Kong released version 1.4 of its enterprise-grade service mesh built on top of CNCF’s Kuma and Envoy.
The new release of Kong Mesh 1.4 enables modern, decentralized, and portable load balancing that works across every cloud, including Kubernetes and VMs, and local development machines that improve accuracy and replicability, according to the company.
The company also unveiled a new concept called ZeroLB, a concept that offers up to four times higher network performance and substantial cost reduction when compared to traditional centralized load balancing-based solutions.
It eliminates the need for elastic cloud, software, and hardware load balancers by removing complexity in applications and offering self-healing abilities.
“Traditional load balancers were built for the monolithic era and don’t work well with today’s cloud native applications because they are slow and create points of failure that cause businesses to lose customers and sales,” said Marco Palladino, the CTO and co-founder of Kong Inc. “I’m thrilled to be spearheading this effort at Kong to evangelize the ZeroLB movement, where legacy load balancing as we know it within the network is dead. Kong is introducing a new way of building load balancers for today’s modern era of software architectures by decentralizing it.”
Other than the new ZeroLB pattern, key features in Kong Mesh 1.4 also include:
- Five different decentralized load balancing algorithms, which include round robin, least request, ring hash, maglev and random
- L4 + L7 load balancing to address every service type, from databases and events to HTTP and gRPC services
- Automatic self-healing capabilities with health checks, circuit breakers, zone-aware load balancing and cross-zone connectivity capabilities
- Comprehensive support for both Kubernetes (automatically injected sidecar) and VMs (manually injected sidecars) across every cloud
- Extension of the connectivity logic for Kong Mesh using any WebAssembly-supported language, thanks to Envoy’s native support