atoti.solution.model.
Cube specific properties are separated by its own prefix: atoti.solution.model.cubes.<cube-name>.*. For example, the VaR cube in Market Risk (Enterprise Risk) uses the prefix atoti.solution.model.cubes.var.*.
This page documents all common properties organized by category. Default values are applied when a property is omitted.
Prerequisites: Familiarity with Spring Boot externalized configuration (application.yml or application.properties).
Enable cubes
Every cube is enabled by default. The Ratings cube of Counterparty Credit Risk (Enterprise Risk) also needs rating components, and fails startup without them unless it is disabled. To disable a cube, use the following property:Currencies
Prefix:atoti.solution.model.cubes.<cube-name>.currencies
Currency properties control the members of the Display Currency simulation dimension.
Confidence levels
Prefix:atoti.solution.model.cubes.<cube-name>.confidence-levels
Confidence level properties control the members of the confidence levels simulation dimension.
Scenarios
Prefix:atoti.solution.model.cubes.<cube-name>.scenarios
Scenario properties control the size and naming of the P&L vector dimension. The scenario count must match the length of the pnl_vector column in the Trade P&L input file.
Rounding
Prefix:atoti.solution.model.cubes.<cube-name>
The rounding method controls how the fractional quantile rank is resolved to an integer index when computing quantile-based measures such as VaR, ES, and PFE.
The following rounding method values are available. The example column shows how each method resolves a fractional rank of
2.5:
Feature flags
Prefix:atoti.solution.model.cubes.<cube-name>
The common layer provides the following features as reusable building blocks. Each solution decides independently which features to include and whether to expose a toggle for them. Not all solutions support all features, and default values vary by solution.
Each feature toggle is a nested object directly under the cube prefix. Set the
enabled sub-property to false to disable a feature. Omitting the property leaves it at its default value (typically true).
When a feature is disabled, the corresponding dimension is excluded from the cube schema and the associated data loading topic is not registered. Input files for disabled features are ignored even if present. The documentation for each solution lists the default values and which features are supported.
Multi-level structure
Some features support a multi-level hierarchical structure, such as book, counterparty, and organization. These features expose additional sub-properties under their respective feature prefix. Prefix:atoti.solution.features.<feature-name>
Counterparty structure display
The counterparty structure additionally chooses what its members display. Book and organization always display identifiers. An organization has no name to show in their place. Prefix:atoti.solution.features.counterparties
Whichever is set, the members themselves carry the counterparty identifier: only their caption changes. Filters, saved dashboards and anything else holding a member coordinate keep resolving when the setting is switched.
Counterparty names need not be unique. Counterparties sharing a name with one of their siblings are displayed as
Name (Identifier), and one with no name is displayed by its identifier. Because a name is only a caption, a counterparty arriving later and colliding with an existing one re-captions it without moving it.
A caption is keyed by counterparty identifier alone, not by date. Where a counterparty is loaded under different names at different as-of dates, the latest as-of date’s name wins. A cube that keeps one effective counterparty set rather than a dated history, such as the Ratings cube of Counterparty Credit Risk (Enterprise Risk), names only the counterparties no loaded date names, so an application showing names from both kinds of cube displays the dated name in both. Unloading a date withdraws exactly that date’s captions: a counterparty no other loaded date names then displays the effective set’s name if it has one, and its identifier otherwise.
Members keep their position: a formatter decides what a member displays, not where it sits, so the
hierarchy stays ordered by counterparty identifier. Where those identifiers are opaque, names
therefore appear in an order that looks arbitrary, and the sort control in Atoti UI does not correct
it: sorting a level there ranks members by their position in the hierarchy’s own order, so ascending
reproduces identifier order and descending reverses it.
Order by what is displayed in the query instead. Atoti UI recognises such a sort and shows it as an
alphabetical sort of the level, provided the calculated measure is named after the level and ends
_for_order:
Query results limit
Prefix:atoti.solution.model.cubes.<cube-name>.query-results-limit
Query result limit properties cap the number of results returned by cube queries to prevent out-of-memory conditions.
As-of-date data load
Controls what an application loads and unloads for a single as-of date, and which dates it loads on startup. Available to an application that importsAsOfDateDataLoadConfiguration. See Data loading for how the load itself runs.
Prefix: atoti.solution.data-load.as-of-date
Example configuration
The following example shows a minimalapplication.yml configuration using <cube-name> as a placeholder. Replace <cube-name> with the actual cube prefix for the solution (for example, var for the Market Risk (Enterprise Risk) VaR cube).