V$VECTOR_GRAPH_INDEX

V$VECTOR_GRAPH_INDEX displays diagnostic information about In-Memory Neighbor Graph vector indexes.

This view is populated only on Oracle Autonomous AI Database Serverless (ADB-S).

Column	Datatype	Description
`OWNER`	`VARCHAR2(129)`	Owner of the index
`INDEX_NAME`	`VARCHAR2(129)`	Name of the index
`PARTITION_NAME`	`VARCHAR2(129)`	Object partition name The value of this column is null for non-partitioned objects.
`INDEX_OBJN`	`NUMBER`	Index object number
`ANCHOR_ADDRESS`	`RAW(8)`	Address of the anchor structure for the vector in the vector pool
`INDEX_GRAPH_TYPE`	`VARCHAR2(129)`	Type of the constructed graph of the index The value of this column is always `HNSW`.
`NUM_LAYERS`	`NUMBER`	Number of layers in the constructed graph
`NUM_VECTORS`	`NUMBER`	Number of indexed vectors in the index
`SPARSE_LAYER_VECTORS`	`NUMBER`	Number of vectors in the sparse layers Note that a sparse layer is defined as any layer above the bottom-most layer.
`NUM_NEIGHBORS`	`NUMBER`	Maximum number of neighbors a vector can have in the sparse layers
`EF_CONSTRUCTION`	`NUMBER`	Maximum number of closest vector candidates considered at each step of the search during insertion
`TOTAL_EDGES`	`NUMBER`	Number of total edges in the constructed graph as of the latest snapshot
`REF_COUNT`	`NUMBER`	Number used to track readers of the index Note that an index can only be dropped when the value of this column is 0.
`QUERY_DIST_COUNT`	`NUMBER`	Number of distance computations performed as part of queries that used the index
`CREATION_DIST_COUNT`	`NUMBER`	Number of distance computations performed as part of index creation
`PRUNED_NEIGHBORS`	`NUMBER`	Number of neighbors pruned out during the neighbor selection phase of index creation
`NUM_SNAPSHOTS`	`NUMBER`	Number of snapshots currently tracked for the index
`MAX_SNAPSHOT`	`NUMBER`	Number of snapshots created for the index
`ALLOCATED_BYTES`	`NUMBER`	Total amount of memory allocated to the index (in bytes)
`USED_BYTES`	`NUMBER`	Total amount of memory currently used by the index (in bytes)
`INDEX_USED_COUNT`Foot 1	`NUMBER`	Number of times the index has been used by queries
`COVERING_COLS`Foot 2	`VARCHAR2(3999)`	Comma-separated list of included columns for the index
`QUANTIZATION_TYPE`Foot 3	`VARCHAR2(129)`	Indicates whether the index is a scalar quantized HNSW index (`SCALAR`) or not (`NONE`)
`QUANTIZATION_MDATA`Foot 3	`JSON`	For scalar quantized HNSW indexes, the quantization metadata, in JSON format. Otherwise, null.
`CON_ID`	`NUMBER`	The ID of the container to which the data pertains. Possible values include: `0`: This value is used for rows containing data that pertain to the entire CDB. This value is also used for rows in non-CDBs. `1`: This value is used for rows containing data that pertain to only the root n: Where n is the applicable container ID for the rows containing data

Footnote 1 This column is available starting with Oracle AI Database 26ai, Release Update 23.26.2.

Footnote 2 This column is available starting with Oracle AI Database 26ai, Release Update 23.26.1.

Footnote 3 This column is available starting with Oracle AI Database 26ai, Release Update 23.26.3.

Note: This view is available starting with Oracle AI Database 26ai, Release Update 23.6.