Sandboxes

Weaviate Cloud Services (WCS) provides two instance types.

• Sandbox clusters. Sandbox clusters are small, free clusters designed for learning and experimentation. Sandbox clusters not scalable and expire after 14 days.
• Managed clusters. Managed clusters are robust, paid clusters designed for production use. Manage clusters scale to meet your needs and don't expire.

Capabilities

Sandbox clusters run a full version of the Weaviate core database. When you create a sandbox, it has the latest version of Weaviate core.

Weaviate core uses modules to connect to vectorizers, rankers, generators, and other services. Modules provide programmatic connections to these services. WCS provides built in modules for more than a dozen services. To see the list of modules that are available for your cluster, check the cluster details panel in the Weaviate web console.

Suggested uses

Sandbox clusters are intended for learning, experimenting, and exploring. Sandboxes are fully featured, but they have a limited lifetime and they run on scaled down infrastructure compared to managed clusters.

If you are working through a course in Weaviate Academy, consider using a sandbox to complete the lessons without having to install and configure a local instance.

Limitations

Sandboxes are not designed for production use. Compared to managed instances, they are limited in size, duration, and operational resilience. Sandboxes have these limits:

• 14 day lifespan
• No backup
• One admin API key
• No read-only API key
• Fixed instance size
• Less memory capacity than the smallest managed cluster
• Less processor capacity than the smallest managed cluster
• Less storage capacity than the smallest managed cluster

Export data

To export your data, iterate through each of your collections. For each collection, iterate through the objects in the collection and export them one by one. The process to iterate through collection objects depends on how your objects are configured. For details, see Read all objects.

The cursor API provides another way to export your data.

Switch to a managed cluster

There is no direct path to convert a sandbox cluster to a managed cluster. Follow these steps to move to a managed cluster.

1. Export your data.
2. Export your schema.
3. Configure billing in your WCS account.
4. Create a managed cluster.
5. Recreate your schema.
6. Import your data.

Export your data

To export your data, iterate through each of your collections. For each collection, iterate through the objects in the collection and export them one by one. The process to iterate through collection objects depends on how your objects are configured. For details, see Read all objects.

The cursor API provides another way to export your data.

Export your schema

To export your collection definitions, follow these steps:

Python (v4)

response = client.collections.list_all()

print(response)

Python (v3)

response = client.schema.get()

print(json.dumps(response, indent=2))

JavaScript/TypeScript

await client.collections.listAll()

Java

Result<Schema> result = client.schema().getter()
  .run();

String json = new GsonBuilder().setPrettyPrinting().create().toJson(result.getResult());
System.out.println(json);

Go

schema, err := client.Schema().Getter().
  Do(ctx)

b, err := json.MarshalIndent(schema, "", "  ")
fmt.Println(string(b))

Configure billing

To configure billing, follow these steps:

1. Click the cogwheel icon at the left hand side of the console.
2. Scroll down to Billing Information.
3. Fill out the form and click Save.
4. Enter your credit card information on the next screen.

Create a managed cluster

Managed clusters require billing details. WCS prompts you to add billing details if you have not already added them.

To create a managed cluster, follow these steps:

1. Select the "Managed cluster" tab.
2. Give your cluster a name.
3. Select a region.
4. Accept the terms and conditions.
5. Click create.

It takes a minute or two to create the new cluster. When the cluster is ready, WCS displays a check mark (✔️) next to the cluster name.

The managed cluster has new API keys. Update the API keys in your client code to use the new keys.

Recreate your schema

Use the exported (sandbox cluster) schema to recreate the collection schema in the managed cluster. For more details on creating schemas, see Manage collections.

Python (v4)

from weaviate.classes.config import Property, DataType

# Note that you can use `client.collections.create_from_dict()` to create a collection from a v3-client-style JSON object
client.collections.create(
    "Article",
    properties=[
        Property(name="title", data_type=DataType.TEXT),
        Property(name="body", data_type=DataType.TEXT),
    ]
)

Python (v3)

class_obj = {
    "class": "Article",
    "properties": [
        {
            "name": "title",
            "dataType": ["text"],
        },
        {
            "name": "body",
            "dataType": ["text"],
        },
    ],
}

client.schema.create_class(class_obj)  # returns null on success

JavaScript/TypeScript

const newCollection = await client.collections.create({
  name: 'Article',
  properties: [
    {
      name: 'title',
      dataType: weaviate.configure.dataType.TEXT,
      vectorizePropertyName: true,
      tokenization: 'lowercase' ,
    },
    {
      name: 'body',
      dataType: weaviate.configure.dataType.TEXT,
      tokenization: 'whitespace',
      skipVectorisation: true
    },
  ],
})
// The returned value is the full collection definition, showing all defaults
console.log(JSON.stringify(newCollection, null, 2));

Import your data

Use the batch import API to import your data into the managed cluster.

If your data file is very large, consider streaming your upload.

Duration

Sandboxes expire after 14 days. When the sandbox expires, the data is deleted and cannot be retrieved. WCS sends an email notification approximately three days before a sandbox is due to expire.

Sandboxes can be extended three times. Each extension is for an additional 14 day period. To extend your sandbox, contact support.

Support

For questions and support, try the following resources:

• Community forum
• GitHub
• Slack

Weaviate also offers paid support services. If you have a contract with Weaviate, contact your sales representative for details.