Home › Studies › AZ-305 › Identity & Governance › Log Analytics Workspace Design

Log Analytics Workspace Design

Single vs multiple workspaces · Access modes · Table plans · Retention

Practice questions for this topic →

When to create another workspace

CriterionEffect
DefaultStart with ONE workspace; use the fewest that meet requirements; data volume doesn't limit performance
Data residency / regionA workspace lives in one region, so residency rules may force more
Multiple Entra tenantsSeveral data sources can send only to a workspace in the same tenant, so usually one per tenant
Security team separationOptional dedicated Sentinel workspace
CostConsolidation can unlock commitment tiers; per-region workspaces can cut egress

Access control

ModeWhat you see
Workspace-contextAll tables you have permission for in the workspace (Logs opened from the workspace)
Resource-contextOnly logs for resources you can access (Logs opened from a resource), no workspace role needed
Table-level RBACRestrict specific tables

Table plans and retention

PlanUse forQueryAlerts
AnalyticsContinuous monitoring, latency-sensitiveFull KQL; analytics retention up to 2 yearsYes
BasicTroubleshooting, incident responseFixed 30 days; older data via search jobsLimited
Auxiliary / LakeHigh-volume verbose data, cheapestQueryable for the whole retention periodNo

Exam Tips

Tick each tip once you can answer it without looking.

“Default design for a new monitoring solution” → One Log Analytics workspace. Trap: Don't create one per app or team without a requirement.

“EU logs must stay in the EU; US logs in the US” → One workspace per required region.

“App owners may only see logs for their own resources, minimal admin” → Resource-context access (permissions on the resources).

“Hide security tables from operations staff in a shared workspace” → Table-level RBAC.

“Keep audit logs 7 years at lowest cost, rarely queried” → Long-term (total) retention up to 12 years, or archive to Storage. Trap: Analytics retention max is 2 years.

“Huge volume of verbose logs, cheapest ingest, no alerts needed” → Auxiliary / Lake table plan.

“Debug logs, queried occasionally during incidents” → Basic table plan. Trap: Basic queries only cover 30 days.

“Large predictable daily ingestion, reduce cost” → Commitment tier (consolidate into fewer workspaces).

“MSP must query workspaces in many customer tenants” → Azure Lighthouse (cross-tenant queries).

“Switch a table's plan repeatedly” → Limited to one plan switch per table per week. Trap: Setting a plan needs workspace write and table write permissions (Log Analytics Contributor).

“Shorten a table's total retention by mistake” → Data is removed only after a 30-day wait, so you can revert. Trap: Total retention can reach 12 years.

“Lower analytics retention below 31 days to save money” → No saving: 31 days are included in the ingestion price. Trap: It can go as low as 4 days through the API or CLI.

“Move a table from Analytics to Auxiliary” → Alerts stop working for that table; queries and summary rules change. Trap: Basic queries cover only 30 days.

Sources

← Azure Monitor & AgentsLog Routing (Diagnostic Settings) →