Removing your Oracle Database Performance Blind Spot with New Relic

by Cameron Jones on June 30, 2017

Oracle Database has been a mainstay within the enterprise datacenter for the last couple of decades. While the database market has exploded over the past five to 10 years with new database types (e.g., NoSQL, Time Series) and database engines (e.g., MongoDB, Couchbase), a large percentage of business-critical data is still hosted, and will continue to be hosted in the future, on Oracle Database. The ability to monitor, visualize, and alert on the health, performance and availability using New Relic is a necessity in many environments.

BindPlane for New Relic with our Oracle Database integration, was designed from the ground up to provide users with deep visibility into the performance of the Oracle databases backing their applications.

More often than not, Oracle database has been a black box to app owners and administrators. There has been little visibility when Oracle performs poorly, which becomes major concern when the application goes to production. In this blog post, I will walk through a number of key use cases that demonstrate how BindPlane can be used with New Relic to ensure Oracle Database is no longer the blind spot within your application infrastructure.

Understanding Oracle Database Health with Contextual Dashboards
For many users, gaining a quick glimpse of Oracle Database serves as a top priority. Monitoring key performance indicators like Oracle Database wait time, sessions and query performance provides enough depth to quickly identify when performance changes. BindPlane includes five out-of-the-box dashboards that allow you to zero in on these aspects of Oracle Database performance.

However, when things go wrong, additional data serves as paramount. To do this, take a look  at New Relic Insights. Within Insights, you gain access to more than 450 different Oracle Database performance metrics, including sessions, waits, tablespaces and read/writes. With these metrics you can create custom dashboards and data apps that fit your specific needs.

In the image below, we have created a data app that provides key metrics to meet the needs of our Oracle Database. Most of our performance issues derived from CPU usage and how many people were connected to the database at any one time. So, on the main screen, you can view key performance indicators like session usage, CPU usage and current active sessions. From there, view key configuration information to know which database we were viewing.

Dig Deeper with New Relic Insights
While high-level health and availability dashboards are nice, the ability to see and understand the actual queries that are executing serves as a common use case for developers. By leveraging New Relic Insights, you can quickly identify the worst performing queries within the Oracle Database. Let’s take a look at another dashboard on the data app we created earlier.

 

This dashboard — the Query Breakdown dashboard — provides an overview of the worst performing Oracle queries, sorted by the by number of executions. This dashboard also indicates essential details about your queries, including which ones that:

  • Execute the most often
  • Have the longest average execution time
  • Use the most CPU
  • Cause the most application wait time

As a result, you can immediately identify if there is a query in the application that needs to be rewritten or altered before sending it off to production. Or, if it is already in production, that it can be fixed immediately!

In this use case, viewing the queries that are executed the most often worked best. However, BindPlane for New Relic allows you to rank your queries by reads or writes. In fact, it allows you to rank your queries by a number of metrics, including:

  • Elapsed time
  • CPU time
  • User IO wait time
  • Buffer gets
  • Executions
  • Parse calls
  • Sharable memory
  • Child count
  • Average time
  • Physical read/write bytes
  • Physical read/write requests

Be proactive!
Have you ever experienced a sudden performance drop between stage and production? Or experienced immediate downtime right after somebody pushed to master? Never have this happen again. BindPlane supports alerting in New Relic that allows you to see the problem before it’s reported to you — meaning the database will never be the cause of a midnight call ever again.

Blue Medora integrates with New Relic Alerts, providing instant notifications on any Oracle performance metric collected by BindPlane, which you can easily integrate with your email to receive alerts in real time. This enables custom thresholds to be alerted on for memory, number of full index scans, application wait time, reads, writes or any other metric that is critical to ensure health of the application.

 

The health of your app is paramount and that means understanding the performance, health, and availability of the database your app depends on. The database is a critical component of understanding the the total health of that app.

BindPlane for New Relic provides the visibility you need to identify sudden performance degradation, ferret out bad query performance, and be notified of data outages the instant they occur. In addition, it works with any of your 11g, 12c, RAC, One Node or Oracle Cloud databases, and is fully supported by Blue Medora.

Get started

Try BindPlane for free. No credit card required.

Sign up
True Visibility
BindPlane for VMware vRealize Operations

True Visibility allows cloud management teams to use VMware vRealize’s powerful machine learning and capacity planning engine across their entire hybrid cloud environment.

Azure Monitor...everything
BindPlane for Microsoft Azure Monitor

Make Azure Monitor your first-pane-of-glass across your entire multi-cloud, multi-database or hybrid platform environment.

Thank you for contacting us. Your information was received. We'll be in touch shortly.