Create and use endpoints

With the SambaStudio platform, you can generate predictions from your models on your data by deploying it to an endpoint. This document describes how to:

Create an endpoint

You can create endpoints to be used for prediction using the GUI or the CLI. Follow the instructions described in the corresponding sections to learn how.

Starting with release 24.1.1, SambaStudio allows you to specify the SambaNova Systems' Reconfigurable Dataflow Units™ (RDUs) generation version to use for endpoints. RDUs can be specified for both GUI and CLI workflows.

Contact your administrator for more information on RDU configurations specific to your SambaStudio platform.

Create a CoE endpoint using the GUI

Starting with SambaStudio release 24.2.1, SN40L External link users can create Composition of Experts (CoE) endpoints. Follow the steps below to create an endpoint using a SambaNova provided CoE, such as Samba-1, using the GUI. After adding your CoE endpoint, adjust the endpoint share settings to share your CoE endpoint with other users and tenants.

Due to the large number of checkpoints involved, CoE endpoints can stay in the Setting Up status for 2-3 hours before deploying to Live status.

  1. Create a new project or use an existing one.

  2. From a project window, click New endpoint. The Create endpoint window will open.

  3. In the Create endpoint window, enter a name for the endpoint into the Endpoint name field.

  4. From the ML App drop-down, select a CoE designated ML App. We selected Symphony CoE App in our example.

  5. From the Select model drop-down, choose SambaNova models and select a downloaded CoE model.

  6. Select the version of the model to use in the Select model drop-down.

  7. The RDU generation will default to SN40L, as COE models only run on the SambaNova SN40L.

  8. Select the number of instances to use when creating the endpoint. The information statement displays the required number of RDUs in a single node along with the total number and available number of RDUs in your SambaStudio configuration. In our example, our CoE model requires one RDU in a node for each instance. Increasing the number of instances will increase the RDU requirements.

  9. Expand Model Parameters to view additional parameters.

    1. The model parallel rdus field designates the number of RDUs used to run in parallel when creating the endpoint. The number directly relates to the number of RDUs available in a single node and the selected number of instances. If the required number of RDUs are not available in a single node, you will receive a warning message.

  10. Click Add an endpoint to queue the endpoint for deployment. Once the endpoint is created, it can take several minutes for the endpoint to be deployed.

Create endpoint
Figure 1. Create endpoint

Insufficient RDUs in a single node

If the required number of RDUs are not available in a single node, you will receive a warning message stating that your endpoint will be in Awaiting RDU status until the RDUs become available in a single node. Contact your administrator for more information on RDU configurations specific to your SambaStudio platform.

Insufficient RDUs in a node
Figure 2. Insufficient RDUs in a node

Create a non-CoE endpoint using the GUI

Follow the steps below to create an endpoint, that is not part of the CoE architecture, using the GUI to be used for prediction. After saving your endpoint, adjust the endpoint share settings to share your endpoint with other users and tenants.

Base models do not support inference and cannot be deployed for endpoints. It is recommended to use Base models for training and not inference.

  1. Create a new project or use an existing one.

  2. From a project window, click New endpoint. The Create endpoint window (Figure 3) will open.

  3. In the Create endpoint window, enter a name for the endpoint into the Endpoint name field, as shown in Figure 3.

  4. Select the ML App from the ML App drop-down, as shown in Figure 3.

  5. From the Select model drop-down, choose My models, Shared models, SambaNova models, or Select from Model Hub.

    1. My models displays a list of models that you have previously added to the Model Hub.

    2. Shared models displays a list of models that have been shared with the selected active tenant.

    3. SambaNova models displays a list of models provided by SambaNova.

      Create Endpoint
      Figure 3. Create endpoint window
    4. Select from Model Hub displays a window with a list of downloaded models that correspond to a selected ML App, as shown in Figure 4, or a list of all the downloaded models if an ML App is not selected. The list can be filtered by selecting options under Field of application, ML APP, Architecture, and Owner. Additionally, you can enter a term or value into the Search field to refine the model list by that input. Choose the model you wish to use and confirm your choice by clicking Use model.

      Model Hub
      Figure 4. Select from Model Hub
  6. Select the version of the model to use in the Select model drop-down.

  7. Enter a value for the Number of instances to designate how many instances will be deployed, as shown in Figure 5. The Number of instances setting helps the endpoint scale to accommodate a high volume number of requests. You can adjust the value based on the expected volume number of requests and available resources.

  8. The RDU generation drop-down allows users to select an available RDU generation version to use for creating an endpoint, as shown in Figure 5. If more than one option is available, the SambaStudio platform will default to the recommended RDU generation version to use based on your platform’s configuration and the selected model. You can select a different RDU generation version to use, if available, than the recommended option from the drop-down. Contact your administrator for more information on RDU configurations specific to your SambaStudio platform.

  9. Click Add an endpoint, as shown in Figure 5, to queue the endpoint for deployment. Once the endpoint is created, it can take several minutes for the endpoint to be deployed.

    Create endpoint
    Figure 5. Completed create endpoint window

    Your endpoint is deployed and will be available to generate predictions when the endpoint status displays Live.

Insufficient RDUs in a single node

If the required number of RDUs are not available in a single node, you will receive a warning message stating that your endpoint will be in Awaiting RDU status until the RDUs become available in a single node. Contact your administrator for more information on RDU configurations specific to your SambaStudio platform.

Insufficient RDUs in a node
Figure 6. Insufficient RDUs in a node

Create an endpoint using the CLI

The example below demonstrates how to create an endpoint using the snapi endpoint create command. You will need to provide the following:

  • A project to assign the endpoint. Create a new project or use an existing one.

  • A new endpoint name for the --endpoint_name input.

  • The model to be used for the --model-checkpoint input.

  • The RDU architecture generation version to use of your SambaStudio platform configuration for the --arch input.

    • Run the snapi tenant info command to view the available RDU generation version(s) specific to your SambaStudio platform. Contact your administrator for more information on RDU configurations specific to your SambaStudio platform.

    • Run the snapi model info command to obtain the --arch input compatible for the selected model.

  • The number of instances to be used for the --instances input.

Example snapi create endpoint command
$ snapi endpoint create \
   --project <project-name> \
   --endpoint_name <endpoint-name> \
   --model-checkpoint <model-name> \
   --arch SN10 \
   --instances <number-of-instances>

Run snapi endpoint create --help to display additional usage and options.

Example snapi model info command

The example snapi model info command snippet below demonstrates where to find the compatible --arch input for the GPT_13B_Base_Model when used in creating an endpoint. The required value is located on the last line of the example snippet and is represented as 'deploy': { 'sn10'. Note that this example snippet contains only a portion of the actual snapi model info command response. You will need to specify:

  • The model name or ID for the --model input.

  • Use deploy for the --job-type input. This returns the 'deploy': { 'sn10' value, which would be entered as --arch SN10 into the snapi create endpoint command.

Click to view the example snapi model info command snippet.
$ snapi model info \
--model GPT_13B_Base_Model \
--job-type deploy

               Model Info
             ============
ID                    : 61b4ff7d-fbaf-444d-9cba-7ac89187e375
Name                  : GPT_13B_Base_Model
Architecture          : GPT 13B
Field of Application  : language
Validation Loss       : -
Validation Accuracy   : -
App                   : 57f6a3c8-1f04-488a-bb39-3cfc5b4a5d7a
Dataset               : {'info': 'N/A\n', 'url': ''}
SambaNova Provided    : True
Version               : 1
Description           : This is a randomly initialized model, meant to be used to kick off a pre-training job.

Generally speaking, the process of pre-training is expensive both in terms of compute and data. For most use cases, it will be better to fine tune one of the provided checkpoints, rather than starting from scratch.


Created Time          : 2023-03-23 00:00:00 +0000 UTC
Status                : Available
Steps                 : 0
Hyperparameters              :
 {   'batch_predict': {},
    'deploy': {   'sn10': {   'imageVariants': [],

View endpoint information

Once your endpoint has been deployed, you can view detailed information about it using the GUI or CLI.

View an endpoint using the GUI

The Endpoint window displays detailed information about your endpoint, as well as providing access to common functions.

  • Open any Endpoint window by selecting the endpoint you wish to view from its associated Project, as shown in Figure 7

    • The Monitoring column provides a link to the Grafana monitoring dashboard. Click the corresponding icon to open the dashboard for that endpoint.

      Endpoints table
      Figure 7. Endpoints displayed in its associated Project
  • Additionally, you can view a list of Live endpoints from the SambaStudio Dashboard.

The Endpoint window displays the following information, as shown in Figure 8:

Details
  • The status of the endpoint is displayed in the Details panel top bar.

  • ML App displays the ML App that was selected when the endpoint was created.

  • Generation displays the RDU generation version used when the endpoint was created.

  • Model/Checkpoint displays the endpoint’s associated model/checkpoint.

  • Model Version displays the current model version associated with the endpoint.

  • Instances displays the number of instances used by the endpoint.

  • Created on displays the date the endpoint was created.

  • Updated on displays the date the endpoint was last updated.

  • Owner displays the owner of the endpoint.

  • Project ID allows you to copy the endpoint’s associated Project ID to your clipboard. Use the Project ID to make requests programmatically.

  • Endpoint ID allows you to copy the Endpoint ID to your clipboard. Use the Endpoint ID to make requests programmatically.

  • Model parameters displays the parameters used during endpoint creation.

  • Predict URL provides the URL of a predict endpoint to use programmatically to make requests. Use the copy icon to temporarily store the URL (copy) to your computer’s clipboard. See API reference for more information.

  • Streaming prediction URL provides the URL of a stream endpoint to use programmatically to make requests. Use the copy icon to temporarily store the URL (copy) to your computer’s clipboard. See API reference for more information.

Logs and monitoring
  • Logs allows you to visualize and analyze an endpoint’s logs in the OpenSearch Dashboard.

  • Monitoring allows you to monitor your endpoint and view metric information in the Grafana dashboard for an endpoint.

API keys
  • The API Keys table displays a list of API keys that can be used to make authenticated requests of the endpoint URL. See Endpoint API keys for information on generating and using API keys. The API Keys table displays the following information:

    • KeyID displays the endpoint’s API key to be used for authenticated requests. Use the copy icon to temporarily store the API key (copy) to your computer’s clipboard.

    • Description displays the API key’s unique description so that it can be easily identified.

    • Created on displays the date the API key was created.

    • Owner displays the owner of the API key.

    • Status displays the current status of the API key.

Endpoint window
Figure 8. Endpoint window

View and download endpoint logs using OpenSearch

SambaStudio provides endpoint integration with the OpenSearch platform External link. This allows you to visualize and analyze endpoint logs in OpenSearch.

Please be aware of the following when interacting with the OpenSearch Dashboard:

  • The OpenSearch Dashboard is only compatible with jobs and endpoints created in SambaStudio 24.5.1 and later.

  • Logs are automatically cleared by the platform after 30 days. Meaning logs older than 30 days will no longer be available.

  • There is a maximum limit of 30K rows when downloading log reports.

  1. Navigate to an endpoint’s detail page and click Logs to view logs in the OpenSearch Dashboard.

    OpenSearch link
    Figure 9. OpenSearch link
  2. The OpenSearch Dashboard will open in a new browser window displaying the log information of the project associated with the job, including all endpoints and jobs of that project.

    Example OpenSearch dashboard
    Figure 10. Example OpenSearch Dashboard
Add filters

You can quickly refine the logs displayed in the OpenSearch Dashboard by adding a filter.

  • Select an available id in the left menu and add it as a filter.

  • You can create a filter using a tracking id to refine the dashboard display.

    Example tracking id filter in OpenSearch Dashboard
    Figure 11. Example tracking id filter in OpenSearch Dashboard
Save and download reports

You can save and download reports defined by criteria you have set in the OpenSearch Dashboard.

  1. Click Save in the top menu. The Save search box will open.

  2. Enter a title into the field and click Save

    Example OpenSearch save search
    Figure 12. Example OpenSearch Dashboard save search
  3. Refresh the page and click Reporting in the top menu to open the Generate and download drop-down. Select a format (Generate CSV is recommended) to generate the report.

    OpenSearch generate and download drop-down
  4. OpenSearch Dashboard will generate your report and download the file to the location configured by your browser.

Endpoint monitoring

SambaStudio provides a monitoring dashboard (Grafana) that displays metric information for all endpoints. From the Details section of an Endpoint window, click Monitoring to open the Grafana monitoring dashboard in a new browser window.

Endpoint monitoring link
Figure 13. Endpoint monitoring link

View endpoint info using the CLI

Use the snapi endpoint info command to view specific information about your endpoint, including:

  • The assigned ID of the endpoint.

  • The ML App used when the endpoint was created.

  • The Model used when the endpoint was created.

  • The number of Instances used when the endpoint was created.

  • The URL path to the endpoint location.

    Please use the GUI to get an endpoint’s Stream URL path.

  • The initial API Key (First Key) created for the endpoint. See Endpoint API keys for information on generating and using API keys.

  • The current Status of the endpoint.

  • The associated API Keys of the endpoint that can be used to make authenticated requests of the endpoint URL. See Endpoint API keys for information on generating and using API keys.

  • The Arch RDU generation version used when the endpoint was created.

The example below demonstrates how to use the snapi endpoint info command to view the info of an endpoint. You will need to provide the following:

  • The project name or ID the endpoint is assigned.

  • The endpoint name or ID.

Example snapi endpoint info command
$ snapi endpoint info \
    --project <project-name> \
    --endpoint <endpoint-name>
          Endpoint Info
         =============
ID             : 4eba9aaa-543f-4f8b-bbc1-30182252f85a
Name           : hav2
Description    :
Project ID     : 87deae92-570e-443f-8ae8-4521fb43ad09
ML App         : Generative Tuning 13B
Model          : GPT_13B_Human_Aligned_Instruction_Tuned_V2
Instances      : 0
URL            : <path-to-endpoint>>
API Key        : 6fbc8f9d-34e2-47e4-983f-19a1845326b4
Status         : Stopped
Created at     : 2023-09-07T21:47:27.581072+00:00
Last Updated   : 2023-11-07T17:02:07.939675+00:00
API Keys       : [{'user_id': 'billb', 'time_created': '2023-11-07T16:51:04.896235+00:00', 'description': 'hav2-shared-key', 'status': 'Live', 'id': '1205d9db-250a-4e71-b1a8-33fa3ffa05cc', 'endpoint_id': '4eba9aaa-543f-4f8b-bbc1-30182252f85a', 'time_updated': '2023-11-07T16:51:04.896235+00:00', 'api_key': 'a0e8e81f-697a-401a-b7c5-688788e0bb8a'}, {'user_id': 'billb', 'time_created': '2023-11-07T13:06:50.209626+00:00', 'description': 'First Key', 'status': 'Live', 'id': '12fffa5a-c615-4271-a920-f4bb1b3da303', 'endpoint_id': '4eba9aaa-543f-4f8b-bbc1-30182252f85a', 'time_updated': '2023-11-07T13:06:50.209626+00:00', 'api_key': '6fbc8f9d-34e2-47e4-983f-19a1845326b4'}]
Hyperparams    : None
Arch           : SN10

Update an endpoint

SambaStudio allows you to update an endpoint while still keeping the endpoint active for use during the update process. For example, if you’re endpoint was created with a version 1 model, you can update the model to version 2 while the endpoint remains active and in use during the update process.

  1. In the Endpoint window, click Update Available or View Updates to start the update process.

    Endpoint Window update available
    Figure 14. Endpoint window with updates available
  2. The Update Available box will open displaying the available updates for the endpoint as well as the Start switch at drop-down.

    1. In our example, a Version 2 model update is available for the endpoint.

    2. The Start switch at drop-down defines when the new updated endpoint will be used to serve requests by setting the number of RDU instances to be used for the updated endpoint. In our example, we selected 2, indicating that we want our endpoint to switch to use the updated model Version 2 once the 2 RDU instances are in set up and live.

    3. Click Update to start the update process.

      Update Available box
      Figure 15. Update Available box
  3. Once your endpoint starts the update process, click View Details in the Update in progress bar to see the status of the update.

    View Details in Update in progress bar
    Figure 16. View Details in Update in progress bar
  4. The Update Endpoint box will open and display the status of the endpoint update. In our example, we selected that our endpoint should switch to the updated endpoint once 2 RDU instances are set up and live.

    1. In our example Update Endpoint box, we can see the following:

      1. None of the Version 2 Instances are Live.

      2. None of the Version 2 Instances have started setting up.

      3. We are waiting on 2/2 RDUs, meaning no RDUs have been assigned yet.

      4. Version 1 of our endpoint is currently serving requests, and as such no RDUs are setting up and we are not waiting on RDUs.

        Update Endpoint box
        Figure 17. Update Endpoint box

Endpoint API keys

An endpoint API key allows you to make authenticated requests of the endpoint URL, such as generating predictions. SambaStudio provides the ability to create multiple API keys for an endpoint. This allows organizations to easily distribute and manage endpoints to different entities and users.

API keys can be added, edited, and revoked using the GUI or the CLI. Follow the instructions described in the corresponding sections to learn how.

The endpoint API key and the platform API key have two distinct implementations.

  1. The endpoint API key is a unique API key generated for each endpoint.

  2. The platform API key is used as an authorization key for authenticated access to the platform. See SambaStudio resources for information about the platform API key.

Add an endpoint API key using the GUI

You can add additional API keys to an endpoint. Please be aware of the following when adding a new API key to an endpoint.

  • Endpoint API keys can be added by:

    • The owner of the endpoint.

    • An organization administrator (OrgAdmin) across all tenants.

    • A tenant administrator (TenantAdmin) within their assigned tenant.

  • The platform creates an initial API key for each endpoint with the description of First Key.

Follow the steps below to add an API key to an endpoint.

  1. From the API keys section of an endpoint window, click Add new.

    API keys section
    Figure 18. API keys section

    The Add new API key box will appear.

  2. Enter a description into the Add description field for the endpoint’s new API key.

  3. Click Add to complete the process and add the endpoint’s new API key.

    Add new API key
    Figure 19. Add new API key
  4. The API keys table will display the new API key and the description you created for your endpoint.

    Endpoint API key added
    Figure 20. API keys table with new API key

Edit an endpoint API key description using the GUI

An endpoint’s API key existing description can be edited. Please be aware of the following when editing an endpoint API key description.

  • The description of an endpoint API key can be edited by:

    • The owner of the endpoint.

    • An organization administrator (OrgAdmin) across all tenants.

    • A tenant administrator (TenantAdmin) within their assigned tenant.

  • The initial API key, and its description of First Key, created for an endpoint by the platform, cannot be edited.

Follow the steps below to edit an endpoint API key description.

  1. From the API keys section of an endpoint window, click the kebob icon (three dots) in the API key you wish to edit and select Edit from the drop-down.

    API key drop-down
    Figure 21. API key drop-down

    The Edit key description box will open.

  2. Enter a new description into the Key description field.

  3. Click Save to complete the operation and update the API key description.

    Endpoint edit key description box
    Figure 22. Endpoint edit key description box

Revoke an endpoint API key using the GUI

Endpoint API keys can be revoked to prevent further usage. Please be aware of the following when revoking an endpoint API key.

  • Endpoint API keys can be revoked by:

    • The owner of the endpoint.

    • An organization administrator (OrgAdmin) across all tenants.

    • A tenant administrator (TenantAdmin) within their assigned tenant.

  • The initial API key (First Key) created for an endpoint by the platform, cannot be revoked.

  • Once an endpoint API key is revoked it cannot be reactivated.

  • The API keys table will display the revoked API key with the status Revoked.

  • All users will lose access to a revoked API key.

Follow the steps below to revoke an endpoint API key.

  1. From the API keys section of an endpoint window, click the kebob icon (three dots) in the API key you wish to edit and select Revoke from the drop-down.

    API key drop-down
    Figure 23. API key drop-down

    The confirmation box will open with a statement describing the consequences of revoking the API key.

  2. Click Revoke to confirm that you want to revoke the endpoint API key.

    Endpoint revoke key confirmation box
    Figure 24. Endpoint revoke key confirmation box

List all endpoint API keys using the CLI

Similar to the endpoint API keys table of the GUI, the SambaStudio command-line interface (CLI) can display a list of API keys for an endpoint.

The example below demonstrates how to list all API keys for an endpoint using the snapi endpoint list-apikeys command. You will need to provide the endpoint name or ID.

Example snapi endpoint list-apikeys command
$ snapi endpoint list-apikeys \
   --endpoint <endpoint-name>

   ------------------------------------------------------------------------
    List of API Keys for Endpoint <endpoint-name>
------------------------------------------------------------------------

+--------------------------------------+--------------------------------------+----------------------+----------------------+----------------------+
|               API Key                |             Description              |      Created On      |        Owner         |        Status        |
+======================================+======================================+======================+======================+======================+
| a0e8e81f-697a-401a-b7c5-688788e0bb8a | hav2-shared-key                      | 07 Nov, 2023         | billb                | Active               |
+--------------------------------------+--------------------------------------+----------------------+----------------------+----------------------+
| 6fbc8f9d-34e2-47e4-983f-19a1845326b4 | First Key                            | 07 Nov, 2023         | billb                | Active               |
+--------------------------------------+--------------------------------------+----------------------+----------------------+----------------------+

Add an endpoint API key using the CLI

Similar to the GUI, the SambaStudio command-line interface (CLI) can be used to add an API key to an endpoint. Please be aware of the following when adding a new API key to an endpoint.

  • Endpoint API keys can be added by:

    • The owner of the endpoint.

    • An organization administrator (OrgAdmin) across all tenants.

    • A tenant administrator (TenantAdmin) within their assigned tenant.

  • The platform creates an initial API key for each endpoint with the description of First Key.

The example below demonstrates how to add an API key to an endpoint using the snapi endpoint add-apikey command. You will need to provide the following:

  • The endpoint name or ID.

  • A description to identify the new API key is recommended.

Example snapi endpoint add-apikey command
$ snapi endpoint add-apikey \
    --endpoint <endpoint-name> \
    --description <endpoint-description>

Edit an endpoint API key description using the CLI

An endpoint’s API key existing description can be edited using the SambaStudio command-line interface (CLI). Please be aware of the following when editing an endpoint API key description.

  • The description of an endpoint API key can be edited by:

    • The owner of the endpoint.

    • An organization administrator (OrgAdmin) across all tenants.

    • A tenant administrator (TenantAdmin) within their assigned tenant.

  • The initial API key, and its description of First Key, created for an endpoint by the platform, cannot be edited.

The example below demonstrates how to edit an endpoint’s API key description using the snapi endpoint edit-apikey command. You will need to provide the following:

  • The endpoint name or ID.

  • The API key ID.

  • A new description for the API key.

Example snapi endpoint edit-apikey command
$ snapi endpoint edit-apikey \
    --endpoint <endpoint-name> \
    --apikey <API-key-ID> \
    --description <new-description>

Revoke an endpoint API key using the CLI

An endpoint’s API key can be revoked using the SambaStudio command-line interface (CLI). Please be aware of the following when revoking an endpoint API key.

  • Endpoint API keys can be revoked by:

    • The owner of the endpoint.

    • An organization administrator (OrgAdmin) across all tenants.

    • A tenant administrator (TenantAdmin) within their assigned tenant.

  • The initial API key (First Key) created for an endpoint by the platform, cannot be revoked.

  • Once an endpoint API key is revoked it cannot be reactivated.

  • The API key info will display the revoked API key with the status Revoked.

  • All users will lose access to a revoked API key.

The example below demonstrates how to revoke an endpoint’s API key using the snapi endpoint revoke-apikey command. You will need to provide the following:

  • The endpoint name or ID.

  • The API key ID.

Example snapi endpoint revoke-apikey command
$ snapi endpoint revoke-apikey \
    --endpoint <endpoint-name> \
    --apikey <API-key>

View endpoint API key info using the CLI

You can view detailed information about the API key of an endpoint using the SambaStudio command-line interface (CLI). This is useful to check the endpoint’s API key status (after revoking an API key) or confirm any changes you might have made to an endpoint API key (such as the description).

The example below demonstrates how to view an endpoint’s API key info using the snapi endpoint info-apikey command. You will need to provide the following:

  • The endpoint name or ID.

  • The API key ID.

Example snapi endpoint info-apikey command
$ snapi endpoint info-apikey \
    --endpoint <endpoint-name> \
    --apikey <API-key>

Edit an endpoint

You can edit an existing endpoint using the GUI or the CLI. Follow the instructions described in the corresponding sections to learn how.

Edit an endpoint using the GUI

Follow the steps below to edit an existing endpoint using the GUI.

If the required number of RDUs are not available in a single node are not available when editing your endpoint, you will receive a warning message stating that your endpoint will be in Awaiting RDU status until the RDUs become available in a single node. Contact your administrator for more information on RDU configurations specific to your SambaStudio platform.

  1. From an Endpoint window, click the kebob icon (three dots) and select Edit from the drop-down, as shown in Figure 25. Or, from the Endpoints table, click the kebob icon (three dots) for the endpoint you wish to edit and select Edit from the drop-down., as shown in Figure 26. The Edit endpoint window will open.

    Edit endpoint
    Figure 25. Edit an endpoint from the endpoint window
    Edit endpoint
    Figure 26. Edit an endpoint from the endpoints table
  2. As shown in Figure 27, from the Edit endpoint window you can:

    1. Enter a new description into the Description field.

    2. Choose a different model from the Select Model drop-down.

    3. Select an available RDU generation version to use. If more than one option is available, the SambaStudio platform will default to the recommended RDU generation version to use based on your platform’s configuration and the selected model. You can select a different RDU generation version to use, if available, than the recommended option from the drop-down. Contact your administrator for more information on RDU configurations specific to your SambaStudio platform.

    4. Update the Number of instances.

  3. Click Edit endpoint to complete process and send the endpoint to the queue.

    Update endpoint window
    Figure 27. Update the endpoint

The endpoint’s updated information will be reflected in the endpoints table.

Edit an endpoint using the CLI

The example below demonstrates how to use the snapi endpoint update command to edit and update an endpoint. You will need to provide the following:

  • The project the endpoint is assigned.

  • The endpoint name or ID.

  • The RDU generation version of the endpoint for the --arch input.

Similar to the GUI, you can:

  • Change the number of instances by using the --instances option.

  • Change the description by using the --description option.

Example snapi endpoint update command
$  snapi endpoint update \
   --project <project-name> \
   --description <new-description> \
   --endpoint <endpoint-name> \
   --arch sn10 \
   --instances <number-of-instances>

Stop and restart an endpoint

To free up Reconfigurable Dataflow Unit™ (RDU) resources in the platform, you may wish to stop an endpoint from running.

Stop an endpoint using the GUI

Do the following to stop an existing endpoint from running using the GUI.

  1. From an Endpoint window, click the kebob icon (three dots) and select Stop from the drop-down, as shown in Figure 28. An alert box will open.

    Edit endpoint
    Figure 28. Stop endpoint
  2. The alert box (Figure 29) displays a warning statement that the models in the endpoint will stop running.

  3. As shown in Figure 29, click Continue to confirm that you want to stop the endpoint from running. Click Cancel to return the Endpoint window.

    Stop endpoint window
    Figure 29. This endpoint will stop running box

Restart an endpoint using the GUI

From an Endpoint window, click the kebob icon (three dots) and select Restart from the drop-down, as shown in Figure 30. The platform will restart the endpoint and will notify you that the endpoint is restarting.

Restart endpoint
Figure 30. Restart endpoint

The status of a restarted endpoint is displayed in the endpoint’s top Details bar.

Stop an endpoint using the CLI

The example below demonstrates how to use the snapi endpoint stop command to stop an endpoint from running. You will need to provide the following:

  • The project the endpoint is assigned.

  • The endpoint name or ID.

Example snapi endpoint stop command
$  snapi endpoint stop \
   --project <project-name> \
   --endpoint <endpoint-name>

Delete an endpoint

You can delete an existing endpoint using the GUI or the CLI. Follow the instructions described in the corresponding sections to learn how.

Delete an endpoint using the GUI

Do the following to delete an existing endpoint using the GUI.

  1. From an Endpoint window, click the kebob icon (three dots) and select Delete from the drop-down, as shown in Figure 31. An alert window box open.

    Delete endpoint
    Figure 31. Delete endpoint

    The alert box (Figure 32) displays a message that the endpoint will be deleted and no longer available. The alert also lists the endpoint that will be deleted.

  2. Click Yes to finalize deleting the endpoint, as shown in Figure 32. Click Cancel to return the Endpoint window.

    Delete alert window
    Figure 32. Delete alert window

Delete an endpoint using the CLI

The example below demonstrates how to use the snapi endpoint delete command to delete an endpoint. You will need to provide the following:

  • The project the endpoint is assigned.

  • The endpoint name or ID.

Example snapi endpoint delete command
$  snapi endpoint delete \
   --project <project-name> \
   --endpoint <endpoint-name> \
Endpoint new-endpoint-snapi successfully marked delete.