Analyzing the Morphing Monitoring Market: APM, Logs, Pricing and Analytics
On this 131st episode of the The New Stack Analysts podcast, Nancy Gohring of 451 Research joined us to talk about the differences between application monitoring and IT operations analytics. Just defining the market is complicated, so Gohring provided some explanation. In many ways, she sees many similarities between the application performance monitoring (APM) and what is going on in server and infrastructure monitoring. However, she sees more differentiation when it comes to providers dealing with log management. Where there is the greatest degree of overlap, and where vendors are focusing hard on innovation, is the area of advanced analytics.
#131: Analyzing the Morphing Monitoring Market: APM, Logs, Pricing and Analytics
In her recent “Application and Infrastructure Performance Market Map 2017” report, Gohring had these takeaways regarding advanced analytics and monitoring:
- Key Challenges: Vendors across market segments, including APM, server monitoring, and log management, are adding analytics capabilities that in some cases address a common set of problems, resulting in new competitive battlefields.
- Innovations: Anomaly detection, automatic baselining, predictive analytics, sophisticated querying capabilities and innovative backend data management techniques are driving this sector.
Although the full report is only available to 451 Research clients, this diagram shows some of the players in the APM analytics space.
Other topics covered included:
- Deployment patterns: For example, how many monitoring tools are being used simultaneously? Is a full ELK stack being used?
- Pricing: Gohring believes “2017 will be a year of pricing experimentation…We’re hearing vendors begin to study different pricing models and expect to see some new plans introduced in the coming year.” The vendor Instrumental describes four approaches:
- Based on Host Count (e.g., Datadog, New Relic)
- Pricing Based on Volume (e.g., SignalFx)
- Pricing Based on Metric Count (e.g., Instrumental, Librato)
- Pricing Based on Resolution (e.g., Librato)
- IT Operations Analytics (ITOA): Gohring believes it is a vendor-driven term. Based on this vendor-created ITOA landscape, we believe she is right. To her, IT Operations Analytics combines application, infrastructure, business data and other sources to provide insights. She talked about how the market is reshaping itself as ITOA 2.0. Since the podcast, she has elaborated on this in the report “Big Data, Machine Learning Shape Performance-monitoring Developments.”
- Amazon Web Services has become a big player in the space. The CloudWatch and X-Ray offerings were mentioned. Gohring said that with AWS you still can’t get 100 percent visibility into application performance. Another challenge with relying on AWS is when you are running an application in a true hybrid cloud environment.
- Two efforts by established tech companies are:
- Cisco Tetration Analytics for data centers.
- CA Analytics Project Jarvis. This is not Mark Zuckerberg’s Facebook project of the same name. Instead is an Analytics Engine that relies on a lot of open source components.
CA’s Analytics Reference Architecture Using Project Jarvis
- The Classes of Container Monitoring by Brian Brazil, an active member of the Prometheus community. See below for an example monitoring architecture:
Feature image via Pixabay.
