Expert Interview: From Reactive to Proactive Configuration Awareness With Four-Dimensional Observability
About
This content is brought to you by Evolven. Evolven Change Analytics is a unique AIOps solution that tracks and analyzes all actual changes carried out in the enterprise cloud environment. Evolven helps leading enterprises cut the number of incidents, slash troubleshoot time, and eliminate unauthorized changes. Learn more
Every organization should know the importance of digital transformation. But do they understand how important observability is to make the digital transformation happen?
More importantly, do they understand that managing configuration change is the missing ingredient in most vendors’ concept of observability, which is the secret to agile organizations?
Observability needs to be understood as four-dimensional (4D) for organizations to truly embrace change with deep awareness from code to customer.
If we’ve got your attention, keep reading to learn:
- Are vendors implementing actual (4D) observability?
- Why aren’t vendors addressing 4D observability head-on?
- Can 4D observability help make a reactive business proactive?
- How does 4D observability help DevSecOps?
- How should legacy configuration and change management evolve and adapt to 4D observability?
Former Gartner Research Director Charley Rich brings his four decades of experience to bear with industry veterans CJ Metz, VP of Strategic Accounts at Trace3, and Sasha Gilenson, Evolven CEO and Founder. They share key insights on how four-dimensional observability can radically transform your organization, keeping it both agile and scalable.
Are Vendors Implementing Actual (4D) Observability?
Many vendors talk about observability. But they likely don’t mean the same thing by it or use it in the same ways.
For organizations to really benefit from observability, it must be four-dimensional and, as Charley explains, “not just reactive monitoring++.” It must go beyond analyzing symptoms for performance issues post fact and begin looking for the risk inherent in configurations as they are and as they change. And detect these before real problems manifest.
Are organizations appropriately utilizing 4D observability?
“The short answer is no,” CJ says.
While most organizations do want the right platform with these capabilities, 4D observability “still remains out of reach for most of the clients that we talk to,” he continues. This even applies to clients with strong change control processes, configuration management databases (CMDBs) and change advisory boards (CABs).
The reason for this disconnect comes down to how organizations view — or more accurately don’t view — configuration change. “We find that most organizations are still highly siloed,” CJ explains, “and that [they] don’t have great partnerships between app dev, security and infrastructure.” Thus, they don’t see the risk that a configuration can have on the organization’s stability, compliance, and security concerns.
Various platforms integrate logs, metrics, and traces, and they apply machine learning (ML) to detect dynamic anomalies. But as Sasha says, “Technology alone is not enough — there should be [emphasis on] the process [that] the organization [is] supporting, and that’s not always there.”
And the logs, metrics and traces examined as part of traditional observability typically do not contain any telemetry data about changes to configurations and the parameters contained within them. Nor do they provide a way to analyze risk and score accordingly.
If great standards are only applied within some teams or business units, but not across the whole organization, you don’t get the desired agility from correctly applying observability or, by extension, automation.
Better integrated platforms come from diligently observing configuration change.
“The right culture… [and] technology that actually helps to bridge the siloes,” Sasha explains. This is achieved with a single pane of glass that everyone can observe and learn from. This single pane of glass must display a democratized view across all configurations regardless of the technology stack and do this in a way that specialized knowledge per technology is not required.
Why Aren’t Vendors Addressing 4D Observability Head-On?
The need for and benefit of 4D observability is clear, but what’s not as clear is why vendors aren’t taking advantage of it.
CJ has an answer: “The reason we don’t see traditional vendors necessarily focusing on this has a lot to do with the fact that it’s adding yet another component to what we would traditionally focus on with metrics, traces, and logs.”
Solution providers with specialized capabilities don’t necessarily want to tackle the challenge of tracking configuration changes even if it provides a more meaningful solution. According to CJ, the issue doesn’t stem from not understanding the problem. He says, “It’s more an issue of [not] having a solution that can properly address the problem.”
“Risky configuration changes typically [result] in the greatest amount of production downtime,” Charley says. So, a platform that can transparently explain these challenges to both vendor and customer will lead to more uptime.
It’s not a blame game — it’s about ensuring the business operates as it should. Avoiding finger-pointing has obvious benefits for business functioning.
Can 4D Observability Help Make a Reactive Business Proactive?
Productivity demands a proactive approach to preventing, rather than fixing, problems.
“When you start seeing symptoms, it’s already too late,” Sasha says.
By integrating configuration change into your concept of observability, however, you can become more proactive and less reactive.
If not, Charley adds, “You’ll risk going down a rabbit hole trying to treat symptoms - those pointed out by metrics, traces, and logs- instead of getting to the true root cause.” Unfortunately, most observability tools are focused on performance issues but don’t relate these issues back to an initial cause of a risky configuration.
The adjustment constitutes a larger cultural shift, though. “We’ve had a lot of good metrics, traces and logs for a long time,” CJ points out. “What we’ve lacked is the [concept of] configuration risk intelligence.”
By including configuration risk intelligence, 4D observability drives better behavior across organizations. It could even lead to fewer employment terminations over unauthorized configuration changes.
“The awareness of what can change in a configuration and what does change is often confused in a lot of organizations,” Charley says. “If we’re looking at a summary view of what changed, we are really missing what might be the cause of the problem. To move from reactive to proactive, that deep understanding of anything in the configuration that could change must be front and center.”
How Does 4D Observability Help DevSecOps?
Observability helps establish stability with DevSecOps teams expanding their purview to look at security and compliance issues. Beyond performance and availability, “security has been focused on anomaly detection for a long time, and configuration change is a critical and core component of that,” CJ says. Establishing whether configuration changes were approved or not is crucial in the security landscape.
Reducing downtime is also critical for infrastructure and security. “We need one lens to look at all configuration change,” CJ says.
If ‘configuration change’ is one of the pillars of observability — as in its 4D form — you can “actively identify configuration risks and take action,” Sasha says. “You don’t need to wait for the symptoms to appear…you start from the cause.”
The technology is there — it’s just a matter of leveraging it with the right culture and process in place to support its implementation.
“DevSecOps is often without a democratized single view into detailed configuration,” Charley says.
Yes, in environments using configuration as code, a repository such as GitHub may be employed to store configurations. However, it and its alternatives have major limitations - most importantly, they do not provide complete visibility of configurations across the end-to-end, hybrid cloud.
Today’s configuration repositories are typically utilized for specific environments and not for configurations across all projects, technologies, and locations. Each tool requires appropriate expertise and access.
A “layer of abstraction” across all the various sources is necessary to deliver a single, usable view of misconfigurations and their risk. Implementing this adds in the missing 4th dimension.
How Should Legacy Configuration and Change Management Evolve and Adapt to 4D Observability?
“There’s a strong connection between monitoring and the CMDB and legacy environments,” Charley says. “To understand the impact of alerts and investigate root causes, the dependencies between configurations and other assets and artifacts must be automatically analyzed to be able to track potential causality”
While some CMDBs support the capture of dynamic configuration changes in cloud native environments, they do not provide a democratized dashboard across all configuration types, nor do they provide consistent access to make this a usable solution. Typically, organizations approach this in a siloed manner with come configuration data in the CMDB, others in Cloud native tools such as AWS, and still others in GitHub.
Furthermore, within every digital transformation are challenges around migration and agility because every organization undergoing one already has a CMDB to rely on.
“In many cases processes [like data review and deduplication] are manual and not automated,” Sasha explains.
Efficient change and configuration management in hybrid, multi-cloud deployments require automation — just as all the observability pillars do. But it also requires configuration awareness and a pervasive layer abstraction across all configurations enabling enterprises with the configuration risk intelligence necessary to successfully conduct their digital transformation.
Rapid expansion of hyper-scalable cloud-based infrastructure simply isn’t possible without configuration awareness CJ highlights. “[Tying] impacted business functions — the key to the future — [to observability] is how we increase agility [and] speed.”
Evolven is an AI-driven platform that uses machine learning to automatically collect information about configuration changes carried out in your environment. Request a demo to see for yourself how Evolven can help you navigate the risk of configuration changes and the challenges of digital transformation.
