AppDynamics CTO shares best practices for managing cloud performance

Gregg Ostrowski, Executive CTO at Cisco AppDynamics has shared the best practices he believes businesses should adopt when managing their IT performance in the complex world of the cloud.

Gaining visibility into dynamic and complex cloud-native applications and technology stacks is arguably a top priority for technologists in organisations across the Emirates. They recognise that their traditional monitoring and APM tools aren’t designed to cope with the complexity of cloud-native environments, and this is limiting their ability to monitor and optimise IT performance and availability and deliver seamless digital experiences to end users.

This is why we’re seeing an acceleration in the adoption of full-stack observability solutions across every industry. The latest AppDynamics report, The Journey to Observability, reveals that 55% of businesses in the UAE have now started the transition to full-stack observability, and a further 38% plan to do so during 2022.

However, in the rush to make the transition to full-stack observability and consequently generate more data and insight into cloud environments, it’s vital for technologists to think carefully about the solutions that they are implementing. They need to ensure they select technology that not only addresses their immediate pain points, but that will also be future-proofed as they continue on their digital transformation journey.

To that end, here are four crucial areas for technologists to consider when thinking about observability in the cloud:

Modern cloud-native environments require new generation observability solutions

Technologists should be looking to implement a new generation solution which can observe distributed and dynamic cloud native applications at scale. Traditional monitoring solutions continue to have a huge role to play, but there will increasingly be certain limits to their functionality in cloud environments. For instance, when new use cases are added to existing solutions, data can sometimes remain disconnected, which requires users to jump between tabs in order to identify the root causes of performance issues.

There is an inherent difference in the way traditional and future applications are built, and they’re being managed by different IT teams with different skills and ways of working — that’s why we’re seeing a massive demand for Site Reliability Engineers, DevOps and CloudOps. These new teams require a completely different kind of technology to track and analyze performance data. They need a solution that is truly customised to the needs of a cloud-native technology stack to decipher short-lived microservices interactions with one another and which can be long gone once troubleshooting is done.

The pivot to open standards

Technologists need to ensure that the solution they choose accepts open standards, giving a full-stack, correlated view of all telemetry data — most notably, OpenTelemetry.

OpenTelemetry is a vendor-agnostic and portable standard for metrics, logs and traces (MLT) telemetry. It allows technologists to collect information from every part of their IT stack and not just the mission-critical applications where most have typically deployed agents to collect performance data until now.

Across all verticals, technologists are realising the benefits of OpenTelemetry as a simple and quick way to generate visibility into performance and availability across every single technical element of their IT environment. As Gartner describes it: OpenTelemetry’s emergence as an open standard for telemetry collection promises improved interoperability and greater observability of microservices, containers and Kubernetes.

This shift towards OpenTelemetry will undoubtedly gather pace over the coming years. And so technologists need a solution capable of consuming, processing and correlating the huge volumes of data that OpenTelemetry delivers. It also needs to be capable of providing a holistic view on all of the trace data which technologists now have at their fingertips, to understand IT performance and availability at a business transaction level.

Ability to deliver business-level insights hinges on advanced AIOps

UAE technologists need to leverage the power of AIOps to prioritise actions for their cloud environments. In the future, organisations will leverage AI-assisted issue detection and diagnosis with insights for faster troubleshooting.

Ultimately, this will allow technologists to identify and remedy the root causes of issues faster, minimising the impact on end user experience.

This technology can also enable prioritisation based on business impact, so technologists can focus on what really matters most to end users and the business, and business leaders can understand the impact of IT performance on business transactions.

The transition to cloud observability is a marathon, not a sprint

Despite the increase in cloud computing initiatives, most enterprise businesses in the region, and across the globe for that matter, are still running the core of their application portfolio in traditional environments. And while technologists may have intentions to make the transition to cloud-native architectures over the next few years, it will more likely be a case of extending rather than entirely replacing their application portfolio. Traditional monitoring solutions are still performing a critical role, enabling technologists to identify issues and take appropriate action on a large part of the IT landscape. This will continue to be the case for some time to come in most organisations.

Technologists need to think about cloud observability as a multi-stage process which supports and enables the organisation’s wider digital transformation journey.

And this means that they need to work with a partner that can guide them as they make the switch, whether that is in two years or 10 years. They need a technology partner that can meet them where they’re currently at and then help them to transition to a cloud-native solution in a gradual and seamless way, when the time is right for them.

As end-user expectations for exceptional digital experiences continue to rise amongst both customers and employees, technologists in the Emirates know that they simply cannot afford any slip-ups when it comes to IT performance and availability. But currently, too many simply don’t have the tools and insights they need to identify and resolve issues within a fragmented and dynamic hybrid cloud environment.

With the accelerated move to cloud-native applications, technologists need to ensure that their monitoring capabilities keep pace. From understanding how highly-distributed cloud-native applications work and preventing incidents, to adopting new ways to gather vast amounts of MELT telemetry data, technologists need contextual insights that provide business context deep within the tech stack.

This is why it is imperative for technologists to implement the right cloud-native observability solution — one that can meet their immediate needs and also allow them to scale functionality as their organisation continues its own unique journey to the cloud.