agentleFS
PatrickGallucciPatrickGallucci/fabric-skills/skills/fabric-pbi-perf-remediate/SKILL.md

fabric-pbi-perf-remediate

Skill17 starsChanged 9 months agoMITView on GitHub

Diagnose and resolve Power BI performance issues in Microsoft Fabric. Use when reports load slowly, DAX queries are inefficient, semantic models need optimization, DirectQuery is underperforming, capacity is throttled, visuals render slowly, or data refresh is slow. Covers Performance Analyzer, DAX Studio, Fabric Capacity Metrics app, VOrder, query folding, incremental refresh, storage modes, Best Practice Analyzer, and Spark resource profiles for Power BI workloads.

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 9 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Look for overload (>100% utilization), throttling events, or queued operations

    line 47
  2. Distinctive:

    Run query with Server Timings enabled (DAX Studio: Server Timings tab)

    line 56
  3. Distinctive:

    Run Best Practice Analyzer (Tabular Editor or Fabric notebook)

    line 68
  4. Distinctive:

    Remove unused columns and tables

    line 70
  5. Distinctive:

    Check for excessive cross-filtering between visuals

    line 87
  6. Distinctive:

    Consider Composite model (Import aggregations + DirectQuery detail)

    line 105
  7. Distinctive:

    Consider Autoscale Billing for Spark if bursty workloads exist

    line 118
  8. Distinctive:

    Run the diagnostic PowerShell script to collect environment and configuration data:

    line 135
  9. Distinctive:

    Use the performance-report-template.md to document findings and recommendations from a performance audit.

    line 151

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Power BI Performance remediate in Microsoft Fabric

Systematic toolkit for diagnosing, analyzing, and resolving Power BI performance bottlenecks across the Microsoft Fabric platform. Covers semantic model optimization, DAX tuning, capacity management, DirectQuery diagnostics, and report design best practices.

When to Use This Skill

  • Power BI reports are slow to load or interact with
  • DAX queries take too long to execute
  • Semantic model refresh is slow or timing out
  • Fabric capacity is throttled or overutilized
  • DirectQuery reports have high latency
  • Visuals render slowly or time out
  • Users report intermittent performance degradation
  • Migrating to Fabric and need to optimize for the new platform
  • Planning capacity sizing for Power BI workloads
  • Conducting a performance audit or health check

Prerequisites

ToolPurposeRequired
Power BI DesktopPerformance Analyzer, DAX query viewYes
DAX StudioAdvanced DAX profiling and server timersRecommended
Fabric Capacity Metrics AppCapacity utilization monitoringYes (admins)
Tabular Editor / Best Practice AnalyzerSemantic model analysisRecommended
SQL Server ProfilerDirectQuery trace analysisOptional
PowerShell 7+Automation scripts included in this skillOptional

Step-by-Step Workflows

Workflow 1: Initial Performance Triage

Determine where the bottleneck lives before diving deep.

  1. Reproduce the issue in Power BI Desktop with Performance Analyzer enabled
    • View ribbon > Performance Analyzer > Start recording > Refresh visuals
  2. Categorize each visual by its dominant cost:
    • DAX query duration > 500ms → Investigate semantic model / DAX
    • Visual display duration > 500ms → Investigate report design
    • Other duration > 500ms → Investigate data source / gateway
  3. Check capacity health using the Fabric Capacity Metrics app
    • Look for overload (>100% utilization), throttling events, or queued operations
  4. Route to the appropriate deep-dive workflow below

Workflow 2: DAX Query Optimization

See dax-optimization-patterns.md for a comprehensive catalog of anti-patterns and fixes.

  1. Capture slow DAX from Performance Analyzer (copy query from visual)
  2. Open DAX query view in Power BI Desktop (or DAX Studio)
  3. Run query with Server Timings enabled (DAX Studio: Server Timings tab)
  4. Analyze the breakdown:
    • Formula Engine (FE) time: DAX calculation overhead
    • Storage Engine (SE) time: Data scan / retrieval overhead
    • SE queries count: High count indicates poor query plan
  5. Apply optimization patterns from the reference guide
  6. Re-test and compare timings

Workflow 3: Semantic Model Optimization

See capacity-optimization.md for Fabric-specific tuning.

  1. Run Best Practice Analyzer (Tabular Editor or Fabric notebook)
  2. Address findings by priority:
    • Remove unused columns and tables
    • Fix incorrect data types (text dates, high-precision decimals)
    • Replace calculated columns with calculated measures where possible
    • Reduce cardinality on high-cardinality columns
  3. Evaluate storage mode (Import vs DirectQuery vs Composite)
  4. Configure incremental refresh for large fact tables
  5. Enable VOrder for read-heavy Power BI workloads in Fabric:

    spark.sql.parquet.vorder.default=true

    Or use the readHeavyForPBI resource profile at the environment level.

Workflow 4: Report Design Optimization

  1. Audit visual count per page (target: 8 or fewer interactive visuals)
  2. Identify high-cardinality visuals (tables/matrices with thousands of rows)
  3. Check for excessive cross-filtering between visuals
  4. Evaluate filter context complexity (many slicers, complex RLS)
  5. Consider:
    • Bookmarks + drill-through instead of dense pages
    • Pre-aggregated measures instead of visual-level calculations
    • Paginated reports for large tabular exports

Workflow 5: DirectQuery Performance

See directquery-tuning.md for detailed guidance.

  1. Enable Performance Analyzer and identify slow DirectQuery visuals
  2. Locate trace files for SQL analysis:
    • File > Options > Diagnostics > Open traces folder
    • Find FlightRecorderCurrent.trc in the active workspace
  3. Open trace in SQL Server Profiler and filter by DirectQuery Begin/End
  4. Analyze generated SQL for inefficient patterns
  5. Optimize at the source (indexes, views, materialized tables)
  6. Consider Composite model (Import aggregations + DirectQuery detail)

Workflow 6: Capacity Monitoring and Sizing

See capacity-optimization.md for detailed guidance.

  1. Install and configure the Fabric Capacity Metrics app
  2. Monitor key metrics:
    • Interactive vs background operation split
    • Throttling events and queue depth
    • Per-item compute consumption
  3. Identify top consumers and optimize or reschedule them
  4. Right-size capacity SKU based on measured utilization
  5. Consider Autoscale Billing for Spark if bursty workloads exist

Quick Reference: Common Fixes

SymptomLikely CauseQuick Fix
All visuals slowCapacity overloadedScale up SKU or reduce concurrency
Single visual slowInefficient DAX measureProfile in DAX Studio, rewrite measure
Slow after slicer changeHigh cardinality filterReduce distinct values or use Top N
Slow first load, fast afterCold cacheEnable query caching; check refresh schedule
Slow in Service, fast in DesktopGateway bottleneck or capacityCheck gateway logs and capacity metrics
Refresh takes hoursNo incremental refreshEnable incremental refresh on fact tables
DirectQuery timeoutsSource query too slowAdd indexes; consider Import aggregations
Intermittent slownessCapacity throttlingReview Capacity Metrics app for spikes

Automation Scripts

Run the diagnostic PowerShell script to collect environment and configuration data:

# Collect Power BI workspace and dataset metadata for analysis
./scripts/Invoke-PBIPerformanceAnalysis.ps1 -WorkspaceId "<workspace-guid>"

Analyze DAX query patterns from a semantic model:

# Extract and evaluate DAX measures for common anti-patterns
./scripts/Get-DAXQueryMetrics.ps1 -DatasetId "<dataset-guid>" -WorkspaceId "<workspace-guid>"

Performance Assessment Template

Use the performance-report-template.md to document findings and recommendations from a performance audit.

remediate

IssueResolution
Performance Analyzer shows no dataEnsure you clicked "Start recording" before refreshing
DAX Studio cannot connectCheck XMLA endpoint is enabled on capacity (requires P1/F64+)
Capacity Metrics app not availableApp requires admin role; install from AppSource
VOrder not improving PBI performanceVerify readHeavyForPBI profile is active; check file format is Delta/Parquet
Best Practice Analyzer missing rulesUpdate to latest Tabular Editor; import community rules from te2.wiki
Scripts fail to authenticateRun Connect-PowerBIServiceAccount first; ensure Power BI Management module installed

References

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.