthedocs.io/en/latest/).
-- [Installation](https://skypilot.readthedocs.io/en/latest/getting-started/installation.html)
-- [Quickstart](https://skypilot.readthedocs.io/en/latest/getting-started/quickstart.html)
-- [CLI reference](https://skypilot.readthedocs.io/en/latest/reference/cli.html)
+You can find our documentation [here](https://docs.skypilot.co/).
+- [Installation](https://docs.skypilot.co/en/latest/getting-started/installation.html)
+- [Quickstart](https://docs.skypilot.co/en/latest/getting-started/quickstart.html)
+- [CLI reference](https://docs.skypilot.co/en/latest/reference/cli.html)
 
 ## SkyPilot in 1 Minute
 
 A SkyPilot task specifies: resource requirements, data to be synced, setup commands, and the task commands.
 
-Once written in this [**unified interface**](https://skypilot.readthedocs.io/en/latest/reference/yaml-spec.html) (YAML or Python API), the task can be launched on any available cloud.  This avoids vendor lock-in, and allows easily moving jobs to a different provider.
+Once written in this [**unified interface**](https://docs.skypilot.co/en/latest/reference/yaml-spec.html) (YAML or Python API), the task can be launched on any available cloud.  This avoids vendor lock-in, and allows easily moving jobs to a different provider.
 
 Paste the following into a file `my_task.yaml`:
 
@@ -135,7 +135,7 @@ Prepare the workdir by cloning:
 git clone https://github.com/pytorch/examples.git ~/torch_examples
 ```
 
-Launch with `sky launch` (note: [access to GPU instances](https://skypilot.readthedocs.io/en/latest/cloud-setup/quota.html) is needed for this example):
+Launch with `sky launch` (note: [access to GPU instances](https://docs.skypilot.co/en/latest/cloud-setup/quota.html) is needed for this example):
 ```bash
 sky launch my_task.yaml
 ```
@@ -152,10 +152,10 @@ SkyPilot then performs the heavy-lifting for you, including:
 </p>
 
 
-Refer to [Quickstart](https://skypilot.readthedocs.io/en/latest/getting-started/quickstart.html) to get started with SkyPilot.
+Refer to [Quickstart](https://docs.skypilot.co/en/latest/getting-started/quickstart.html) to get started with SkyPilot.
 
 ## More Information
-To learn more, see [Concept: Sky Computing](https://docs.skypilot.co/en/latest/sky-computing.html), [SkyPilot docs](https://skypilot.readthedocs.io/en/latest/), and [SkyPilot blog](https://blog.skypilot.co/).
+To learn more, see [Concept: Sky Computing](https://docs.skypilot.co/en/latest/sky-computing.html), [SkyPilot docs](https://docs.skypilot.co/en/latest/), and [SkyPilot blog](https://blog.skypilot.co/).
 
 <!-- Keep this section in sync with index.rst in SkyPilot Docs -->
 Runnable examples:

{skypilot_nightly-1.0.0.dev20241217/sky/setup_files → skypilot_nightly-1.0.0.dev20241219}/setup.py

@@ -186,6 +186,6 @@ setuptools.setup(
         'Homepage': 'https://github.com/skypilot-org/skypilot',
         'Issues': 'https://github.com/skypilot-org/skypilot/issues',
         'Discussion': 'https://github.com/skypilot-org/skypilot/discussions',
-        'Documentation': 'https://skypilot.readthedocs.io/en/latest/',
+        'Documentation': 'https://docs.skypilot.co/',
     },
 )

{skypilot_nightly-1.0.0.dev20241217 → skypilot_nightly-1.0.0.dev20241219}/sky/__init__.py

@@ -5,7 +5,7 @@ from typing import Optional
 import urllib.request
 
 # Replaced with the current commit when building the wheels.
-_SKYPILOT_COMMIT_SHA = 'f0ebf13b5b164d012f4b033609eb1b364dc6caf3'
+_SKYPILOT_COMMIT_SHA = 'a73a9cb37152c5f05d06bd1afe29a685ef86f85c'
 
 
 def _get_git_commit():
@@ -35,7 +35,7 @@ def _get_git_commit():
 
 
 __commit__ = _get_git_commit()
-__version__ = '1.0.0.dev20241217'
+__version__ = '1.0.0.dev20241219'
 __root_dir__ = os.path.dirname(os.path.abspath(__file__))
 
 

{skypilot_nightly-1.0.0.dev20241217 → skypilot_nightly-1.0.0.dev20241219}/sky/adaptors/cloudflare.py

@@ -177,7 +177,7 @@ def check_credentials() -> Tuple[bool, Optional[str]]:
             hints += f'\n{_INDENT_PREFIX}  $ mkdir -p ~/.cloudflare'
             hints += f'\n{_INDENT_PREFIX}  $ echo <YOUR_ACCOUNT_ID_HERE> > ~/.cloudflare/accountid'  # pylint: disable=line-too-long
         hints += f'\n{_INDENT_PREFIX}For more info: '
-        hints += 'https://skypilot.readthedocs.io/en/latest/getting-started/installation.html#cloudflare-r2'  # pylint: disable=line-too-long
+        hints += 'https://docs.skypilot.co/en/latest/getting-started/installation.html#cloudflare-r2'  # pylint: disable=line-too-long
 
     return (False, hints) if hints else (True, hints)
 

{skypilot_nightly-1.0.0.dev20241217 → skypilot_nightly-1.0.0.dev20241219}/sky/adaptors/oci.py

@@ -1,9 +1,16 @@
 """Oracle OCI cloud adaptor"""
 
+import logging
 import os
 
 from sky.adaptors import common
 
+# Suppress OCI circuit breaker logging before lazy import, because
+# oci modules prints additional message during imports, i.e., the
+# set_logger in the LazyImport called after imports will not take
+# effect.
+logging.getLogger('oci.circuit_breaker').setLevel(logging.WARNING)
+
 CONFIG_PATH = '~/.oci/config'
 ENV_VAR_OCI_CONFIG = 'OCI_CONFIG'
 

{skypilot_nightly-1.0.0.dev20241217 → skypilot_nightly-1.0.0.dev20241219}/sky/backends/cloud_vm_ray_backend.py

@@ -1092,7 +1092,7 @@ class FailoverCloudErrorHandlerV2:
                     'having the required permissions and the user '
                     'account does not have enough permission to '
                     'update it. Please contact your administrator and '
-                    'check out: https://skypilot.readthedocs.io/en/latest/cloud-setup/cloud-permissions/gcp.html\n'  # pylint: disable=line-too-long
+                    'check out: https://docs.skypilot.co/en/latest/cloud-setup/cloud-permissions/gcp.html\n'  # pylint: disable=line-too-long
                     f'Details: {message}')
                 _add_to_blocked_resources(
                     blocked_resources,
@@ -1389,8 +1389,7 @@ class RetryingVmProvisioner(object):
                 f'in {to_provision.cloud}. '
                 f'{colorama.Style.RESET_ALL}'
                 f'To request quotas, check the instruction: '
-                f'https://skypilot.readthedocs.io/en/latest/cloud-setup/quota.html.'  # pylint: disable=line-too-long
-            )
+                f'https://docs.skypilot.co/en/latest/cloud-setup/quota.html.')
 
         for zones in self._yield_zones(to_provision, num_nodes, cluster_name,
                                        prev_cluster_status,

{skypilot_nightly-1.0.0.dev20241217 → skypilot_nightly-1.0.0.dev20241219}/sky/check.py

@@ -146,7 +146,7 @@ def check(
                 dim=True) + click.style(f'sky check{clouds_arg}', bold=True) +
             '\n' + click.style(
                 'If any problems remain, refer to detailed docs at: '
-                'https://skypilot.readthedocs.io/en/latest/getting-started/installation.html',  # pylint: disable=line-too-long
+                'https://docs.skypilot.co/en/latest/getting-started/installation.html',  # pylint: disable=line-too-long
                 dim=True))
 
         if disallowed_clouds_hint:

{skypilot_nightly-1.0.0.dev20241217 → skypilot_nightly-1.0.0.dev20241219}/sky/clouds/aws.py

@@ -617,7 +617,7 @@ class AWS(clouds.Cloud):
                 'Failed to fetch the availability zones for the account '
                 f'{identity_str}. It is likely due to permission issues, please'
                 ' check the minimal permission required for AWS: '
-                'https://skypilot.readthedocs.io/en/latest/cloud-setup/cloud-permissions/aws.html'  # pylint: disable=
+                'https://docs.skypilot.co/en/latest/cloud-setup/cloud-permissions/aws.html'  # pylint: disable=
                 f'\n{cls._INDENT_PREFIX}Details: '
                 f'{common_utils.format_exception(e, use_bracket=True)}')
         return True, hints

{skypilot_nightly-1.0.0.dev20241217 → skypilot_nightly-1.0.0.dev20241219}/sky/clouds/cudo.py

@@ -42,8 +42,7 @@ class Cudo(clouds.Cloud):
         f'{_INDENT_PREFIX}  $ cudoctl init\n'
         f'{_INDENT_PREFIX}For more info: '
         # pylint: disable=line-too-long
-        'https://skypilot.readthedocs.io/en/latest/getting-started/installation.html'
-    )
+        'https://docs.skypilot.co/en/latest/getting-started/installation.html')
 
     _PROJECT_HINT = (
         'Create a project and then set it as the default project,:\n'
@@ -51,8 +50,7 @@ class Cudo(clouds.Cloud):
         f'{_INDENT_PREFIX} $ cudoctl init\n'
         f'{_INDENT_PREFIX}For more info: '
         # pylint: disable=line-too-long
-        'https://skypilot.readthedocs.io/en/latest/getting-started/installation.html'
-    )
+        'https://docs.skypilot.co/en/latest/getting-started/installation.html')
 
     _CLOUD_UNSUPPORTED_FEATURES = {
         clouds.CloudImplementationFeatures.STOP: 'Stopping not supported.',

{skypilot_nightly-1.0.0.dev20241217 → skypilot_nightly-1.0.0.dev20241219}/sky/clouds/gcp.py

@@ -167,7 +167,7 @@ class GCP(clouds.Cloud):
         # ~/.config/gcloud/application_default_credentials.json.
         f'{_INDENT_PREFIX}  $ gcloud auth application-default login\n'
         f'{_INDENT_PREFIX}For more info: '
-        'https://skypilot.readthedocs.io/en/latest/getting-started/installation.html#google-cloud-platform-gcp'  # pylint: disable=line-too-long
+        'https://docs.skypilot.co/en/latest/getting-started/installation.html#google-cloud-platform-gcp'  # pylint: disable=line-too-long
     )
     _APPLICATION_CREDENTIAL_HINT = (
         'Run the following commands:\n'
@@ -175,7 +175,7 @@ class GCP(clouds.Cloud):
         f'{_INDENT_PREFIX}Or set the environment variable GOOGLE_APPLICATION_CREDENTIALS '
         'to the path of your service account key file.\n'
         f'{_INDENT_PREFIX}For more info: '
-        'https://skypilot.readthedocs.io/en/latest/getting-started/installation.html#google-cloud-platform-gcp'  # pylint: disable=line-too-long
+        'https://docs.skypilot.co/en/latest/getting-started/installation.html#google-cloud-platform-gcp'  # pylint: disable=line-too-long
     )
 
     _SUPPORTED_DISK_TIERS = set(resources_utils.DiskTier)
@@ -836,7 +836,7 @@ class GCP(clouds.Cloud):
                 'The following permissions are not enabled for the current '
                 f'GCP identity ({identity_str}):\n    '
                 f'{diffs}\n    '
-                'For more details, visit: https://skypilot.readthedocs.io/en/latest/cloud-setup/cloud-permissions/gcp.html')  # pylint: disable=line-too-long
+                'For more details, visit: https://docs.skypilot.co/en/latest/cloud-setup/cloud-permissions/gcp.html')  # pylint: disable=line-too-long
         return True, None
 
     def get_credential_file_mounts(self) -> Dict[str, str]:

{skypilot_nightly-1.0.0.dev20241217 → skypilot_nightly-1.0.0.dev20241219}/sky/clouds/kubernetes.py

@@ -395,7 +395,7 @@ class Kubernetes(clouds.Cloud):
                 tpu_requested = True
                 k8s_resource_key = kubernetes_utils.TPU_RESOURCE_KEY
             else:
-                k8s_resource_key = kubernetes_utils.GPU_RESOURCE_KEY
+                k8s_resource_key = kubernetes_utils.get_gpu_resource_key()
 
         port_mode = network_utils.get_port_mode(None)
 

{skypilot_nightly-1.0.0.dev20241217 → skypilot_nightly-1.0.0.dev20241219}/sky/clouds/oci.py

@@ -390,7 +390,7 @@ class OCI(clouds.Cloud):
         short_credential_help_str = (
             'For more details, refer to: '
             # pylint: disable=line-too-long
-            'https://skypilot.readthedocs.io/en/latest/getting-started/installation.html#oracle-cloud-infrastructure-oci'
+            'https://docs.skypilot.co/en/latest/getting-started/installation.html#oracle-cloud-infrastructure-oci'
         )
         credential_help_str = (
             'To configure credentials, go to: '

{skypilot_nightly-1.0.0.dev20241217 → skypilot_nightly-1.0.0.dev20241219}/sky/clouds/paperspace.py

@@ -258,7 +258,7 @@ class Paperspace(clouds.Cloud):
             return False, (
                 'Failed to access Paperspace Cloud with credentials.\n    '
                 'To configure credentials, follow the instructions at: '
-                'https://skypilot.readthedocs.io/en/latest/getting-started/installation.html#paperspace\n    '
+                'https://docs.skypilot.co/en/latest/getting-started/installation.html#paperspace\n    '
                 'Generate API key and create a json at `~/.paperspace/config.json` with \n     '
                 '    {"apiKey": "[YOUR API KEY]"}\n    '
                 f'Reason: {str(e)}')

{skypilot_nightly-1.0.0.dev20241217 → skypilot_nightly-1.0.0.dev20241219}/sky/clouds/runpod.py

@@ -253,7 +253,7 @@ class RunPod(clouds.Cloud):
                     '    Credentials can be set up by running: \n'
                     f'        $ pip install runpod \n'
                     f'        $ runpod config\n'
-                    '    For more information, see https://skypilot.readthedocs.io/en/latest/getting-started/installation.html#runpod'  # pylint: disable=line-too-long
+                    '    For more information, see https://docs.skypilot.co/en/latest/getting-started/installation.html#runpod'  # pylint: disable=line-too-long
                 )
 
             return True, None

{skypilot_nightly-1.0.0.dev20241217 → skypilot_nightly-1.0.0.dev20241219}/sky/clouds/service_catalog/common.py

@@ -296,7 +296,7 @@ def validate_region_zone_impl(
                         faq_msg = (
                             '\nIf a region is not included in the following '
                             'list, please check the FAQ docs for how to fetch '
-                            'its catalog info.\nhttps://skypilot.readthedocs.io'
+                            'its catalog info.\nhttps://docs.skypilot.co'
                             '/en/latest/reference/faq.html#advanced-how-to-'
                             'make-skypilot-use-all-global-regions')
                         error_msg += faq_msg + _get_all_supported_regions_str()

{skypilot_nightly-1.0.0.dev20241217 → skypilot_nightly-1.0.0.dev20241219}/sky/clouds/vsphere.py

@@ -266,7 +266,7 @@ class Vsphere(clouds.Cloud):
                 'Run the following commands:'
                 f'\n{cls._INDENT_PREFIX}  $ pip install skypilot[vSphere]'
                 f'\n{cls._INDENT_PREFIX}Credentials may also need to be set. '
-                'For more details. See https://skypilot.readthedocs.io/en/latest/getting-started/installation.html#vmware-vsphere'  # pylint: disable=line-too-long
+                'For more details. See https://docs.skypilot.co/en/latest/getting-started/installation.html#vmware-vsphere'  # pylint: disable=line-too-long
                 f'{common_utils.format_exception(e, use_bracket=True)}')
 
         required_keys = ['name', 'username', 'password', 'clusters']

{skypilot_nightly-1.0.0.dev20241217 → skypilot_nightly-1.0.0.dev20241219}/sky/data/storage.py

@@ -1157,7 +1157,7 @@ class S3Store(AbstractStore):
                     'Storage \'store: s3\' specified, but ' \
                     'AWS access is disabled. To fix, enable '\
                     'AWS by running `sky check`. More info: '\
-                    'https://skypilot.readthedocs.io/en/latest/getting-started/installation.html.' # pylint: disable=line-too-long
+                    'https://docs.skypilot.co/en/latest/getting-started/installation.html.' # pylint: disable=line-too-long
                     )
 
     @classmethod
@@ -1588,7 +1588,7 @@ class GcsStore(AbstractStore):
                     'Storage \'store: gcs\' specified, but '
                     'GCP access is disabled. To fix, enable '
                     'GCP by running `sky check`. '
-                    'More info: https://skypilot.readthedocs.io/en/latest/getting-started/installation.html.')  # pylint: disable=line-too-long
+                    'More info: https://docs.skypilot.co/en/latest/getting-started/installation.html.')  # pylint: disable=line-too-long
 
     @classmethod
     def validate_name(cls, name: str) -> str:
@@ -2110,7 +2110,7 @@ class AzureBlobStore(AbstractStore):
                     'Storage "store: azure" specified, but '
                     'Azure access is disabled. To fix, enable '
                     'Azure by running `sky check`. More info: '
-                    'https://skypilot.readthedocs.io/en/latest/getting-started/installation.html.'  # pylint: disable=line-too-long
+                    'https://docs.skypilot.co/en/latest/getting-started/installation.html.'  # pylint: disable=line-too-long
                 )
 
     @classmethod
@@ -2813,7 +2813,7 @@ class R2Store(AbstractStore):
                     'Storage \'store: r2\' specified, but ' \
                     'Cloudflare R2 access is disabled. To fix, '\
                     'enable Cloudflare R2 by running `sky check`. '\
-                    'More info: https://skypilot.readthedocs.io/en/latest/getting-started/installation.html.'  # pylint: disable=line-too-long
+                    'More info: https://docs.skypilot.co/en/latest/getting-started/installation.html.'  # pylint: disable=line-too-long
                     )
 
     def initialize(self):

{skypilot_nightly-1.0.0.dev20241217 → skypilot_nightly-1.0.0.dev20241219}/sky/provision/gcp/constants.py

@@ -142,7 +142,7 @@ FIREWALL_RULES_TEMPLATE = [
 ]
 
 # A list of permissions required to run SkyPilot on GCP.
-# Keep this in sync with https://skypilot.readthedocs.io/en/latest/cloud-setup/cloud-permissions/gcp.html # pylint: disable=line-too-long
+# Keep this in sync with https://docs.skypilot.co/en/latest/cloud-setup/cloud-permissions/gcp.html # pylint: disable=line-too-long
 VM_MINIMAL_PERMISSIONS = [
     'compute.disks.create',
     'compute.disks.list',

{skypilot_nightly-1.0.0.dev20241217 → skypilot_nightly-1.0.0.dev20241219}/sky/provision/kubernetes/instance.py

@@ -180,6 +180,7 @@ def _raise_pod_scheduling_errors(namespace, context, new_nodes):
                             #  case we will need to update this logic.
                             # TODO(Doyoung): Update the error message raised
                             # with the multi-host TPU support.
+                            gpu_resource_key = kubernetes_utils.get_gpu_resource_key()  # pylint: disable=line-too-long
                             if 'Insufficient google.com/tpu' in event_message:
                                 extra_msg = (
                                     f'Verify if '
@@ -192,14 +193,15 @@ def _raise_pod_scheduling_errors(namespace, context, new_nodes):
                                                        pod,
                                                        extra_msg,
                                                        details=event_message))
-                            elif (('Insufficient nvidia.com/gpu'
+                            elif ((f'Insufficient {gpu_resource_key}'
                                    in event_message) or
                                   ('didn\'t match Pod\'s node affinity/selector'
                                    in event_message)):
                                 extra_msg = (
-                                    f'Verify if '
-                                    f'{pod.spec.node_selector[label_key]}'
-                                    ' is available in the cluster.')
+                                    f'Verify if any node matching label  '
+                                    f'{pod.spec.node_selector[label_key]} and '
+                                    f'sufficient resource {gpu_resource_key} '
+                                    f'is available in the cluster.')
                                 raise config_lib.KubernetesError(
                                     _lack_resource_msg('GPU',
                                                        pod,
@@ -722,13 +724,13 @@ def _create_pods(region: str, cluster_name_on_cloud: str,
                        'Continuing without using nvidia RuntimeClass.\n'
                        'If you are on a K3s cluster, manually '
                        'override runtimeClassName in ~/.sky/config.yaml. '
-                       'For more details, refer to https://skypilot.readthedocs.io/en/latest/reference/config.html')  # pylint: disable=line-too-long
+                       'For more details, refer to https://docs.skypilot.co/en/latest/reference/config.html')  # pylint: disable=line-too-long
 
     needs_gpus = False
     limits = pod_spec['spec']['containers'][0].get('resources',
                                                    {}).get('limits')
     if limits is not None:
-        needs_gpus = limits.get(kubernetes_utils.GPU_RESOURCE_KEY, 0) > 0
+        needs_gpus = limits.get(kubernetes_utils.get_gpu_resource_key(), 0) > 0
 
     # TPU pods provisioned on GKE use the default containerd runtime.
     # Reference: https://cloud.google.com/kubernetes-engine/docs/how-to/migrate-containerd#overview  # pylint: disable=line-too-long

{skypilot_nightly-1.0.0.dev20241217 → skypilot_nightly-1.0.0.dev20241219}/sky/provision/kubernetes/utils.py

@@ -438,7 +438,7 @@ def detect_accelerator_resource(
     nodes = get_kubernetes_nodes(context)
     for node in nodes:
         cluster_resources.update(node.status.allocatable.keys())
-    has_accelerator = (GPU_RESOURCE_KEY in cluster_resources or
+    has_accelerator = (get_gpu_resource_key() in cluster_resources or
                        TPU_RESOURCE_KEY in cluster_resources)
 
     return has_accelerator, cluster_resources
@@ -972,7 +972,7 @@ def is_kubeconfig_exec_auth(
                     '~/.sky/config.yaml:\n'
                     '    kubernetes:\n'
                     '      remote_identity: SERVICE_ACCOUNT\n'
-                    '    More: https://skypilot.readthedocs.io/en/latest/'
+                    '    More: https://docs.skypilot.co/en/latest/'
                     'reference/config.html')
         return True, exec_msg
     return False, None
@@ -2253,10 +2253,11 @@ def get_node_accelerator_count(attribute_dict: dict) -> int:
         Number of accelerators allocated or available from the node. If no
             resource is found, it returns 0.
     """
-    assert not (GPU_RESOURCE_KEY in attribute_dict and
+    gpu_resource_name = get_gpu_resource_key()
+    assert not (gpu_resource_name in attribute_dict and
                 TPU_RESOURCE_KEY in attribute_dict)
-    if GPU_RESOURCE_KEY in attribute_dict:
-        return int(attribute_dict[GPU_RESOURCE_KEY])
+    if gpu_resource_name in attribute_dict:
+        return int(attribute_dict[gpu_resource_name])
     elif TPU_RESOURCE_KEY in attribute_dict:
         return int(attribute_dict[TPU_RESOURCE_KEY])
     return 0
@@ -2415,3 +2416,18 @@ def process_skypilot_pods(
         num_pods = len(cluster.pods)
         cluster.resources_str = f'{num_pods}x {cluster.resources}'
     return list(clusters.values()), jobs_controllers, serve_controllers
+
+
+def get_gpu_resource_key():
+    """Get the GPU resource name to use in kubernetes.
+    The function first checks for an environment variable.
+    If defined, it uses its value; otherwise, it returns the default value.
+    Args:
+        name (str): Default GPU resource name, default is "nvidia.com/gpu".
+    Returns:
+        str: The selected GPU resource name.
+    """
+    # Retrieve GPU resource name from environment variable, if set.
+    # Else use default.
+    # E.g., can be nvidia.com/gpu-h100, amd.com/gpu etc.
+    return os.getenv('CUSTOM_GPU_RESOURCE_KEY', default=GPU_RESOURCE_KEY)

{skypilot_nightly-1.0.0.dev20241217 → skypilot_nightly-1.0.0.dev20241219/sky/setup_files}/setup.py

@@ -186,6 +186,6 @@ setuptools.setup(
         'Homepage': 'https://github.com/skypilot-org/skypilot',
         'Issues': 'https://github.com/skypilot-org/skypilot/issues',
         'Discussion': 'https://github.com/skypilot-org/skypilot/discussions',
-        'Documentation': 'https://skypilot.readthedocs.io/en/latest/',
+        'Documentation': 'https://docs.skypilot.co/',
     },
 )

{skypilot_nightly-1.0.0.dev20241217 → skypilot_nightly-1.0.0.dev20241219}/sky/skypilot_config.py

@@ -238,7 +238,7 @@ def _try_load_config() -> None:
                 _dict,
                 schemas.get_config_schema(),
                 f'Invalid config YAML ({config_path}). See: '
-                'https://skypilot.readthedocs.io/en/latest/reference/config.html. '  # pylint: disable=line-too-long
+                'https://docs.skypilot.co/en/latest/reference/config.html. '  # pylint: disable=line-too-long
                 'Error: ',
                 skip_none=False)
 

{skypilot_nightly-1.0.0.dev20241217 → skypilot_nightly-1.0.0.dev20241219}/sky/utils/kubernetes/deploy_remote_cluster.sh

@@ -1,5 +1,5 @@
 #!/bin/bash
-# Refer to https://skypilot.readthedocs.io/en/latest/reservations/existing-machines.html for details on how to use this script.
+# Refer to https://docs.skypilot.co/en/latest/reservations/existing-machines.html for details on how to use this script.
 set -e
 
 # Colors for nicer UX