ShardGraphs

Header File/Library File

• Header file: #include <ge/ge_api.h>
• Library file: libge_runner.so

Function Usage

This API is deprecated. Avoid using it.

This API is not recommended. ShardGraphsToFile is recommended, which delivers better performance than ShardGraphs.

This API shards graphs in a session based on the AddGraph sequence. Sharded graphs are saved in the memory.

The sharding mode is specified by the ge.graphParallelOptionPath option in Command-Line Options. If the parallel graph function is disabled, this API does not work.

Naming rules for sharded graphs:

• In SPMD mode: The naming format is {Original graph name}_{ClusterId}_{ItemId}_{ChipId}_{VirtualStageId}_{Original GraphId}. The number of sharded graphs equals the number of original graphs multiplied by the number of shards.
• In non-SPMD mode: The naming format is {Original graph name}_{Original GraphId}. The number of sharded graphs equals the number of original graphs.

After a graph is sharded, it does not exist in the session. For the ID of the newly generated graph after sharding, the original graph ID is replaced by another ID.

Prototype

1

Status Session::ShardGraphs()

Parameters

N/A

Returns

Restrictions

This API works only when the parallel graph function is enabled.

Example

1
2
3
4
5

Session session(options); // The parallel graph function is enabled through options.
Graph init_graph("init_graph");
Graph first_graph("first_graph");
Graph second_graph("second_graph");
session.ShardGraphs();