Groundcover aims to improve observability and monitoring with eBPF and microservices

More organizations are becoming digitally transformed and migrating their workloads to the cloud. However, successfully building applications in the cloud requires a comprehensive view of development pipelines, from workflows to deployment and more. A major challenge development teams have faced over the years is figuring out how to effectively detect and address the issues that occur along their production pipelines, leading to a greater need for more agile application performance monitoring (APM). This is why observability and monitoring are critical to application development, management and performance.

The problem, however, is that many IT teams still rely on traditional monitoring tools that are often slow and not agile. This problem is what Israel-based startup Groundcover wants to solve by combining eBPF (extended Berkeley Packet Filter) and microservices architecture. Groundcover says it’s redefining traditional application monitoring by giving IT teams access to monitor and fix issues within their application pipelines in record time.

Yechezkel Rabinovich, CTO and cofounder at Groundcover, told VentureBeat that Groundcover approaches application monitoring from a new perspective, using an eBPF-based agent and distributed computing to improve application observability. 

Groundcover’s approach to observability

eBPF is a technology designed to run sandboxed programs in the Linux kernel without having to manipulate the kernel’s behavior or load any kernel modules. Basically, it lets you run sandboxed programs using your operating system. With the ability to make the Linux kernel programmable, developers get a wide range of tools to aid their tasks. 

Given the complexity of software monitoring, it’s become essential to rethink the process, resulting in the architectural bridging between eBPF and microservices, which deliver a more refined application monitoring process.

eBPF first became a staple for the Linux kernel in 2014 as an extension to the original Berkeley Packet Filter, which was introduced in 1993. Almost three decades on, it’s evolved to cover more use cases across more than a dozen projects. With Kubernetes playing a significant role in managing individual application containers today, the approach to monitoring and observability must change to reflect current realities. 

Groundcover’s promise is to offer a new approach that helps developers up their Kubernetes observability game with eBPF. Rabinovich said Groundcover’s solution could instantaneously pinpoint bleeding issues in production and troubleshoot quickly, all with zero code changes.

Evolving with a changing application monitoring industry

Highlighting the changes in the industry and the need for organizations to evolve accordingly, Rabinovich said organizations depending on a traditional observability strategy for distributed apps end up facing what he likes to call “the cost-depth trade-off.”

“You can observe quickly and cheaply. Or you can observe in-depth, but at a high cost in terms of time and effort. You can’t have it all,” said Rabinovich, who added that Groundcover gives developers a better playing field. With the Groundcover solution, he said, instead of collecting and analyzing every bit of data available or sampling it randomly, developers can intelligently sample it by identifying the most interesting data right at the source, then select only that data to send to their observability platform. 

“They can also translate the data into granular, actionable metrics so that it’s primed to be analyzed as soon as it hits your observability portal,” added Rabinovich.

Maintaining visibility in Kubernetes clusters 

Groundcover helps IT teams track container state information continuously to monitor what’s happening in containers and pods. Ordinarily, this requires a deep understanding of how Kubernetes tracks container and pod status, how it reports error information and how you can efficiently collect all of the information, but Groundcover says it helps to simplify this complex process. As Rabinovich puts it, “Groundcover is key for any team that wants to maintain reliable visibility into its systems without paying a fortune in monitoring and data storage costs.”  

Founded in 2021, Groundcover has competition in ContainIQ, Pixie, DataDog and NewRelic, with the latter two recently named leaders in Gartner’s 2022 Magic Quadrant for APM and Observability. However, Rabinovich claims Groundcover is a step ahead because its competitors require some sampling mechanism or to send/save everything. 

On the other hand, he noted that Groundcover uses edge computing to convert traces to metrics on the node itself. Rabinovich claims that, unlike its competitors, Groundcover can handle larger volumes of data with much fewer resources.