Add vertex persistent resource to settings for step operator

docs/book/component-guide/step-operators/vertex.md

@@ -136,3 +136,31 @@ For more information and a full list of configurable attributes of the Vertex st
 Note that if you wish to use this step operator to run steps on a GPU, you will need to follow [the instructions on this page](../../how-to/pipeline-development/training-with-gpus/README.md) to ensure that it works. It requires adding some extra settings customization and is essential to enable CUDA for the GPU to give its full acceleration.
 
 <figure><img src="https://static.scarf.sh/a.png?x-pxid=f0b4f458-0a54-4fcd-aa95-d5ee424815bc" alt="ZenML Scarf"><figcaption></figcaption></figure>
+
+#### Using Persistent Resources for Faster Development
+
+When developing ML pipelines that use Vertex AI, the startup time for each CustomJob can be significant since Vertex needs to provision new compute resources for each run. To speed up development iterations, you can use Vertex AI's [Persistent Resources](https://cloud.google.com/vertex-ai/docs/training/persistent-resource-overview) feature, which keeps compute resources warm between runs.
+
+To use persistent resources with the Vertex step operator, you can configure it either when registering the step operator or through the step settings:
+
+```python
+from zenml.integrations.gcp.flavors.vertex_step_operator_flavor import VertexStepOperatorSettings
+
+@step(step_operator=<STEP_OPERATOR_NAME>, settings={"step_operator": VertexStepOperatorSettings(
+    persistent_resource_id="my-persistent-resource",  # specify your persistent resource ID
+    machine_type="n1-standard-4",
+    accelerator_type="NVIDIA_TESLA_T4",
+    accelerator_count=1,
+)})
+def trainer(...) -> ...:
+    """Train a model."""
+    # This step will use the persistent resource and start faster
+```
+
+This is particularly useful when:
+* You're developing locally and want to iterate quickly on steps that need GPU/TPU resources
+* You have a local orchestrator but want to leverage Vertex AI for specific compute-intensive steps
+
+{% hint style="warning" %}
+Remember that persistent resources continue to incur costs as long as they're running, even when idle. Make sure to monitor your usage and configure appropriate idle timeout periods.
+{% endhint %}

docs/book/how-to/project-setup-and-management/setting-up-a-project-repository/connect-your-git-repository.md

@@ -12,7 +12,7 @@ A code repository in ZenML refers to a remote storage location for your code. So
 
 Code repositories enable ZenML to keep track of the code version that you use for your pipeline runs. Additionally, running a pipeline that is tracked in a registered code repository can [speed up the Docker image building for containerized stack components](../../infrastructure-deployment/customize-docker-builds/use-code-repositories-to-speed-up-docker-build-times.md) by eliminating the need to rebuild Docker images each time you change one of your source code files.
 
-Learn more about how code repositories benefit development [here](../../infrastructure-deployment/customize-docker-builds/use-code-repositories-to-speed-up-docker-build-times.md).
+Learn more about how code repositories benefit development [here](../../customize-docker-builds/how-to-reuse-builds.md).
 
 ## Registering a code repository
 

src/zenml/integrations/gcp/flavors/vertex_step_operator_flavor.py

@@ -51,14 +51,16 @@ class VertexStepOperatorSettings(BaseSettings):
             https://cloud.google.com/vertex-ai/docs/training/configure-compute#boot_disk_options
         boot_disk_type: Type of the boot disk. (Default: pd-ssd)
             https://cloud.google.com/vertex-ai/docs/training/configure-compute#boot_disk_options
-
+        persistent_resource_id: The ID of the persistent resource to use for the job.
+            https://cloud.google.com/vertex-ai/docs/training/persistent-resource-overview
     """
 
     accelerator_type: Optional[str] = None
     accelerator_count: int = 0
     machine_type: str = "n1-standard-4"
     boot_disk_size_gb: int = 100
     boot_disk_type: str = "pd-ssd"
+    persistent_resource_id: Optional[str] = None
 
 
 class VertexStepOperatorConfig(

src/zenml/integrations/gcp/step_operators/vertex_step_operator.py

@@ -265,6 +265,7 @@ def launch(
             }
             if self.config.encryption_spec_key_name
             else {},
+            "persistent_resource_id": settings.persistent_resource_id,
         }
         logger.debug("Vertex AI Job=%s", custom_job)