Once upon a time, ‘service mesh’ was a term that applied to containerized applications. In this article, we look at why its definition needs to change,and what VMware’s Office of the CTO is doing to improve cloud runtime.
Extending the Meaning of Service Mesh for Multi-Cloud Environments
Like so many things in tech, the meaning behind certain words evolves and morphs over time. Take ‘service mesh’. The most prevalent understanding of a service mesh today is bound up with the idea of providing a low latency, reliable and secure network infrastructure for containerized, microservice-oriented applications. But, as the world moves towards multi-cloud, is this current definition too limiting?
Emad Benjamin and Mark Schweighardt explore this question in a new blog series. Emad, from the VMware Office of the CTO, is the Chief Technologist of Cloud Application Platforms, while Mark is a Director of Product Management, leading efforts around VMware Tanzu Service Mesh. Together, they share deep experience around architecting a multi-cloud environment, and their blog series promises to be valuable reading.
In their first blog, they introduce the idea that CIOs need a new way to tame large complex systems to improve reliability and service delivery. According to them, the current definition of a service mesh is no longer enough. Instead we need to move towards the idea of a service mesh as a multi-cloud runtime.
Next-generation cloud runtime
Emad and Mark open their discussion by exploring the issue of Service Level Objectives (SLO) attainment within organizations responsible for business-critical applications. Most enterprises struggle to meet their SLO commitments. It doesn’t seem to matter if the applications are traditional, modern or hybrid in nature. Nor does it matter whether those apps run on-premise or on a public cloud. Meeting SLOs remains a major challenge.
To help address this challenge, VMware’s Office of the CTO team is currently working on a project in conjunction with multiple VMware business units to design and build a new cloud runtime. This runtime would extend the concept of a service mesh from something that only focuses on containerized applications to one that can run across a multi-cloud environment and support all application types. In other words, the environment found in most enterprises.
Emad talked with Eric Nielsen and me about the idea of cloud runtime on a podcast we did together late in 2019. He and others in the CTO community also authored a whitepaper on the same topic. As well as reading the first blog on cloud runtime, watch out for upcoming blogs in the series – where Emad, Mark and other VMware technical leaders will explore the capabilities required and how VMware is evolving its current portfolio of technologies to address this real-world need.
To learn about how VMware is helping organizations address multi-cloud requirements, visit the Cloud Solutions page on VMware.com. Another good multi-cloud resource is Architect Central, featuring blogs written for architects responsible for building out cloud environments.
Looking to better understand VMware’s unique approach to multi-cloud architecture? Get the definitive guide.