All in on APM

It’s been just over six months since Splunk disrupted the Application Performance Monitoring (APM) market with the launched SignalFx Microservices APM, combining the technologies of SignalFx and Omnition. We have pushed ourselves harder and continued to invest in creating more value for our customers by making it easier for them to ingest ALL data and providing ever more powerful analytics on top of that data. All of this has been done with the help of the entire Splunk team, which is one of the reasons we’ve decided to change the product name to Splunk APM. Excited to learn about our innovations? Let’s dive right in!

ALL Your Data In with No Lock-In

Our goal is to help our customers monitor and modernize their applications regardless of where they are in their cloud journey. Traditional applications tend to use one set of programming languages, and modern applications use another. Additionally, when creating modern applications developers often write different services using different languages. That is why over the summer we have invested in creating auto-instrumentation client libraries in more languages, and now support all major web development languages: Java, Python, .Net (Core and Framework), Node.js, GoLang, Ruby and PHP. 

We are big believers that your data should be under your control, which is why we, as Omnition,  co-founded OpenTelemetry (the second most popular CNCF project after Kubernetes) and are the biggest contributors to this project. We contribute all our instrumentation libraries to OpenTelemetry. In the past few months we created the Splunk distribution of OpenTelemetry Java and Python auto-instrumentation, and helped launch the .NET auto-instrumentation project. OpenTelemetry-based instrumentation is lightweight, easy to use, gives you full control over your data and maximum flexibility when creating new apps, and eliminates vendor lock-in.

Full-Fidelity Tracing, End-to-End

Splunk APM leverages OpenTelemetry-based instrumentation to ingest ALL trace data. Splunk APM is the only solution in the market that collects, analyzes and stores 100% of the trace data with absolutely no sampling. Full-fidelity tracing provides our customers with unmatched levels of visibility into their applications, enables them to find and understand ANY issue (even those that they did not think about in advance), and is the basis for the most powerful analytics in the APM market. 

Full-fidelity trace ingestion is a powerful capability. By expanding it to the front-end with the beta release of Splunk Real User Monitoring (RUM) SREs, developers, and Ops teams would achieve something that has never been offered before — the ability to track every single user action all the way from the front-end browser traces to the corresponding backend traces with absolutely no sampling! This groundbreaking capability gives you new levels of visibility into user experience, and enables you to detect and resolve user-facing issues or bottlenecks faster than ever before! 

Faster Troubleshooting & Business Insights

Other APM tools have many limitations, including on the cardinality of the data — the ability to slice and dice the data based on different attributes (also known as tags). At best, they can only handle tens of thousands of dimensions, lumping together thousands of users with no ability to discern between them. As the category-defining product, Splunk APM is not bound by these limitations and offers Infinite Cardinality, with millions of dimensions per minute, enabling unprecedented visibility into user behavior. 

Full-fidelity tracing, and Infinite Cardinality power the analytics behind the newly released Tag Spotlight functionality. Before using Splunk APM, many of our customers spent hours looking at traces and logs to try and figure out the root cause of latency or errors. Tag Spotlight reduces those hours to minutes, and provides users with a powerful way to quickly discover granular trends contributing to latency or errors on a service. You now have a single pane of glass to view infrastructure, application or business-specific tags in APM to quickly correlate peaks in latency or errors with specific tag values. For instance, you can easily identify that root errors could all be coming from the same K8s node or affect just a subset of tenants (see image below). 

^{Tag Spotlight provides a one-stop shop to analyze all tags}

Adding to the exciting news, is the beta availability of Business Workflows (planned to be Generally Available later this year), enabling our customers to visualize and monitor the health of their business, see business trends over time, and alert if any revenue-impacting issue is detected. Business Workflows data is automatically extracted from traces and  its seamless linking between charts, dashboards, and alerts then make it easier to identify the culprit service and shorten troubleshooting time. Our unique streaming analytics, full-fidelity tracing, and infinite cardinality, provide you with immediate feedback on whether the resolution has indeed solved the problem. The amount of time it takes users to authenticate or to go through the checkout process are just a couple of examples of business KPIs that Business Workflows can track and troubleshoot when there’s an issue. 

^{Tracking the Frontend Check KPI with Business Workflows}

Enterprise-Grade Control

Usage management has always been a part of Splunk Infrastructure Monitoring (formerly known as SignalFx Infrastructure Monitoring), and later this year we’ll be expanding it to APM as well. We strive to give our customers full control over their data, access, usage, and costs. As part of that goal, we’ll provide usage graphs from the beginning of each billing cycle to help them better manage and understand their costs and avoid any overage surprises. We’re also helping our customers better plan for the future and understand their usage trends through a monthly billing history graph. 

^{Billing History}

Forging a Path

All these innovations provide you with deeper understanding of your users and reduce troubleshooting time so that you can invest more time forging ahead with new innovations.

It is very exciting to see so many innovations come to fruition in such a short amount of time. You can be sure that there are many more exciting innovations to come.

Interested in learning more? Check out the links below:

• Splunk .conf20 sessions:
      1. Observability Super Session: Observability in the Data Age
      2. Get started with Microservices APM in minutes
      3. A Sneak Peek Into End-to-End Visibility with Splunk
      4. 5 Reasons Why OpenTelemetry is the Future of Observability
      5. Microservices, a fairy tale with a “cautiously optimistic” ending
  
• Splunk APM Website
  
• Sign up for our beta program: Observability Suite, Splunk RUM
  
• Splunk’s contribution to OpenTelemetry