Expert
Prepare and test your skills
Expert
Prepare and test your skills
Cost-Optimized Monitoring for Spring Apps and VMs
Free sample question, answered and explained below. The AZ-305 Designing Microsoft Azure Infrastructure Solutions Exam practice bank has 124 more like it. Try 10 free: instant score, no account.
A large enterprise has deployed a critical, multi-tier application on Azure, comprising Azure Spring Apps microservices, Azure SQL Database, and several Azure Virtual Machines hosting legacy components. They are currently experiencing unexpectedly high data ingestion costs due to default monitoring configurations. The operations team requires a comprehensive monitoring solution that provides deep application performance insights, infrastructure health, and user behavior analytics, while significantly optimizing costs.
The design must fulfill the following specific requirements:
- Collect detailed application telemetry, including custom events and metrics, from Azure Spring Apps.
- Monitor infrastructure performance counters and system logs from Azure Virtual Machines efficiently.
- Implement cost-effective data retention for frequently collected logs that are less frequently queried.
- Ensure critical alerts are configured to be both proactive and cost-efficient.
- Provide unified dashboard visualization for all collected telemetry.
Which integrated monitoring solution design best meets these requirements?
Explanations
Integrate third-party APM tools for Azure Spring Apps and Azure Virtual Machines, forwarding only critical alerts to Azure Event Hubs. Use Azure Resource Health for infrastructure status and Service Health for platform issues. Store all raw telemetry in Azure Storage Accounts for long-term retention and use a custom application to query and visualize data.
Integrate third-party APM tools for Azure Spring Apps and Azure Virtual Machines, forwarding only critical alerts to Azure Event Hubs. Use Azure Resource Health for infrastructure status and Service Health for platform issues. Store all raw telemetry in Azure Storage Accounts for long-term retention and use a custom application to query and visualize data.
What this option is/does
This option proposes a solution heavily reliant on third-party APM tools for data collection, with Azure Event Hubs for forwarding alerts, Azure Resource Health and Service Health for basic health checks, and Azure Storage Accounts for raw data retention. It also suggests a custom application for querying and visualization.
Why it doesn't meet the requirements
This design deviates significantly from leveraging Azure's integrated monitoring capabilities and introduces unnecessary complexity and cost:
- Third-party APM tools: While valid in some scenarios, the question asks for an "integrated monitoring solution design" within the context of Azure, implying a preference for native Azure Monitor services. Relying solely on third-party tools means foregoing the deep integration, native cost optimization features, and unified experience of Azure Monitor.
- Forwarding only critical alerts to Azure Event Hubs: Event Hubs are a streaming data platform, not an alerting solution. While they can be a destination for alerts, they don't generate or manage alerts themselves. This approach would require significant custom development to process events from Event Hubs and trigger actual notifications, making it inefficient and complex. It also implies that non-critical telemetry might not be monitored or alerted upon, which contradicts the need for "comprehensive visibility."
- Azure Resource Health and Service Health: These services provide valuable insights into the health of Azure resources and the Azure platform itself. However, they are high-level status indicators and do not provide the detailed application telemetry (custom events, metrics, traces) or granular infrastructure performance counters needed for deep troubleshooting and performance analysis, as required by the scenario.
- Storing all raw telemetry in Azure Storage Accounts for long-term retention: While Storage Accounts are cost-effective for raw data storage, querying and analyzing this data requires significant effort. Building a custom application for querying and visualization from raw storage is a complex, time-consuming, and expensive endeavor. It bypasses the powerful querying capabilities of Log Analytics (KQL) and the flexible visualization of Azure Workbooks, which are designed precisely for these tasks within Azure Monitor.
Key characteristics/limitations
- High complexity and cost: Requires significant custom development for data processing, alerting, and visualization.
- Lack of native integration: Does not leverage the inherent strengths and integrations of Azure Monitor services.
- Limited visibility: Resource Health and Service Health are insufficient for deep application and infrastructure insights.
- Inefficient alerting: Event Hubs are a data transport, not an alerting engine, leading to a gap in proactive notifications.
- Vendor lock-in (third-party APM): Introduces dependencies on external vendors.
This answer would be correct if
The organization had a strong existing investment in a specific third-party APM solution they wished to retain, a highly specialized need for raw data storage and custom analytics, and a willingness to invest heavily in custom development for monitoring and alerting. It would also be suitable if the primary goal was simply long-term archival of raw data rather than active, integrated operational monitoring.
Use Azure Monitor Application Insights (classic) for Azure Spring Apps, relying on its default collection. Install the Log Analytics agent on Azure Virtual Machines to collect all performance counters and system logs. Configure Azure Security Center to generate alerts for all security-related events. Visualize all data using Power BI dashboards connected directly to Application Insights and Log Analytics.
Use Azure Monitor Application Insights (classic) for Azure Spring Apps, relying on its default collection. Install the Log Analytics agent on Azure Virtual Machines to collect all performance counters and system logs. Configure Azure Security Center to generate alerts for all security-related events. Visualize all data using Power BI dashboards connected directly to Application Insights and Log Analytics.
What this option is/does
This option suggests using classic Application Insights for application monitoring and the Log Analytics agent for VM monitoring. It proposes Azure Security Center for alerts and Power BI for visualization.
Why it doesn't meet the requirements
This solution has several drawbacks that prevent it from being the best fit:
- Outdated Application Insights: Classic Application Insights does not support newer cost optimization features like Basic Logs or commitment tiers that are available with workspace-based Application Insights. This directly conflicts with the requirement to significantly optimize costs.
- Outdated VM Agent: The Log Analytics agent (also known as the Microsoft Monitoring Agent) is a legacy agent. Microsoft recommends migrating to the Azure Monitor Agent (AMA), which offers superior capabilities for granular data filtering via Data Collection Rules (DCRs), allowing for more efficient cost optimization and flexible configurations. The Log Analytics agent collects data based on workspace-wide settings, making granular filtering for cost reduction challenging.
- Limited Alerting Scope: Azure Security Center (now part of Microsoft Defender for Cloud) is excellent for security-related alerts and posture management. However, it does not cover the full spectrum of operational alerts needed for application performance (e.g., high latency, specific error codes) or general infrastructure health (e.g., high CPU, low memory) that are not security-related. This fails to meet the requirement for comprehensive and proactive critical alerts.
- Suboptimal Visualization: While Power BI can connect to Application Insights and Log Analytics, Azure Workbooks are a native, more integrated, and often more agile solution for creating unified monitoring dashboards within the Azure portal. Workbooks are specifically designed for operational monitoring scenarios, allowing for interactive data exploration and combining data from various Azure Monitor sources more seamlessly without requiring external tools.
Key characteristics/limitations
- Legacy components: Relies on older versions of Application Insights and the Log Analytics agent, missing out on modern features and cost efficiencies.
- Narrow alerting focus: Alerts are limited to security, neglecting operational performance and health.
- External visualization: Power BI is powerful but adds an external dependency and might not be as integrated for operational dashboards as Azure Workbooks.
- Less granular cost control: The older agent and classic AI offer fewer options for fine-grained data filtering and cost optimization.
This answer would be correct if
The organization had existing investments in classic Application Insights and the Log Analytics agent, and the primary focus was on security alerting with external visualization, without a strong emphasis on modern cost optimization techniques or comprehensive operational alerting.
Deploy Azure Monitor Log Analytics workspace for all telemetry, sending all application logs from Azure Spring Apps and all VM logs and performance counters directly to it. Use Activity Log Alerts for all critical incidents and create custom KQL queries for dashboard visualization in Log Analytics. Implement a daily data cap on the Log Analytics workspace to control costs.
Deploy Azure Monitor Log Analytics workspace for all telemetry, sending all application logs from Azure Spring Apps and all VM logs and performance counters directly to it. Use Activity Log Alerts for all critical incidents and create custom KQL queries for dashboard visualization in Log Analytics. Implement a daily data cap on the Log Analytics workspace to control costs.
What this option is/does
This option proposes using a Log Analytics workspace as the primary and almost exclusive destination for all telemetry. It suggests sending all application logs, VM logs, and performance counters to this single workspace. For alerting, it relies on Activity Log Alerts, and for visualization, it uses custom KQL queries within Log Analytics. Cost control is attempted through a daily data cap.
Why it doesn't meet the requirements
This solution falls short in several key areas:
- Limited Application Telemetry: While Log Analytics can store application logs, it lacks the deep, out-of-the-box Application Performance Monitoring (APM) capabilities of Application Insights, especially for detailed dependency tracking, request tracing, and user behavior analytics from Azure Spring Apps. Sending "all application logs" directly to Log Analytics without Application Insights' processing would be less effective for gaining deep insights.
- Inefficient Alerting: Activity Log Alerts are free but are limited to monitoring control plane operations and service health events (e.g., resource creation, deletion, service outages). They cannot monitor application-specific metrics (like response times, error rates) or detailed infrastructure performance counters (like CPU utilization, disk I/O) from within the VMs. This would fail to provide proactive and comprehensive critical alerts for application and infrastructure issues.
- Suboptimal Cost Optimization: A daily data cap is a blunt instrument for cost control. Once the cap is reached, data ingestion stops, potentially leading to a loss of critical monitoring data during peak periods or incidents. This is less sophisticated and effective than granular filtering with DCRs, sampling, or using Basic Logs for specific tables, which optimize what data is ingested or how it's stored rather than just cutting it off.
- Less Integrated Visualization: While custom KQL queries can be used for basic visualization within Log Analytics, Azure Workbooks offer a far more flexible, interactive, and integrated dashboarding experience, capable of combining data from various Azure Monitor sources seamlessly.
Key characteristics/limitations
- Centralized, but undifferentiated storage: All data goes to one place, but without proper pre-processing or categorization, it can be hard to manage.
- Lack of APM depth: Does not provide the rich application insights offered by Application Insights.
- Ineffective alerting: Activity Log Alerts are too narrow in scope for comprehensive application and infrastructure monitoring.
- Risk of data loss: Daily data caps can lead to missed critical telemetry.
This answer would be correct if
The primary requirement was simply to centralize all raw log data for basic querying, and the organization was willing to accept significant limitations in deep application insights, comprehensive alerting, and potentially losing data due to caps. It would also be more suitable if the monitoring needs were exclusively focused on Azure resource management operations rather than application and infrastructure performance.
Implement Azure Monitor Application Insights (workspace-based) for Azure Spring Apps with sampling enabled and unneeded modules disabled. Deploy the Azure Monitor Agent on Azure Virtual Machines with Data Collection Rules (DCRs) to filter unnecessary data and send performance counters to both Azure Monitor Metrics and Log Analytics. Configure Basic Logs for less critical log tables in Log Analytics. Utilize Azure Monitor Metric Alerts for high-frequency, critical infrastructure issues and Log Search Alerts for application-specific anomalies, optimizing their frequency. Consolidate all data for visualization in Azure Workbooks.
Implement Azure Monitor Application Insights (workspace-based) for Azure Spring Apps with sampling enabled and unneeded modules disabled. Deploy the Azure Monitor Agent on Azure Virtual Machines with Data Collection Rules (DCRs) to filter unnecessary data and send performance counters to both Azure Monitor Metrics and Log Analytics. Configure Basic Logs for less critical log tables in Log Analytics. Utilize Azure Monitor Metric Alerts for high-frequency, critical infrastructure issues and Log Search Alerts for application-specific anomalies, optimizing their frequency. Consolidate all data for visualization in Azure Workbooks.
What this solution is
This solution leverages a comprehensive suite of Azure Monitor services, configured with best practices for cost optimization and deep observability. It integrates Application Insights for application-level telemetry, the Azure Monitor Agent (AMA) for infrastructure monitoring, Log Analytics as a central data store, and Azure Monitor's alerting and visualization tools.
How it meets the requirements
- Detailed application telemetry from Azure Spring Apps: Workspace-based Application Insights is the native and most effective solution for monitoring Azure Spring Apps. It automatically collects application performance data, dependencies, and requests. Enabling custom events and metrics through the Application Insights SDK allows for deep, application-specific insights into user behavior and business logic. Being workspace-based also allows for leveraging Log Analytics cost-saving features.
- Monitor infrastructure performance counters and system logs from Azure Virtual Machines efficiently: The Azure Monitor Agent (AMA) is the recommended agent for VMs, offering granular control over data collection. Data Collection Rules (DCRs) allow precise filtering of logs and performance counters, sending only necessary data to Log Analytics, thus reducing ingestion volume. Sending performance counters to Azure Monitor Metrics provides near real-time, high-frequency data for immediate infrastructure health checks, while also sending to Log Analytics for longer-term analysis and correlation with other logs.
- Cost-effective data retention for frequently collected logs that are less frequently queried: Configuring Basic Logs for specific tables in Log Analytics that contain high-volume but infrequently queried data (e.g., verbose diagnostic logs) significantly reduces ingestion costs. This is a key feature available with workspace-based Application Insights and DCRs. Additionally, sampling in Application Insights reduces the volume of telemetry sent from applications without distorting metrics, further optimizing costs. Disabling unneeded modules in Application Insights also reduces unnecessary data collection.
- Critical alerts are configured to be both proactive and cost-efficient: Azure Monitor Metric Alerts are highly efficient and cost-effective for high-frequency, numerical data (like CPU utilization or request rates), providing immediate notification for critical infrastructure issues. Log Search Alerts are suitable for detecting application-specific anomalies or patterns in collected logs, with their frequency optimized to balance responsiveness and cost.
- Unified dashboard visualization for all collected telemetry: Azure Workbooks provide a flexible and powerful canvas to create interactive, integrated dashboards that combine data from Application Insights, Log Analytics, and Azure Monitor Metrics, offering a single pane of glass for comprehensive visibility across the entire application and infrastructure stack.
Key advantages/capabilities
- Comprehensive Coverage: Monitors both application (PaaS) and infrastructure (IaaS) components.
- Cost Optimization: Utilizes sampling, DCR filtering, Basic Logs, and efficient alert types to manage ingestion and alerting costs.
- Granular Control: AMA and DCRs offer fine-grained control over what data is collected from VMs.
- Deep Application Insights: Application Insights provides rich APM capabilities, including custom telemetry.
- Centralized Data: Log Analytics workspace acts as a unified repository for all logs and metrics.
- Flexible Visualization: Azure Workbooks enable highly customizable and integrated dashboards.
- Proactive Alerting: Combines metric and log alerts for efficient and timely notifications.
Why this is optimal
This design is optimal because it directly addresses all stated requirements, particularly the need for cost optimization without sacrificing comprehensive visibility. It leverages the strengths of each Azure Monitor component: Application Insights for deep application telemetry, AMA for efficient VM monitoring, Log Analytics for centralized storage and advanced querying, and Azure Monitor's native alerting and visualization tools. The specific recommendations like sampling, DCR filtering, and Basic Logs are direct responses to the high ingestion cost problem, making it a well-rounded and efficient integrated monitoring architecture.
Try 10 interactive questions like this one, free
Answer them yourself and get your score the moment you finish. No account needed.
Full access includes
- 124 practice questions across all 4 exam domains
- Every answer explained: why it is right, and why the others are not
- Scout, an AI tutor that answers your follow-up questions
- Readiness tracking by domain, so you know when to book the exam
Related questions
Keep the momentum going with these hand-picked practice scenarios
ExpertAZ-305 Designing Microsoft Azure Infrastructure Solutions Exam
Azure Container Apps vs AKS
Recommend a container-based solution
IntermediateAZ-400 Designing and Implementing Microsoft DevOps Solutions Exam
GitHub Branch Protection Rules and SAML SSO
Design and implement permissions and roles in GitHub
IntermediateAZ-400 Designing and Implementing Microsoft DevOps Solutions Exam
Managed Identity for Azure DevOps Service Connections
Implement and manage Azure DevOps service connections and personalaccess tokens
Last updated