Skip to content

MemoryUsageMonitoring.md

Latest commit

 

History

History
117 lines (83 loc) · 3.97 KB

MemoryUsageMonitoring.md

File metadata and controls

117 lines (83 loc) · 3.97 KB

Memory Usage Monitoring

This document explains how the profiler monitors memory usage to trigger profiling sessions when memory exceeds a configured threshold.

Overview

The memory monitoring pipeline follows the same three-layer architecture as CPU monitoring:

  1. Metrics Reading — reads current memory usage on demand
  2. Baseline Tracking — aggregates readings into a rolling average every 30 seconds
  3. Scheduling Policy — compares the average against the threshold every 5 seconds
Memory Reader (on-demand)
        ↓
Baseline Tracker (30s rolling average)
        ↓
Memory Monitoring Policy (5s polling)
        ↓ memoryUsage > threshold?
Start Profiling Session

How Memory is Measured

Windows

Uses the Win32 API to get total physical memory and Process.WorkingSet64 for current process memory. Returns usage as:

Memory% = (processWorkingSet / totalPhysicalMemory) × 100

Linux

Reads from /proc/meminfo to get system-wide total and available memory:

Memory% = (1 - MemAvailable / MemTotal) × 100

This measures system-wide memory pressure, not just the current process.

Unlike CPU sampling, memory readings are computed synchronously on demand — there is no separate sampling thread. Each time the baseline tracker requests a new value, a fresh reading is taken directly.

Baseline Tracking (Sliding Window)

Memory readings are aggregated into a smoothed average using the same circular buffer mechanism as CPU monitoring.

Configuration

Setting Default Description
MemoryRollingHistorySize 10 minutes Total window size for the circular buffer
MemoryRollingHistoryInterval 30 seconds How often a reading is recorded into the buffer
MemoryAverageWindow 30 seconds Sliding window for computing the reported average

How it works

  1. Every 30 seconds, the latest memory reading is recorded into a circular buffer (sized at 10min / 30s = 20 slots).
  2. The average is computed over the most recent 30 seconds of entries.
  3. This average is what the scheduling policy uses for threshold comparison.

Memory Monitoring Scheduling Policy

The policy checks the memory baseline every 5 seconds:

  • If memoryUsage > MemoryThreshold → starts a profiling session, then enters a cooldown period.
  • If memoryUsage ≤ MemoryThreshold → stands by for 5 seconds, then checks again.

Configuration

These settings can be configured via app settings or environment variables:

Setting Default App Setting Key
Memory Threshold 80% ApplicationInsightsProfiler_MemoryThreshold
Trigger Cooldown 14400s (4h) ApplicationInsightsProfiler_Memory_TriggerCooldownInSeconds
Profiling Duration 30s ApplicationInsightsProfiler_Memory_ProfilingDurationInSeconds
Enabled true ApplicationInsightsProfiler_MemoryTriggerEnabled

User-level overrides can also be set via code:

builder.Services.AddServiceProfiler(options =>
{
    options.MemoryTriggerThreshold = 70;  // Trigger at 70% memory
    options.MemoryTriggerCooldown = TimeSpan.FromMinutes(30);
});

Diagnostic Logging

To observe the memory monitoring pipeline, enable debug logging:

{
  "Logging": {
    "LogLevel": {
      "Microsoft.ServiceProfiler": "Debug",
      "Microsoft.ApplicationInsights.Profiler": "Debug"
    }
  }
}

Or via environment variables (e.g., in Azure App Service):

Logging__LogLevel__Microsoft.ServiceProfiler=Debug
Logging__LogLevel__Microsoft.ApplicationInsights.Profiler=Debug

Key log messages to look for:

  • "Get memory usage: free/total: X/Y" — raw readings from the provider
  • "Baseline update: metric=X, old=Y, new=Z" — 30-second rolling average updates
  • "Getting current Memory usage: X" — the value the scheduling policy sees
  • "MemoryMonitoringSchedulingPolicy request delay for ... StartProfilingSession" — threshold exceeded, profiling triggered