Pipeline Architecture

S2DCartographer is built as a sequential pipeline with a shared session cache at its core. Understanding how the pieces fit together helps when using the module programmatically or troubleshooting unexpected results.

Pipeline Overview

flowchart TD
    A(["Connect-S2DCluster"]) --> B[("Module Session\nS2DSession")]

    B --> C["Get-S2DPhysicalDiskInventory"]
    B --> D["Get-S2DStoragePoolInfo"]
    B --> E["Get-S2DVolumeMap"]
    B --> F["Get-S2DCacheTierInfo"]

    C --> G[("S2DClusterData")]
    D --> G
    E --> G
    F --> G

    G --> H["Get-S2DCapacityWaterfall"]
    G --> I["Get-S2DHealthStatus"]
    H --> G
    I --> G

    G --> J["New-S2DReport"]
    G --> K["New-S2DDiagram"]

    J --> L[/"HTML · Word · PDF · Excel"/]
    K --> M[/"SVG Diagrams"/]

    N(["Invoke-S2DCartographer"]) -. "orchestrates" .-> A
    N -. "orchestrates" .-> C
    N -. "orchestrates" .-> D
    N -. "orchestrates" .-> E
    N -. "orchestrates" .-> F
    N -. "orchestrates" .-> H
    N -. "orchestrates" .-> I
    N -. "orchestrates" .-> J
    N -. "orchestrates" .-> K

---

Module Session (S2DSession)

Connect-S2DCluster creates a script-scoped hashtable ($Script:S2DSession) that all other commands read and write:

$Script:S2DSession = @{
    ClusterName   = 'c01-prd-bal'
    ClusterFqdn   = 'c01-prd-bal.contoso.com'
    Nodes         = @('n01','n02','n03','n04')
    CimSession    = <CimSession>
    PSSession     = <PSSession>
    IsConnected   = $true
    IsLocal       = $false
    CollectedData = @{
        PhysicalDisks      = $null   # populated by Get-S2DPhysicalDiskInventory
        StoragePool        = $null   # populated by Get-S2DStoragePoolInfo
        Volumes            = $null   # populated by Get-S2DVolumeMap
        CacheTier          = $null   # populated by Get-S2DCacheTierInfo
        CapacityWaterfall  = $null   # populated by Get-S2DCapacityWaterfall
        HealthChecks       = $null   # populated by Get-S2DHealthStatus
        OverallHealth      = $null   # set by Get-S2DHealthStatus
    }
}

Every collector checks CollectedData before querying the cluster. If data is already present, it is returned from cache without a network round-trip. This means calling Get-S2DCapacityWaterfall twice only hits the cluster once.

---

S2DClusterData Object

Invoke-S2DCartographer -PassThru and the individual collectors all build toward the same S2DClusterData object passed to report and diagram generators:

Property	Type	Source
ClusterName	string	Session
NodeCount	int	Session
PhysicalDisks	S2DPhysicalDisk[]	Get-S2DPhysicalDiskInventory
StoragePool	S2DStoragePool	Get-S2DStoragePoolInfo
Volumes	S2DVolume[]	Get-S2DVolumeMap
CacheTier	S2DCacheTier	Get-S2DCacheTierInfo
CapacityWaterfall	S2DCapacityWaterfall	Get-S2DCapacityWaterfall
HealthChecks	S2DHealthCheck[]	Get-S2DHealthStatus
OverallHealth	string	Get-S2DHealthStatus
CollectedAt	datetime	Orchestrator

---

Collector Call Order

Invoke-S2DCartographer calls collectors in dependency order:

1. Get-S2DPhysicalDiskInventory — no dependencies
2. Get-S2DStoragePoolInfo — no dependencies
3. Get-S2DVolumeMap — no dependencies
4. Get-S2DCacheTierInfo — no dependencies
5. Get-S2DCapacityWaterfall — depends on 1, 2, 3
6. Get-S2DHealthStatus — depends on 1, 2, 3, 4, 5

When called individually, each collector auto-invokes its prerequisites if their results are not already cached.

---

CIM vs PS Sessions

S2DCartographer uses two session types in parallel:

Session	Used For
CimSession	Storage queries (Get-PhysicalDisk, Get-StoragePool, Get-VirtualDisk, etc.)
PSSession	Node-level queries that require per-node execution (disk inventory loop, reliability counters)

Both are established by Connect-S2DCluster and closed by Disconnect-S2DCluster. If you pass an existing CimSession, no PSSession is created — node-level queries fall back to per-node Invoke-Command using the CIM session's credentials.

---

Using the Pipeline Manually

You do not need Invoke-S2DCartographer — all commands are composable:

Connect-S2DCluster -ClusterName "c01-prd-bal" -Credential $cred

$disks  = Get-S2DPhysicalDiskInventory
$pool   = Get-S2DStoragePoolInfo
$vols   = Get-S2DVolumeMap
$wf     = Get-S2DCapacityWaterfall
$health = Get-S2DHealthStatus

# Build the data object manually
$data = [S2DClusterData]::new()
$data.ClusterName      = 'c01-prd-bal'
$data.PhysicalDisks    = $disks
$data.StoragePool      = $pool
$data.Volumes          = $vols
$data.CapacityWaterfall = $wf
$data.HealthChecks     = $health

New-S2DReport  -InputObject $data -Format Html -OutputDirectory "C:\Reports\"
New-S2DDiagram -InputObject $data -DiagramType All -OutputDirectory "C:\Reports\"

Disconnect-S2DCluster