Cisco and Spectro Cloud: Easy Kubernetes® observability

February 13 2024

Deploying Kubernetes in production environments at scale is a common enterprise challenge. Learn how Cisco and Spectro Cloud ease deployment and management of Kubernetes clusters in the cloud, bare metal and virtualized data center, and at the edge.

Begin with the end in mind and lessons learned

“We may be very busy, we may be very ‘efficient’, but we will also be truly ‘effective’ only when we begin with the end in mind.” — Stephen Covey, The 7 Habits of Highly Effective People.

While it may seem unusual to tie rapid tech adoption to life, the theme brings me back to a challenging yet rewarding time. I was living overseas, working full time and completing my MBA with two young children. I was at a tipping point because the MBA program I chose was finance-heavy, and I have a business-tech background. Just like any complex problem, I was able to break it down into manageable pieces, get the help I needed, and start with the end in mind.

Similarly, our customers are faced with information overload and lots of noise when it comes to compliance, security, maintenance, and observability. Keeping up with the pace of change can be unmanageable and the adoption of new technologies, infrastructure platforms, and environments extrapolate the challenge. This can be overwhelming for IT Ops and platform engineering. However, there are ways to reduce the risk — by incorporating business needs into observability up-front, teams can begin with the end in mind!  To deliver on this promise, Cisco Cloud Observability customers can enjoy observability and gain complete peace of mind for Kubernetes environments, wherever they are on day zero, by integrating with the Spectro Cloud Palette.

Kubernetes adoption (at scale) is growing fast

In the past few years, Kubernetes (K8s) and cloud native application development have emerged to play an ever-larger role in application infrastructure. Although Kubernetes has been around for quite some time, only recently have more organizations been using it in production environments across multiple locations.

And the adoption is substantial. According to a recent independent research report, revealing how businesses use Kubernetes:

    • 56% have more than 10 clusters.
    • 69% run Kubernetes in multiple clouds or other environments.
    • 80% expect their scale to grow.
    • Almost half use Kubernetes at the edge, citing AI as a key driver in Kubernetes adoption.

How do you ensure SLOs are being met in complex K8s environments?

Kubernetes is not easy, nor is deploying it in production environments at the scale enterprises need today. It can be a real challenge to keep clusters and the cloud native stack healthy to support application availability and performance needs aligned with business requirements. When deploying Kubernetes in production, it’s a constant struggle to control complexity as you manage the cluster lifecycle across diverse environments while always ensuring the health and availability of the complete infrastructure and the application stack.

Indeed, the same research found:

    • The majority (75%) of Kubernetes users suffer issues that affect their clusters, caused by infrastructure complexity.
    • IT Ops and platform engineering teams say they spend too much time troubleshooting issues and resolving day two operations tasks.

If only there was an easy way to roll out Kubernetes monitoring across all your clusters while keeping business goals across service level objectives (SLOs) in check.

A new solution to a complex problem

We have been working with Spectro Cloud and its next-generation enterprise Kubernetes management platform, Palette. Spectro Cloud Palette makes it easy to deploy and manage Kubernetes clusters in the cloud, bare metal and virtualized data center — and at the edge. It also takes away the complexity of lifecycle management of any Kubernetes distribution and the cloud native integrations deployed to your clusters. Palette is used by global organizations across multiple industries.

To address our joint customer requirements and market demand, Cisco Cloud Observability now integrates with Palette out-of-the-box. The result? You can deploy Cisco Cloud Observability to any Kubernetes location with just a few clicks, giving you a single platform to monitor Kubernetes clusters and complete peace of mind about the health of your cloud native infrastructure and application performance.

Research shows that 47% of organizations report an average hourly cost of downtime at $250,000 or more. That means whenever a performance issue affects your Kubernetes environment, whether it’s a single node in the cluster or a broader resource utilization problem you need to act fast. With a comprehensive view of insights needed to get to root cause and mean time to repair (MTTR) much faster, Pallete helps organizations save time and money.

“As more and more organizations are leveraging Kubernetes for production environments to build new modern apps or modernize existing ones, the need to reduce complexity and risk is paramount. The integration with Cisco Cloud Observability and our Palette platform enables teams to easily deploy and manage Kubernetes-based apps at scale across data centers, clouds, and the edge, while always staying ahead of problems.” 
—Dave Cope, CRO/CMO, Spectro Cloud

Deploying observability couldn’t be easier

The integration is seamless, just add the Cisco AppDynamics “pack” as your preferred monitoring integration in a reusable “Cluster Profile” before you even start deploying your clusters (day zero). Palette will take care of deploying the Cisco AppDynamics operator and collectors to your infrastructure.

The final step is configuring the Cisco Cloud Observability portal, and you are good to go! All new and existing Kubernetes clusters that Palette deploys will collect Kubernetes telemetry data and be proactively monitored to ensure healthy operation.

Figure 1: Spectro Cloud Palette cluster profile with Cisco AppDynamics Collectors pre-configured


Figure 2: Cisco Cloud Observability Kubernetes dashboard

This integration of Cisco Cloud Observability and Spectro Cloud Palette brings the best of the two platforms to your IT Ops and platform engineering teams:

    • Easy: Native out-of-the-box integration means no time wasted in manual configuration of collectors. With a few clicks, you can have any cluster deployed and telemetry proactively monitored, wherever they are, at scale, for tens of thousands of clusters.
    • Comprehensive: Get deep and wide observability across the full stack of your Kubernetes clusters, from the hardware (CPU, memory, disk I/O, etc.) to your preferred OS and Kubernetes distribution, to the applications and any other associated cloud native integration. — including correlation to your holistic application landscape and cloud services.
    • You are in control: Granular controls ensure governance; from tracking usage of Kubernetes pods based on defined requests and limits to detailed logging and powerful day zero to day two lifecycle management features, stay ahead of problems and focus on what’s important.
    • Reduces risk: Gain confidence knowing the integration is validated by Cisco and Spectro Cloud, ensuring day zero operations include observability seamlessly.

Managing technical demands consisting of largely heterogeneous and potentially ephemeral environments while being held to ever rigid business demands tied to SLOs and SLAs is challenging. Therefore, it makes sense to focus on the outcomes of meeting business objectives such as acceptable latency and expected uptime for business-critical applications and infrastructure at the onset. Including observability as part of your initial K8s deployments makes sense since you can’t meet goals if you don’t measure them!

“Strategically it’s a great alliance with Spectro Cloud since collectors come prepackaged providing the quick ‘easy button’ to start sending Kubernetes cluster telemetry to Cisco Cloud Observability, saving valuable time for our joint customers” 
— Amit Patel, VP Engineering, Cisco

Our work with Spectro Cloud reflects a ‘commitment and shared vision to simplify customers’ cloud native and Kubernetes journeys. For more information about how we simplify cluster monitoring, read the datasheet and learn more about Cisco Cloud Observability.

Craig Ginsberg received his MBA from The University of Chicago and Bachelor of Science majoring in Management Information Systems from the University at Albany New York. At AppDynamics, Craig leads ecosystem development and the Integration Partner Program. In his past, he led the SD-WAN global channel strategy for Cisco, developed and led pre-sales teams while living in Asia and helped double revenue in two years. He was a CTO for an IPTV startup and led the product launch for a drone analytics startup. He was also an Associate Partner for IBM Global Business Services focused on Telecom. Currently, Craig lives in Denver with his family where he is active with his MBA alumni and is an avid mountain biker and skier.

Thank you! Your submission has been received!

Oops! Something went wrong while submitting the form