Application Monitoring( AIAPM)

Go to AI Monitoring → AI APM

You will see tabs for:

• Application
• Microsoft IIS
• API Monitoring

By default, the Application Dashboard will displays.

The AI Application Performance Monitoring (AI APM) dashboard provides real-time insights into application health, performance, and anomalies for selected Applications.

Top-Level Metrics:

• Total Transactions
  Displays the number of transactions handled over a recent time window, offering a quick snapshot of system activity.

• Active Services
  Indicates how many application services are currently under monitoring.

• Databases
  Shows the count of connected databases being tracked for performance.

• Anomalies Detected
  Highlights any detected irregularities in performance or behavior that may require attention.

• Average Error Rate
  Reflects the overall error percentage observed across monitored transactions.

Service Overview Section

This table provides insights into each monitored service, including:

• Service Name, Host Identifier

• Average Latency – time taken to process requests

• Throughput – requests handled per minute

• Error Rate – percentage of failed transactions

• Anomalies – count of unusual events or performance issues

Recent Anomalies

Lists recent anomalies detected by the system, including:

• Date and Time of detection

• Type of anomaly 

• Affected Service

• Details for further investigation

Status Codes Distribution

A visual chart representing the distribution of HTTP status codes returned by services, helping to quickly assess success and failure rates of requests.

The Applications tab in AI APM provides a service-level breakdown of application health and performance. It enables users to visualize and monitor individual services, latency, throughput, and response status distribution.

 Services Overview

• Displays a donut chart representing all monitored services.

• Each service is color-coded and listed for quick reference.

• Helps identify how many services are actively being tracked.

Host View vs Service View

• Host View:
  Focuses on metrics aggregated by host. Useful for identifying host-specific issues such as latency spikes or processing delays on a particular machine.

• Service View:
  Filters and displays metrics by service, giving a clearer picture of how individual services are performing regardless of the host they run on.

• A dropdown allows selection of specific hosts or services for focused monitoring.

Application Metrics

• Offers metrics like Application Latency, allowing users to view performance trends over time.

• Supports switching between Host View and Service View to isolate specific metrics.

• Dropdown filters allow choosing specific hosts or services for detailed analysis.

Throughput Graph

• Visual representation of Transactions Per Minute (TPM) across the selected time window.

• Shows how transaction volume changes over time for the chosen service or host.

Status Codes Distribution

• Bar chart showing the distribution of HTTP response status codes (e.g., 200, 400, 500).

• Allows users to understand response quality and detect errors.

Status Code Drill-Down:

When a status code bar (e.g., 200) is clicked:

• A detailed Status Code Transactions Table appears below.

• This table includes:

  • Timestamp

  • Service Name

  • Host

  • Transaction ID

  • Trace ID (with a link to view full trace)

  • Response Code

  • Duration

  • URL (if available)

This feature helps with granular debugging and traceability of individual transactions contributing to a specific response code.

The Service Map visually represents the relationships and dependencies between various services and components within the application ecosystem. It helps users quickly understand service interactions, communication flow, and architecture topology.

Service Map Layout

• Nodes represent individual services.

• Edges/Connections show communication paths between services and dependencies.

• Layout Options (Right Panel):

  • Mini map Toggle – Enable/disable mini navigation map.

  • Layout Type – Hierarchical (default) or other visual flow options.

  • Direction – Choose flow direction (e.g., left-to-right, top-down).

  • Node Type Filters – Filter by tech stack or service types.

When you click on a connection (edge) between two services:

1. A host selector appears, allowing you to focus on a specific host involved in the communication.

2. A detailed Dependency Transactions Table opens below, showing:

   • Timestamp

   • Service name

   • Host

   • Transaction name

   • Duration

   • Status code 

   • Trace access (View Timeline)

Timeline & Transaction Details

Clicking "View Timeline" on a transaction entry brings up:

• Timeline Graph:
  Visual representation of transaction duration, phases, and where time is spent.

• Transaction Details Panel:
  Provides technical metadata including:

  • Container and process information

  • Agent and source IP

  • URL details

  • Labels and observer data

This data is essential for root-cause analysis, tracing delays, and understanding infrastructure-level behavior.

The Traces tab provides deep visibility into individual application requests and how they are processed by services. This is critical for tracing performance bottlenecks and debugging specific transaction paths.

 Trace Table

Each row in the table represents a single trace captured from a service and includes:

• Timestamp – When the trace occurred

• Trace ID – Unique identifier for the trace

• Service Name – The service involved in the trace

• Duration – Time taken for the transaction

• URL – The endpoint associated with the trace

• Transactions Count – Number of transactions in the trace

• Spans Count – Number of sub-operations or spans within the trace

1. Click on a Trace Row: Opens a Timeline Graph below the table, showing the time breakdown of operations involved in that trace.
2. Click on the Graph Bar (Span or Transaction): Displays Transaction Details in a side panel

This drilldown helps users trace individual service calls, analyze latency, and debug operational issues.

The Infrastructure tab provides detailed visibility into how application services are distributed across infrastructure and how they're consuming system resources like CPU and memory. 

Applications Table

• Lists all deployed services and their associated:

  • Application Name

  • Service Name

  • Host Name

  • IP Address

Helps understand the infrastructure footprint of applications.

Service Distribution Chart

• A donut chart visualizes the number of monitored services.

• Color-coded segments represent different services for easier identification.

Host View vs Service View

• Host View:
  Aggregates metrics per host machine.

• Service View:
  Filters metrics per individual service.

You can toggle between views and use the dropdown to focus on specific services.

1. CPU Usage Graph

• Shows percentage of CPU usage over time.

• Helps identify CPU-intensive behaviors or potential overloads.

2. Memory Usage Graph

• Displays memory consumption trends across services or hosts.

• Highlights any memory anomalies visually.

Anomaly Detection & Root Cause Analysis

When you click on an anomaly indicator (e.g., red dot in the memory usage graph):

• A Root Cause Analysis modal is displayed with:

  1. What’s Happening – Summary of the application’s current state and observed behavior.

  2. Likely Cause – Explanation of the root cause based on the observed metrics.

  3. Recommended Fix – Suggestions for resolving or monitoring the issue further (e.g., profiling, logging, scaling).

This enables quick diagnosis and action on performance degradations or spikes.

The AI Insights tab provides intelligent anomaly detection based on system and application performance metrics. It helps identify unusual patterns such as memory spikes, high CPU usage, or degraded performance using AI-driven logic.

AI Anomalies Table

Each row in the anomaly list includes:

• Timestamp – When the anomaly was detected

• Service Name – Affected microservice

• Anomaly Type – Type of metric anomaly (e.g., CPU, memory)

• Details – Quick view of what triggered the anomaly (e.g., high CPU usage)

The table supports sorting, pagination, and quick search filtering for faster investigation.

When you click on a specific anomaly, a detailed Root Cause Analysis panel expands below the table.

Database Tab:

Database Transactions Table

• Lists of active queries and interactions between services and databases
• Displays database name, type (e.g., MSSQL), action performed (e.g., query), and latency

Database Latency Graph

• Tracks average query latency over time per database
• Helps identify slow or overloaded databases Database Throughput Graph
• Show transactions per minute (TPM) handled by each database
• Useful for analyzing load trends and spikes

Additional Metrics Panels:

• Database Space Usage
• User Connections
• Batch Requests
• Cache Hit Ratio

RUM Tab:

• End-user latency trends across pages
• Breakdown of backend vs. frontend load times
• Core Web Vitals like LCP, INP, and CLS
• Detailed RUM traces and span breakdowns for slow interactions

Synthetic monitoring Tab:

• Monitor Status Summary - Displays total monitors along with their statuses (Up, Down, Disabled, Pending)
• Test Statistics - Total test executions, number of errors, and triggered alerts
• Duration Overview - Visual timeline of test durations to spot trends and performance shifts
• Errors & Alerts - Graphs showing the count of failed test runs and triggered alerts by monitor
• Test Runs by Time - Histogram representing the frequency of synthetic test executions
• Monitor List Table - Detailed list of all configured monitors