Last reviewed: April 9, 2025
📘 In this article:
TechLytics: The Core of ArchTechLytics Scoring
At the heart of ArchTechLytics is the concept of a TechLytic — a small but powerful unit of architectural evaluation.
🧱 What is a TechLytic?
A TechLytic represents a single architectural check aligned to a best practice or principle.
Each TechLytic answers a precise question, such as:
- “Is the VM using a managed OS disk?”
- “Is the system’s RPO defined?”
- “Are all data disks encrypted?”
These evaluations are run automatically across your system and cloud resources.
🧬 TechLytic Structure
A TechLytic is composed of:
| Property | Description |
|---|---|
TechLyticId | Unique ID used for mapping to metadata |
ArchitecturalPillar | Mapped pillar (e.g., Reliability, Security) |
State | Result: Passed, Failed, PartiallyPassed, NotApplicable, NotConfigured |
Score / MaxScore | Scored value and maximum possible value |
TechLyticType | Scoring (affects score) or Informational (for visibility only) |
LastEvaluated | Timestamp of evaluation |
🧠 Backed by Metadata
TechLytics are metadata-driven. Each supported resource type has a set of metadata-defined TechLytics that describe:
- Name and Description
- Architectural Pillar
- Recommendation
- Microsoft Learn reference link
This allows ArchTechLytics to adapt as guidance evolves — without rewriting evaluation code.
✨ Scoring vs. Informational
| Type | Description |
|---|---|
Scoring | Directly impacts system or resource score |
Informational | Surface best practices that aren’t scored (yet) |
For example, being part of a load balancer may not be scored directly, but it’s shown as a passed/failed informational TechLytic.
🕹️ System-Level vs. Resource-Level
| Scope | Description |
|---|---|
| Resource-Level | Tied to a specific cloud resource like a VM, disk, or NSG |
| System-Level | Based on declared system intent (e.g., RPO defined, threat model exists) |
ArchTechLytics combines both to form a holistic architectural evaluation.
🔁 Evaluated Continuously
TechLytics are re-evaluated:
- When changes are detected
- When systems are rescored
- When cloud metadata updates
This ensures living architectural integrity — not just point-in-time compliance.
Next up: