Monitoring Performance Insights Management Guide on the Elven Platform

The Monitoring Performance feature of the Elven Platform provides a comprehensive view of application performance, using essential metrics to assess system efficiency and stability. This analysis enables quick identification of areas for improvement, helps prevent failures, and optimize resources, ensuring the service operates in a consistent and reliable manner.

Tracking these metrics is a powerful tool to anticipate issues and ensure a high-quality experience for users, with minimal impact and maximum operational efficiency.

Accessing the Insights Center in the Monitoring Performance section

• Navigate to the main menu and click on Insights.

• In the submenu, select the item Monitoring Performance.

Understanding the Metrics

The study of performance metrics in Monitoring Performance is essential to ensure that the system operates efficiently and without major interruptions. Through the analysis of metrics such as MTBF, Uptime, Downtime, Outages, Max Hour Latency, Min Hour Latency, and Avg Hour Latency, it is possible to gain valuable insights into the application's health. These metrics help identify areas for improvement, anticipate potential issues, and optimize performance, ensuring that users have stable and responsive access to the service. By closely monitoring these metrics, the platform can evolve proactively, minimizing impacts and maximizing resource efficiency.

MTBF (Mean Time Between Failures)

MTBF represents the mean time between failures of a system or component, providing a measure of the reliability and stability of the service. This metric is important for predicting the frequency of failures and, therefore, planning preventive maintenance and improvement strategies. A high MTBF indicates that the system or component is operating in a stable manner, with failures occurring less frequently.

Example: During the 30-day analysis period, the MTBF was 150 hours, which means that, on average, failures occurred every 150 hours of continuous operation. This value indicates that the system demonstrated reliable performance, with occasional failures that did not significantly compromise service continuity. Identifying the most frequent failure points and optimizing the most vulnerable components can contribute to increasing the MTBF, resulting in a more stable platform with greater availability for users.

Downtime

Downtime refers to the period when the service or platform is offline or inaccessible to users. This period can be caused by a variety of factors, such as system failures, scheduled maintenance, or unexpected technical issues. Monitoring downtime is essential to ensure service continuity and minimize the impact on operations.

Example:

On a given day, the platform experienced 2 hours of downtime due to a database server failure. During this period, users were unable to access certain application features. Quickly identifying the causes of the downtime and implementing corrective actions is crucial to reduce the duration of the issue and improve the platform’s reliability in the future.

Outages

Outages represent significant service failure events, such as system crashes or critical errors that prevent the platform from functioning properly. Unlike downtime, which may occur due to scheduled inactivity or minor issues, outages are unexpected and impactful events that require rapid response to restore normal service operation.

Example:

During November 2024, the platform experienced a 30-minute outage due to a configuration error on one of the main servers. This failure temporarily affected all users, causing service unavailability. After resolving the issue, the team implemented preventive measures to avoid similar failures in the future, such as configuration reviews and monitoring process improvements.

Uptime

Uptime measures the period during which the service was available and operating normally, without interruptions or failures. This metric is essential for evaluating the reliability and stability of the platform, ensuring that users can access the service continuously and without issues. A high uptime rate indicates that the platform is functioning efficiently and without significant interruptions.

Example:

On November 25 and 26, 2024, the platform maintained 100% uptime, demonstrating a period of uninterrupted operation during those two days. This level of availability reflects a stable operation and the team’s ability to keep the service failure-free, providing a reliable experience for users.

Max Hour Latency

Max Hour Latency indicates the highest response time recorded within one hour during the analyzed period. This metric is essential for identifying performance spikes and potential bottlenecks that may impact the user experience. High latency may suggest system issues, such as overload or component failures, which need to be addressed to ensure service efficiency and smooth operation.

Example:

During the monitoring period, the Max Hour Latency was 62ms, representing the highest response time recorded in one hour. Although this value falls within an acceptable range, it may indicate the need to pay attention to possible demand spikes or request processing optimization to ensure service consistency.

Min Hour Latency

Min Hour Latency represents the lowest response time recorded within an hour during the monitoring period. This metric is important for understanding the system's optimal performance under normal conditions and identifying periods when the service operated with greater efficiency. A low minimum latency value indicates that the system was able to process requests quickly, providing a more agile experience for users.

Example: During the analyzed period, the Min Hour Latency was 52ms, reflecting the lowest response time recorded in one hour. This value suggests that, at certain moments, the system was optimized, delivering agile and efficient performance, which contributes to a smoother experience for users.

Avg Hour Latency

Avg Hour Latency is the average response time recorded during each hour of the analyzed period. This metric provides an overall view of the system's performance over time, allowing the identification of efficiency trends or possible service degradations. A lower average latency indicates a consistent and efficient response time, which is essential to ensure a good user experience.

Example: During the monitoring period, the Avg Hour Latency was 62ms, indicating that, on average, the system was able to maintain a reasonably fast and stable response time throughout the analyzed hours. This value suggests that the system had consistent performance, without major fluctuations, which contributes to a user experience free from noticeable delays.

Uptime per Day

The Uptime per Day metric provides an essential view of the service’s reliability, ensuring that users have continuous access to the application and strengthening trust in the platform.

Example: On November 25 and 26, 2024, the system achieved 100% uptime, demonstrating stable and uninterrupted operation throughout the analyzed period. This consistency reinforces the efficiency of monitoring and the robustness of the infrastructure.

Outages per Day

The Outages per Day metric is essential for identifying failures and service interruptions, allowing for a detailed analysis of their occurrence and the adoption of corrective measures to prevent future impacts.

Example: On November 25 and 26, 2024, no interruptions were recorded in the system, reinforcing the application's reliability and the effectiveness of failure prevention mechanisms. This stability contributes to a continuous and satisfactory experience for users.

Latency per Day

A métrica Latency per Day é essencial para monitorar o tempo de resposta do sistema e garantir a Efficiency in service delivery allows for the identification of performance variations, optimizing the user experience.

Example: On November 25 and 26, 2024, the system maintained an average latency of 98ms, with a maximum peak of 62ms and a minimum of 52ms. These values indicate consistent and efficient performance, ensuring quick responses and smooth navigation for users.

Latency per Hour

The Latency per Hour metric provides a detailed analysis of system performance throughout the day, allowing the identification of specific variations and periods of higher or lower efficiency. This granular view is essential for resource optimization and ensuring a high-quality experience for users.

Example:

On November 26, 2024, the hourly analysis showed an average latency of 98ms, with values ranging from a peak of 62ms to a minimum of 52ms. This hourly consistency reflects a well-balanced system, capable of efficiently handling demands throughout the day.

TOP Downtime Resources

The TOP Downtime Resources feature allows the identification of specific components or services that experienced the most downtime. With this information, it becomes possible to prioritize corrective actions and optimize infrastructure to reduce interruptions, ensuring a more stable experience for users.

Example:

In a recent analysis, the Authentication API resource was identified as the main component with accumulated downtime, totaling 2 hours and 30 minutes during the week. This data highlights the need to review critical dependencies and implement mitigation measures, such as redundancy or architecture improvements.

Top AVG Latencies per Resources

The Top AVG Latencies per Resources functionality provides a detailed analysis of the system components that show the highest average response times. With this view, it is possible to identify performance bottlenecks and prioritize optimizations to improve the user experience.

Example: During the analysis, the Front-end resource showed an average latency of 62ms, being the component with the highest response time in the evaluated period. Although this value is within acceptable limits, it highlights the importance of reviewing critical routes, optimizing queries, and implementing caching techniques to further reduce latency and improve the system’s overall performance.

Status

The Status feature allows monitoring the state of each application component over time, providing information about the availability and operating time of specific resources. This view is essential to ensure that the application’s critical resources are functioning without interruptions, optimizing the user experience and the service reliability.

Example:

The Front resource maintained continuous uptime between 2024-11-25, 00:00:00 and 2024-11-26, 13:15:25, with a total duration of 1 day, 13 hours and 15 minutes. This high operating time without failures indicates considerable resource stability, allowing users to have uninterrupted access to the application during this period. Continued monitoring of this status will help ensure the service remains available and efficient.

Glossary of Technical Terms

Insights Center: Central module of the Elven Platform that provides in-depth analysis of operational and business data, supporting strategic decision-making and performance improvement.

Monitoring Performance: Feature that offers a comprehensive view of system performance through key metrics such as uptime, downtime, latency, and outages, aiming to optimize application stability and efficiency.

MTBF (Mean Time Between Failures): Metric that indicates the average time between failures of a system or component, reflecting its reliability and operational stability. A high MTBF value suggests the system can operate for long periods without interruptions, contributing to greater availability and service efficiency.

Uptime: Metric that indicates the period during which the service is available and operating normally, without failures or interruptions. A high uptime rate reflects system stability and reliability.

Downtime: Period when the service is offline or inaccessible to users. It may be caused by system failures, scheduled maintenance, or unexpected issues. Monitoring downtime is essential to minimize operational impact.

Outages: Critical events that cause significant service failure, such as system crashes or failures in essential components. Outages require quick actions to restore normal operation.

Latency: The response time of the system to a request. It can be measured in different ways, including maximum, minimum, and average latency, and is fundamental for evaluating system speed and efficiency.

Max Hour Latency: The highest latency recorded in one hour during the analyzed period. This metric helps identify latency spikes that may indicate performance issues.

Min Hour Latency: The lowest latency recorded in one hour during the monitored period. It reflects the ideal system performance during periods of high efficiency.

Avg Hour Latency: The average latency recorded during each hour of the analyzed period. This metric helps identify overall performance trends and system efficiency over time.

Uptime per Day: Metric that evaluates the daily reliability of the service, showing the percentage of time the system remained available on a given day.

Outages per Day: Metric that quantifies the number of service interruptions during a day, helping monitor the frequency of failures and the effectiveness of preventive actions.

Latency per Day: Metric that provides a detailed view of daily latency, allowing the identification of performance variations throughout the day.

Latency per Hour: Metric that offers a granular analysis of system latency, helping identify patterns and periods of higher or lower efficiency throughout the day.

TOP Downtime Resources: Feature that identifies the resources or components with the highest downtime, allowing prioritization of corrective actions to optimize infrastructure.

Top AVG Latencies per Resources: Feature that provides an analysis of system components with the highest average latencies, helping identify bottlenecks and areas that need optimization to improve the user experience.

Status: Feature that monitors the state of each application component, providing information about the availability and operating time of critical resources, ensuring service continuity.