Monitoring ^Enterprise

Monitoring in Memgraph Lab provides a comprehensive, real-time view of system performance and database activity. It’s designed to help developers, administrators, and data engineers optimize workloads, identify issues quickly, and better understand how their Memgraph instance is behaving.

The database is being monitored during an active Lab session and the data is refreshed automatically every 5 seconds. You can customize this interval using the Change time interval option in the top-right corner of the dashboard which will lead you to the Monitoring settings in the Settings section.

Metrics summary

The summary panel on the right-hand side of the Monitoring dashboard displays key metrics at a glance:

• Available memory – The current amount of memory available to the Memgraph instance.
• Peak memory – The highest amount of RAM used since the instance started.
• Query status – A pie chart showing the overall percentage of successful and failed queries.
• Query type – A pie chart summarizing the proportion of read, write, and read-write queries executed.

Metrics charts

The main area of the Monitoring dashboard consists of a set of real-time charts that track system performance and database activity over time. These charts provide deeper insights into how Memgraph is running, helping you identify trends, spikes, and potential issues as they happen.

The following metrics are visualized:

• Memory usage – Tracks the amount of memory used by Memgraph over time, helping identify memory pressure or unusual behavior.
• Data metrics, nodes and edges – Displays the number of nodes and edges currently stored in the database, giving insight into graph growth.
• Query responses – Shows the number of successful and failed queries, allowing you to monitor execution stability.
• Query type – Breaks down the volume of read, write, and read-write queries to better understand query workload.
• Disk usage – Tracks how much disk space is used by the database engine.
• Transaction types – Displays counts of committed vs rolled-back transactions to help detect abnormal patterns or failures.
• Active sessions – Shows the number of concurrent Bolt sessions (including Memgraph Lab), helping track active client connections.

You can click the eye (👁️) icon next to a chart legend item to hide specific lines you’re not interested in. For example, to focus on failed query responses, simply hide the successful ones.