Values of a simulation metric, aggregated along with specific dimensions.

Item Summary
Name	Type	Multiplicity	Description
ref	container SimulationMetricReferenceType	[0,1]	Identifier of the metric.
aggregationFunction	property SimulationMetricAggregationFunctionType	[0,1]	What function is used to produce the aggregated value.
sourceDimensions	container SimulationMetricPartitionDimensionsType	[0,1]	What are the dimensions of the "most granular" (source) data?
partition	container SimulationMetricPartitionType	[0,-1]	Partial or total aggregation of values for given metric.

Items
- ref
  
  Flags: RAM,runtime
  
  Multiplicity: [0,1]
  
  Identifier of the metric. This is either the identifier of the new (aggregation) metric, like "percent-of-users-disabled", or the name of the original (source) metric that is being aggregated, like "number-of-attributes-modified" or an event mark reference.
- aggregationFunction
  
  Flags: RAM,runtime,AVals:13
  
  Multiplicity: [0,1]
  
  What function is used to produce the aggregated value. It is provided here to be ready when it should be re-applied e.g. during data aggregation.
- sourceDimensions
  
  Flags: RAM,runtime
  
  Multiplicity: [0,1]
  
  What are the dimensions of the "most granular" (source) data?
- partition
  
  Flags: RAM,runtime
  
  Multiplicity: [0,-1]
  
  Partial or total aggregation of values for given metric. The exact values that are stored here are configurable, e.g. according to what dimensions we want to do the partitioning, and whether we should also provide summarization (collapsing) along some or all of these dimensions. TODO item name