Key-Value Storage Layer¶
Most TensorStore drivers access the underlying storage for array data through a key-value store layer that supports a variety of underlying storage systems, such as the local filesystem and cloud storage systems, through different drivers.
Drivers¶
- json KvStore : object¶
Key-value store specification.
This describes the common properties supported by all key-value stores. Refer to the driver documentation for the supported
driveridentifiers and driver-specific properties.- Subtypes:¶
kvstore/file— Read/write access to the local filesystem.kvstore/gcs— Read/write access to Google Cloud Storage (GCS).kvstore/http— Read-only access to arbitrary HTTP servers.kvstore/memory— Read/write, non-persistent in-memory storage.kvstore/neuroglancer_uint64_sharded— Read/write adapter for the Neuroglancer Precomputed sharded format.kvstore/ocdbt— Read/write adapter for the OCDBT format.kvstore/s3— Read/write access to Amazon S3-compatible object stores.kvstore/zarr_sharding_indexed— Read/write adapter for the zarr v3 sharded_indexed format.kvstore/zip— Read-only adapter for the ZIP archive format.
- Required members:¶
- driver : string¶
Key-value store driver identifier
- json KvStoreUrl : string¶
URL representation of a key-value store.
In addition to the
JSON objectrepresentation, a key-value store may be specified more concisely by a URL.Warning
The URL representation of a key-value store specification may exclude certain parameters, such as concurrency limits.
- Subtypes:¶
KvStoreUrl/file—file://KvStore URL schemeKvStoreUrl/gs—gs://KvStore URL schemeKvStoreUrl/http—http://andhttps://KvStore URL schemeKvStoreUrl/memory—memory://KvStore URL schemeKvStoreUrl/s3—s3://KvStore URL scheme