Fabric Batch Overview

The Batch process is a Fabric utility that can be used to run multiple commands on a list of instances, regardless of its size or type.

The Batch process is used to trigger the following:

Sync instances, to perform multiple syncs on all (or a subset of) instances (IIDs) for a specific Logical Unit. This process is also referred to as the Migration Process, handled by the migrate command.
Broadway flow, to run a given flow on multiple instances.
Republish data changes using the CDC notification mechanism.

Fabric Batch Features

Fabric's integrated Batch process provides the following advantages to ensure the smooth execution of heavy-resources demanding processes:

Efficient distribution of jobs between nodes.
Dynamic adjusting of work balance configuration between nodes during execution phases.
Monitoring of instances' synchronization progress (batch sync) at cluster, DC or node levels from the command line or using the Batch Monitor Dashboard.
Process failure recovering. (e.g non-responding nodes).
Stopping and resuming of the migration process (LUI synchronization).
Process tracking at entity level. (e.g. timing, duration, handling node or failure management).

Instances Sync (Migration)

BATCH LU ('LUI','LUI2','LUI3','LUI4') FABRIC_COMMAND="sync_instance LU.?" with ASYNC='true';

Broadway Flows Execution

BATCH LU FABRIC_COMMAND="broadway LU.SampleFlow SampleIID=?" with async=true;

CDC Republish

BATCH LU from fabric FABRIC_COMMAND="cdc_republish_instance Customer.?" with async=true;