/repartition/graph

URL: https://<aws.fqdn>/<aws.cluster.name>/gpudb-0/repartition/graph

Rebalances an existing partitioned graph.

IMPORTANT: It's highly recommended that you review the Network Graphs & Solvers concepts documentation, the Graph REST Tutorial, and/or some graph examples before using this endpoint.

Input Parameter Description

Name Type Description
graph_name string Name of the graph resource to rebalance.
options map of string to strings

Optional parameters. The default value is an empty map ( {} ).

Supported Parameters (keys) Parameter Description
new_graph_name If a non-empty value is specified, the original graph will be kept (non-default behaviour) and a new balanced graph will be created under this given name. When the value is empty (default), the generated 'balanced' graph will replace the original 'unbalanced' graph under the same graph name. The default value is ''.
source_node The distributed shortest path solve is run from this source node to all the nodes in the graph to create balaced partitions using the iso-distance levels of the solution. The source node is selected by the rebalance algorithm automatically (default case when the value is an empty string). Otherwise, the user specified node is used as the source. The default value is ''.
sql_request_avro_json The default value is ''.

Output Parameter Description

The GPUdb server embeds the endpoint response inside a standard response structure which contains status information and the actual response to the query. Here is a description of the various fields of the wrapper:

Name Type Description
status String 'OK' or 'ERROR'
message String Empty if success or an error message
data_type String 'repartition_graph_response' or 'none' in case of an error
data String Empty string
data_str JSON or String

This embedded JSON represents the result of the /repartition/graph endpoint:

Name Type Description
result boolean Indicates a successful rebalancing on all servers.
info map of string to strings Additional information.

Empty string in case of an error.