Linkerd Service Mesh Now Spans Multiple Kubernetes Clusters

The Linkerd community in collaboration with Buoyant Inc. announced the availability of an update that enables the open source service mesh to be deployed across multiple Kubernetes clusters.

Bouyant CEO William Morgan says version 2.8 of Linkerd is arriving at a time when more IT organizations are starting to distribute microservices-based applications across multiple clusters. The latest version of Linkerd facilitates connectivity between clusters while at the same time providing a framework for governing a large number of microservices, he says.

While rival service meshes such as Istio can be employed to connect multiple clusters. Morgan notes Linkerd provides a lighter-weight approach that is easier to deploy. In fact, Morgan says a large percentage of the IT organizations that have embraced Linkerd formally tried to employ Istio.

Morgan says Linkerd also configures itself and automatically encrypts traffic between applications within a cluster. Organizations that have embraced Linkerd include Chase Bank, EverQuote, Expedia, GoDaddy, Nordstrom and Walmart.

Linkerd is currently an incubating project being advanced under the auspices of the Cloud Native Computing Foundation (CNCF). Istio is also an open source project that has yet to be formally submitted to a governing body. Istio was originally launched as an open source project by Google, IBM and Lyft.

Of course, there also other open source and commercial service mesh platforms. As Kubernetes adoption starts to increase, many organizations are just now starting to appreciate the need for a service mesh to integrate and manage services both within a Kubernetes cluster and between them.

At the same time, some organizations are also looking toward application delivery controllers (ADCs) to provide similar capabilities. ADCs represent more of a legacy approach to managing services, says Morgan.

Microservices as a concept provide a powerful way to build applications in a way that serves to make them more flexible and secure. However, many IT teams are also discovering microservices can be too much of a good thing unless they are properly managed and governed. As microservices become more widely employed, more than a few DevOps teams are struggling with how to manage modules of code that are being continuously updated by different teams within the organization. In some cases, that complexity has led to IT teams reverting to building and deploying a monolithic application that they are more comfortable managing.

However, in the wake of the COVID-19 pandemic many organizations are embracing cloud-native applications to build applications that are not only more accessible to remote users, but also scale them up and down more elegantly as part of an effort to reduce cloud computing costs. IT teams are also savvier about not becoming locked into a specific virtual machine platform on a public cloud computing platform.

It’s not clear to what degree the proliferation of microservices versus the increased number of Kubernetes clusters is driving increased interest in service meshes. Regardless of motivation, however, it’s apparent many more organizations are starting to enter a new phase of adoption that requires tools to manage both the Kubernetes clusters and the microservices deployed on them at scale.

Mike Vizard

Mike Vizard is a seasoned IT journalist with over 25 years of experience. He also contributed to IT Business Edge, Channel Insider, Baseline and a variety of other IT titles. Previously, Vizard was the editorial director for Ziff-Davis Enterprise as well as Editor-in-Chief for CRN and InfoWorld.

Mike Vizard has 1604 posts and counting. See all posts by Mike Vizard