Industry experts offer thoughtful, insightful, and often controversial predictions on how APM, AIOps, Observability, OpenTelemetry and related technologies will evolve and impact business in 2023. Part 4 covers monitoring, site reliability engineering and ITSM.

Start with: 2023 Application Performance Management Predictions - Part 1

Start with: 2023 Application Performance Management Predictions - Part 2

Start with: 2023 Application Performance Management Predictions - Part 3

MONITORING ALIGNS WITH BUSINESS

Organizations will refocus their IT Infrastructure Monitoring efforts on helping the organization achieve its business objectives — Everything within an organization should be there to help achieve defined business objectives. It does not matter if you are the custodian sweeping the floor, the developer writing lines of code, or the CEO making decisions of what strategies to pursue, each person is there to help achieve specific objectives. Dirty floors may attract rodents and insects that distract developers from writing good code, which impacts the CEO's ability to traverse a specific path. ITIM tools should be no different. The ultimate objective of an ITIM tool is to ensure the performance of the infrastructure, and when that performance lags, help identify and resolve the issues. These approaches will move from a focus on metrics (i.e. 99.999% uptime) to a focus on business objectives (we need 100% uptime from 9am to 5pm and 50% uptime from 5pm to 9am). By transitioning the focus away from fixed values (which are likely unrelated to the businesses objectives) to a focus on helping achieve those same objectives the organization is more likely to succeed.
Josh Chessman
VP, Strategy & Innovation, Netreo

THE NEED FOR FULL VISIBILITY

Cyber incidents were at an all-time high in 2022, topping the Allianz Risk Barometer as the number-one cause of enterprise risk (44% of responses), beating business interruptions and natural disasters. Teams are under pressure to put minimum cyber hygiene standards into practice, and failure to do so can lead to loss of data, additional breaches and poor application performance. In 2023, organizations will recognize the need to have full visibility into all aspects of their IT environment, and invest in application performance monitoring to proactively identify red flags before they become issues. This will save businesses time and money, enable security teams to remediate vulnerabilities and decrease the risk of disruption to critical business operations.
Adrian Moir
Sr. Product Management Consultant & Technology Strategist, Quest Software

MONITORING FOR DEVELOPERS

"You build it, you own it!" has never been more true than today. Engineers are responsible for both writing code and owning their services on production. In 2023 I think we will see a breakthrough in dead simple monitoring made for developers. Tools that do not require engineers to have a complete understanding of the whole stack of their complex and interconnected applications. These tools will give developers the confidence that their system works from an end-user perspective.
Hannes Lenke
CEO and Co-Founder, Checkly

MONITORING AS CODE

I predict that 2023 will be the year of monitoring as code, where DevOps teams will ship their monitors alongside their application code. Developers can easily configure and customize their own monitoring without requiring operations teams to do so. To enable this, the repository will become the single source of truth for all things software development. This Developer-centric approach to monitoring will lead to more stable and performant applications.
Hannes Lenke
CEO and Co-Founder, Checkly

VISIBILITY INTO BUGS IS IMPERATIVE

Whether bugs are found in pre- or post-production, it's critical for teams to be able to remediate bugs quickly and with minimal disruption to efficiency, their business, or their customers. I predict that greater visibility into all the context that led to that bug, as well as the visibility into potential business impact, will be viewed as increasingly imperative, and therefore increasingly demanded from vendors. "This failed," is no longer helpful-enough info, not when advancements in intelligent automation can provide development teams with location, root cause, level of risk, and other predictive information that can prevent similar bugs in future releases.
Noel Wurst
Quality Evangelist, SmartBear

EBPF CHANGES THE WAY SYSTEMS ARE MONITORED

eBPF is poised to drastically change the way systems are monitored. By allowing small programs to run securely directly in the Linux kernel across multiple kernel versions both vendors and organizations will benefit. Vendors benefit by simplifying the process of creating and aggregating different types of data which will allow significantly more data to be collected faster and more easily. Organizations will benefit due to the simplification of deploying data collectors, faster collection times, drastically more data types available, and an overall improvement in process. All these benefits combine to allow organizations to improve the level of visibility across applications, networks, infrastructure, and more.
Josh Chessman
VP, Strategy & Innovation, Netreo

INCREASING THE SIGNAL TO NOISE RATIO

I think 2023 (and beyond) will be about increasing signal to noise ratio. Or actionable data. Or reducing false positives. They are all addressing the same issue: we have a mountain of data points across performance, reliability and security but what is important and urgent? Across the industry I see many initiatives starting to address this core issue, but we have a long way to go. Data overload, alerting fatigue, endless logs. In the past, we didn't have enough data. Now we can record and scan anything and everything economically. So we are presented with the next problem: what do my engineers act on today? What about next week? How do we structure that? What is actually impacting users and the business, what is “just a blip?” In summary, I think in 2023 we'll see developments that begin to help engineers increase the signal to noise ratio by removing needless distractions and helping them spend resources on what matters most.
Tim Nolet
CTO/CPO and Co-Founder, Checkly

FACING THE BIG DATA CHALLENGE

Over the last decade the industry has produced a large number of tools that analyze code, traffic, usage patterns and more for integrity, reliability, and security. The proliferation of tools is producing an ever-increasing flow through an already unmanageable set of firehoses. The new use of AI to look at code, development and test practices, coverage data, monitoring data, traffic, etc., should finally start bearing fruit with AI/ML augmented capabilities that will help prioritize what is important and what can safely be ignored to save developers, testers, and ops from drowning in a sea of data.
Arthur Hicken
Evangelist, Parasoft

Organizations are processing more data and workloads than ever before, and in 2023, IT infrastructure monitoring will become more complex as organizations scramble to manage the influx of data from on-prem, the cloud, and edge and IoT devices. Thanks to regulation and additional oversight from non-technical leadership, real-time monitoring of IT infrastructure will be seen as a must-have, not a nice-to-have. To ensure this, in 2023, IT teams must prioritize implementing data observability and IT diagnostic capabilities to reduce blind spots and eliminate silos.
Adrian Moir
Sr. Product Management Consultant & Technology Strategist, Quest Software

SRE SUPPLY MEETS DEMAND

Will there be a shortage of SREs? I think the number of SREs will grow to meet the demand. Increasing the supply will require education and training. But I also think the layoffs of the past months in the tech industry have been overemphasized. A few prominent companies made some bets that didn't pay off, and now their balance sheets are out of whack. Layoffs are easy. But those layoffs are still relatively small compared to the industry as a whole, and the companies doing those layoffs are (with very few exceptions) aware that the last thing they want to do is compromise their online presence. So I don't foresee lots of newly unemployed SREs on the job market. But as I said, I do expect the supply to grow to meet the demand.
Mike Loukides
VP of Emerging Tech Content, O'Reilly Media

SRE CHALLENGED BY MICROSERVICE ARCHITECTURE

In 2023 and beyond, enterprises will be challenged by the massively distributed nature of microservice architectures and meeting stringent service level objectives (SLOs) in production. To diagnose issues and restore service after a failure, Site Reliability Engineering (SRE) teams will need a fine-grained view of how transactions perform as they flow across the many microservices that make up an application. Traditional static monitoring approaches aren't up to the task, especially since microservices are often ephemeral — multiple instances of microservices are automatically spun up and down, scaling processing capacity to match increasing or decreasing workloads. This observability challenge is magnified by the sheer frequency of changes originating from multiple distributed development teams.
Richard Hawes
Product Marketing Director, DevOps, ServiceNow

INCORPORATING HA INTO SRE

With large organizations now managing many hundreds of servers and cloud VMs, all requiring increased availability, means that incorporating high availability (HA) into Site Reliability Engineering principles will become a standard part of DevOps projects. Using SRE, DevOps teams will standardize on HA tools that are capable of decreasing complexity, increase availability and reliability, and automate application aware failovers. The vendors who have products that support multiple OS versions, clouds, applications, and databases will be baked into vendor best practices.
Cassius Rhue
VP, Customer Experience, SIOS Technology

COLLABORATION BETWEEN SRE AND SHARED INFRASTRUCTURE PROVIDER

In 2023, cooperation between SRE and shared infrastructure providers will be increasingly necessary to keep digital services running across hybrid architectures. More and more cloud-native applications will also run in shared cloud infrastructure that will often be managed by separate technology operations teams. Issues with this infrastructure can affect cloud-native services, requiring close cooperation between SRE and the shared infrastructure provider (for example, IT Operations). In practice, this collaboration can be difficult since SRE and technology operations teams often use different monitoring systems, and even mapping microservices to underlying infrastructure is challenging. And in many cases, cloud-native applications also communicate with non-cloud-native applications managed by IT Operations, again requiring tight collaboration when there's an issue at the boundary.
Richard Hawes
Product Marketing Director, DevOps, ServiceNow

ITSM ABSORBED BY DEVSECOPS

DevSecOps platforms will absorb IT Service Management (ITSM) platform requirements in order to continue shifting ops processes left. In order to provide enterprises with much-needed visibility into their value stream, service management will become a critical part of the DevSecOps platform. This will allow service desk data to reside within the unified data model of DevSecOps platforms.
David DeSanto
VP of Product, GitLab

Got to: 2023 Application Performance Management Predictions - Part 5, covering AIOps.