We are excited to announce the addition of instance-level datastore visibility to New Relic APM. We’ve made performance monitoring and troubleshooting between applications and databases quicker and easier by providing this information in New Relic APM Service Maps, Transaction Traces, and Slow Query Traces.
Why instance-level visibility into datastores is so important
Software complexity is exploding. With the accelerating adoption of microservices, cloud computing, and containerization, applications are becoming smaller and changing more rapidly than ever. Application architectures are becoming more distributed and software is more complicated than ever. Today’s highly complex applications rely on multiple programming languages, third-party APIs, and increasingly even multiple datastores with multiple instances to store and process data.
Modern applications process a tremendous amount of data, making them more and more reliant on highly sophisticated databases/datastores. Any performance degradation in the datastore instances can have a huge impact on the performance of these applications. Today’s operations, development, and DevOps teams need tools that provide complete visibility into the application architecture, including the performance of each datastore.
In addition to the visibility needed to help figure out the cause of the problem when something goes wrong, the tool should allow users to quickly navigate through all the layers of the application to the relevant details. In case of an issue, it should gather enough actionable data to assess the issue’s impact and begin the root-cause analysis.
What’s new in New Relic APM
With this release, New Relic customers now have complete visibility into their application environments (including datastore performance) along with the ability to determine the impact of a performance issue with a specific datastore instance and to more quickly identify database bottlenecks. This new capability is available in two places:
- Within New Relic Service Maps, customers will now be able to gain insight into the health and performance of their application architecture all the way down to the specific datastore instances to which the applications are connecting. These Service Maps are a customizable, real-time visualization, which means that users now can filter to individual datastore instances and quickly troubleshoot the performance bottlenecks.
- With Transaction traces and slow query traces, New Relic APM now includes datastore instance-level details to ensure that teams have the actionable, code-level insights they need to understand the cause of an issue or to determine where best to focus their performance-improvement efforts.
Going beyond application visibility
In addition to application health, metrics, and dependency information, New Relic Service Maps now provide deeper visibility into the database tier of your application architecture. Customers can see metrics for datastore calls all the way down to the individual datastore instance. Response time and throughput can be viewed on a timeseries chart that can be filtered and faceted to focus on specific applications, instances, or a combination of factors. This is particularly useful when attempting to identify anomalous behavior at the application or datastore instance level.
In addition to the additional metrics and charting, we’ve added a health indicator that gives you an at-a-glance indication of the health of the connections between your applications and datastore instances. The new health indication is computed using a simple baseline technique that compares the performance over the past 15 minutes with the average over the past week.
And it doesn’t end there! The same language agent-reported metrics that drive the new visualizations on service maps can also be viewed on custom dashboards or in New Relic Insights.
You can even go a step further and create alerts on these new metrics to notify you of any potential performance issue before it starts impacting customer experience.
Datastore instance-level details are also available within transaction traces and slow query traces. This information includes details on the specific database instance (host + port) involved and the database name against which the query was run.
These are powerful new capabilities that can help reduce MTTR for issues such as “noisy neighbor” problems or database configuration issues (such as a missing index) by allowing customers to easily identify outliers and obtain slow-query information down to the individual instance level.
Try it out and let us know!
Datastore instance information in transaction traces and slow query traces is available to New Relic customers with at least an APM Essentials subscription and who meet the minimum agent versions requirements. Datastore instance level visibility on service maps is available to New Relic customers with at least an APM Pro subscription.
We would love for you to try out the new capabilities and let us know what you think. Please provide your feedback directly in the New Relic Community Forum.
Neha Duggal contributed to this post.