Last reviewed: April 9, 2025
π In this article:
- Scoring Model
Scoring Model
ArchTechLytics uses a TechLytics-based scoring model to evaluate the architectural integrity of your systems. This model breaks away from traditional static checklists, instead analyzing real-time cloud configurations and aligning them with your declared system intent.
π What is a TechLytic?
A TechLytic is an atomic unit of architectural insight. It represents a single architectural best practice or concern, such as:
- Is the VM part of an availability set?
- Are all data disks encrypted?
- Does backup frequency meet the systemβs declared RPO?
Each TechLytic has:
TechLyticId
β unique reference (e.g.,reliability-vm-haszoneredundancy
)ArchitecturalPillar
β mapped pillar like Reliability, Security, etc.State
β outcome of the check (Passed, Failed, PartiallyPassed, Info)Score
β actual score achievedMaxScore
β possible scoreTechLyticType
β determines if it affects the score (Scoring) or is advisory (Informational)LastEvaluated
β timestamp for transparency
π‘ Note: Only a subset of significant resources are currently scored.
ArchTechLytics monitors many important resources for architecture awareness,
but scoring support requires detailed TechLytics to be implemented for each type.
Weβre actively expanding scoring coverage, initially supporting Azure VMs.
π§ How Scoring Works
1. TechLytic Metadata-Driven Evaluation
Every resource is evaluated based on its type and the associated TechLyticMetadata
. Metadata defines:
- Human-readable name and description
- Scoring pillar/category
- Microsoft Learn references
- Architectural recommendations
Evaluations are decoupled from hardcoded logic β enabling ArchTechLytics to evolve as guidance evolves.
2. System-Level vs. Resource-Level TechLytics
Scoring happens across two dimensions:
𧱠Resource-Level TechLytics
These reflect how each cloud resource is configured β like VM backup frequency, disk encryption, or availability zones.
π System-Level TechLytics
These reflect the architectural intent declared in the System Profile β such as RTO/RPO values, threat model existence, or backup vault presence.
ArchTechLytics combines both dimensions to assess the whole system.
3. Aggregation and Weighting
The scoring engine aggregates TechLytics across:
- Resource Type
- Category (e.g., Compute, Storage, Network)
- Cloud Provider
- Architectural Pillar
Each unit tracks its own scoring history for time-based analytics.
Final scoring uses a weighted aggregation:
Area | Weight |
---|---|
Architectural Pillars (Intent) | 50% |
Cloud Resources (Implementation) | 50% |
The result is normalized to a 0β100 score.
π Transparency by Design
- You can inspect every TechLytic and its outcome.
- Informational TechLytics highlight things that arenβt scored (yet), but are architecturally significant.
- Failed TechLytics link directly to recommendations and Microsoft guidance.
- The scoring model is transparent and explainable β every score can be traced back to its underlying TechLytics.
βοΈ Adaptive Scoring
As new TechLytics are added or metadata evolves, scores automatically reflect the latest architectural standards β without any system owner needing to rewire their evaluations.
β Currently Scored Resource Types
The following table lists cloud resource types that ArchTechLytics actively scores today.
Cloud Provider | Product | Resource Type | Category |
---|---|---|---|
Azure | Virtual Machines | Microsoft.Compute/virtualMachines | Compute |
π‘ We are actively expanding resource scoring coverage. Scored resources will be included here once their TechLytics are implemented.
π Planned Scoring Coverage
The following resource types are significant to architectural integrity and are planned for future TechLytic scoring support.
Cloud Provider | Product | Resource Type | Category | Target ETA |
---|---|---|---|---|
Azure | Virtual Machine Scale Sets | Microsoft.Compute/virtualMachineScaleSets | Compute | CY 2025 |
Azure | App Service | Microsoft.Web/sites | Compute | CY 2025 |
Azure | SQL Database | Microsoft.Sql/servers | Data | CY 2025 |
Azure | Key Vault | Microsoft.KeyVault/vaults | Security | CY 2025 |
Azure | Storage Account | Microsoft.Storage/storageAccounts | Storage | CY 2025 |
Azure | Cosmos DB | Microsoft.DocumentDB/databaseAccounts | Data | CY 2025 |
Azure | Kubernetes Service (AKS) | Microsoft.ContainerService/managedClusters | Compute | CY 2025 |
Azure | Application Gateway | Microsoft.Network/applicationGateways | Network | CY 2025 |
Azure | Virtual Network | Microsoft.Network/virtualNetworks | Network | CY 2025 |
Azure | Network Security Groups (NSGs) | Microsoft.Network/networkSecurityGroups | Security | CY 2025 |
Azure | Logic Apps | Microsoft.Logic/workflows | Integration | CY 2025 |
Azure | Log Analytics Workspace | Microsoft.OperationalInsights/workspaces | Management | CY 2025 |
Azure | Entra ID Applications | Microsoft.Graph/applications | Identity | CY 2025 |
π οΈ We are actively expanding TechLytic support across these resources based on best practices, demand, and architectural relevance.
β‘οΈ Learn more about: