Skip to content

Exploration and Dependencies

Mercator provides two complementary tools for navigating the map:

  • Exploration — displays a free-form graph around an asset: clicking a node in the graph reveals all objects connected to it (physically or logically) along with all incoming and outgoing flows.
  • Dependencies — imposes a layered structure (upstream → downstream) starting from a selected asset.

These tools allow you to navigate dynamically through the graph of relationships between assets. They reveal dependencies across layers: upward (abstract / business) or downward (concrete / physical).

Hierarchical views are available in both cases: starting from a macro-process, you can visualise all the processes, activities and operations that depend on it; or starting from a site, all the rooms and equipment it contains.

Exploration

The Exploration tool (Tools ➡️ Exploration) displays a relational graph with no imposed hierarchy.
Nodes are positioned by a physics engine (attraction / repulsion), which allows you to:

  • freely explore relationships
  • identify clusters
  • navigate complex dependencies
  • visualise the full graph around an asset

This mode is ideal for intuitive exploration.

🔍 See also the interface overview

Dependencies

The Dependencies tool (Tools ➡️ Dependencies) imposes a layered structure (upstream → downstream).
Dependencies are directed and organised by level, which allows you to:

  • quickly understand data flows
  • analyse upstream or downstream dependencies
  • document an architecture
  • clarify complex relationships

This mode is ideal for structured analysis.

🔍 See also interface overview


Interfaces

Description of the Exploration and Dependencies interfaces.

UI Element Type 🕸️ Exploration ⟁ Dependencies Role
Filter Dropdown list Filters on map views (e.g. Applications)
Attributes Dropdown list Filters on asset attributes (e.g. Opensource)
Object Dropdown list Selects the asset to explore (e.g. HR-Solution)
Add Green button Adds the selected asset to the graph window
Physical Green toggle Enables/disables display of physical network links
Level (1–5) Numeric picker Number of relationship levels to expand
Deploy Green button Launches exploration with the defined parameters
Analyze Green button Launches dependency analysis with the defined parameters
↑ Up Cyan button Expands toward upper layers (abstract / business)
↓ Down Cyan button Expands toward lower layers (concrete / physical)
↕ Both Cyan button Expands in both directions simultaneously
Remove Red button Removes the selected asset from the graph
Reset Yellow button Completely resets the graph
Show IP Grey button Displays IP addresses on network assets

Filter

⚠️ Critical point: the Filter field has a dual effect that is essential to understand before using these tools.

Effect 1 — Restricting the "Object" Dropdown

This is the most intuitive use. By selecting a view in the Filter field (e.g. Applications), the "Object" dropdown only shows assets belonging to that view. This avoids searching for an asset across the entire CMDB.

Effect 2 — Limiting Asset Visibility in the Graph (common pitfall)

This is the least expected effect, and the most frequent source of errors. The filter does not simply restrict the "Object" list: it also controls which asset types will appear in the graph during exploration.

Concretely: if you select a logical-server with only Logical Infrastructure in the filter, applications linked to that server will never appear in the graph, even if they exist in Mercator and are correctly associated. They are simply excluded because their type is not covered by the active filter.

Illustrated example:

Active filter Explored asset Result in the graph
Logical Infrastructure LOGICAL-SERVER-HR-11 Visible: NETWORK-CORE-11, SUBNET-CORE-11, SUBNET-VIRT-111 — but not HR-Solution
Applications + Logical Infrastructure LOGICAL-SERVER-HR-11 Also visible: HR-Solution and DB-HR-PROD
(empty) Any asset All linked assets are visible, across all layers
Filter 🕸️ Exploration ⟁ Dependencies
Logical Infrastructure only:
HR-Solution does not appear.
Applications + Logical Infrastructure:
HR-Solution and DB-HR-PROD appear.

Practical Rule: Which Filter to Choose?

Goal Recommended filter
Quickly find an asset within a specific view Enter only the target view (e.g. Applications)
Cross-layer exploration or analysis (application + infrastructure) Enter all relevant views, or leave empty
Full impact analysis (all layers) Leave the filter empty to exclude nothing
Exploration or analysis limited to a single layer (e.g. network only) Enter only that layer's view

💡 Tips: - When in doubt, always start with an empty filter. You can narrow it down afterwards if the graph becomes too dense. - Trick: use the filter to easily locate the asset in the list, add it to the graph, then clear the filter before clicking 🕸️ Deploy or ⟁ Analyze — assets from all layers will then be visible. - You can also type the beginning of an asset name directly in the "Object" field to filter without using the Filter field.


Direction

The direction of dependency analysis is relative to the Mercator view hierarchy. This hierarchy is aligned with ArchiMate:

↑ UP    ═══════════════════════════════════════════════  toward 100 (Business)
         Layer 1 : Entities, Processes, Actors  (100–260)
         Layer 2 : Applications, Modules, DBs   (300–340)
         Layer 3 : IAM, Active Directory         (400–460)
         Layer 4 : Networks, VMs, Containers     (500–580)
         Layer 5 : Physical, Sites, Racks         (600–675)
↓ DOWN  ═══════════════════════════════════════════════  toward 675 (Physical)

The directional buttons work as follows:

Button Direction Meaning Example from HR-Solution
↑ Up Toward upper layers Goes up toward the business layer: processes and actors that use this asset Application → HR Process → HR Director actors
↓ Down Toward lower layers Goes down toward infrastructure: servers and networks that support this asset Application → VM → Physical server → Rack → Site
↕ Both Bidirectional Complete view: who uses this asset AND what it relies on Process ← Application → Server → Physical

Usage

Step-by-step usage instructions.

Exploration

Step 1 ─ Enter a filter (optional)
         └─ "Filter" field: e.g. "Applications"
            Restricts the object list to a specific view
            ⚠️ See Filter section: the filter also limits which assets appear in the graph

Step 2 ─ Select the starting asset
         └─ "Object" dropdown: e.g. "HR-Solution"

Step 3 ─ Add the asset to the graph
         └─ Click the green "Add" button
            The asset appears in the graph window

Step 4 ─ Select the asset in the graph
         └─ Click the asset icon in the graph area
            (it must be active / selected)

Step 5 ─ Choose the direction
         └─ Click: ↑ Up  |  ↓ Down  |  ↕ Both

Step 6 ─ Set the number of levels
         └─ Numeric picker: 1 to 5
            Start with 1 or 2 for highly connected assets

Step 7 ─ Deploy
         └─ Click the green "Deploy" button
            The graph builds with the relationships found

Step 8 ─ Iterate (optional)
         └─ Click another node in the graph
            Repeat from step 4

Dependencies

Step 1 ─ Enter a filter (optional)
         └─ "Filter" field: e.g. "Applications"
            ⚠️ See Filter section: the filter also limits which assets appear in the graph

Step 2 ─ Select the starting asset
         └─ "Object" dropdown: e.g. "HR-Solution"

Step 3 ─ Choose the direction
         └─ Click: ↑ Up  |  ↓ Down  |  ↕ Both

Step 4 ─ Set the number of levels
         └─ Numeric picker: 1 to 5
            Start with 1 or 2 for highly connected assets

Step 5 ─ Analyze
         └─ Click the green "Analyze" button
            The layered graph builds with the dependencies found
            (no need to select an asset in the graph)

💡 Dependencies mode has no "Add" button: the starting asset is selected directly from the "Object" dropdown.


Examples

Business Impact of an Application

Context: Which business processes and actors depend on HR-Solution?

Parameter Value
Filter Applications
Object HR-Solution
Direction ↑ Up
Levels 3

The Applications filter is sufficient here since the ↑ Up navigation stays within the business layer (processes, actors), which is covered by this view.

Expected result:

HR-Solution
 └── APPLI-HR-MOD-1 (module)
      └── HR Management Process
           └── Actor: HR Director

You get the full list of processes and actors that functionally depend on this application.


Infrastructure Supporting an Application

Context: What physical hardware does HR-Solution run on? (BCP, ITIL impact analysis, CMDB)

Parameter Value
Filter Applications + Logical Infrastructure + Physical Infrastructure (or empty)
Object HR-Solution
Direction ↓ Down
Levels 3

An Applications-only filter would exclude servers and racks from the graph. All traversed layers must be included, or the filter left empty.

Expected result:

HR-Solution
 ├── APPLI-HR-SRV-1 (application service)
 └── DB-HR-PROD (database)
      └── VM-APP-HR-01 (logical server)
           └── SRV-PROD-01 (physical server)
                └── RACK-A3 (rack)
                     └── Site DC-Paris

You get the complete chain from the application down to the physical site.


Full Impact Analysis (Before Maintenance / Incident)

Context: Complete view of ALL dependencies of HR-Solution before a maintenance window.

Parameter Value
Filter (empty)
Object HR-Solution
Direction ↕ Both
Levels 3

The filter must be empty so that all layers (business, application, physical infrastructure) are simultaneously visible in the graph.

Expected result:

         Macro-process: Personnel Management
              └── HR Process → HR Director actor
                       ↑
              [HR-Solution]  ← central asset
                       ↓
         APPLI-HR-SRV-1 · APPLI-HR-MOD-1 · DB-HR-PROD
              └── Logical servers → Physical → Racks → Site

Ideal for architecture dossiers and impact analyses.


Choosing the Direction by Use Case

Use case Direction Layers traversed
Who uses this application? ↑ Up Application → Processes → Actors
What does this server run on? ↓ Down Logical server → Physical → Rack → Site
Full impact analysis ↕ Both Business ↔ Application ↔ Infrastructure
Which processes depend on a database? ↑ Up Database → Application → Processes
Which network equipment is under a VLAN? ↓ Down VLAN → Switch → Physical router
Map an AD domain ↕ Both AD Forest → Domain → Admin Zone → Servers
Which assets does an actor use? ↓ Down Actor → Process → Application → Infrastructure

Mapping

Mercator Asset BPMN 2.0 ArchiMate 3.1 TOGAF
entities (100) Pool / Lane Business Actor Organizational Unit
macro-processes (200) Process (level 1) Business Process Business Function
processes (210) Sub-Process Business Process Business Service
activities (220) Task / Activity Business Function Business Function
tasks (240) Task (atomic) Business Interaction
actors (250) Lane / Performer Business Role Business Actor
information (260) Data Object Business Object Data Entity
applications (310) Application Component Application Component
application-services (320) Application Service Application Service
databases (340) Data Store Data Object Data Store
logical-servers (580) System Software Platform Service
sites (600) Location Geography
physical-servers (615) Device Technology Component

ArchiMate Relations in Mercator

ArchiMate Relation Direction 🕸️ ⟁ Mercator Example
Serving ↑ Up Application Service serves a Business Process
Realization ↕ Both Application realizes a Business Service
Assignment ↓ Down Logical Server assigned to Physical Server
Composition ↓ Down Site contains Buildings contains Racks
Association ↕ Both Application associated with Database

Best Practices

Tool Recommendations

Start at level 1 or 2
For highly connected assets (e.g. a central application), starting at 1 or 2 levels prevents an unreadable graph. Increase progressively from there.

Use the filter with care
As explained in the Filter section, the filter controls not only the list of available objects but also the assets visible in the graph. For cross-layer exploration or analysis, include all relevant views or leave the filter empty.

"Physical" Toggle (Exploration mode only)
This toggle, available only in 🕸️ Exploration mode, enables or disables the display of physical network links (WAN/LAN/MAN). When disabled, exploration stays at the logical level. It does not constitute a third visualisation mode.

Asset Entry Recommendations

  • Follow Mercator's numbering: create physical layer assets (600+) before associating them with logical layers.
  • Physical containment relationships must be entered in order: Site → Building → Bay → Physical Server.
  • A logical-server (VM) must always be linked to a physical-server for ↓ Down navigation to work correctly.
  • certificates (570) are cross-layer: associate them with both applications AND logical servers for complete exploration or analysis.
  • external-connected-entities (540) can be attached to the business layer (partners) or the network layer (connections) depending on context.

Use Cases by User Profile

Profile Preferred direction Typical use case
Enterprise Architect ↑ Up Trace alignment between infrastructure and business processes
Infrastructure Architect ↓ Down Identify the physical chain supporting an application
CISO / Risk Manager ↕ Both Map dependencies for risk analysis
CMDB Manager ↓ Down Verify completeness of physical chaining
Crisis Manager / BCP ↕ Both Impact analysis before/after an incident