Use Cases

Decentralized architectures
require decentralized LBs

Introducing ZeroLB, a better way to build load balancing via service mesh that is more perfomant and reliable than legacy load balancers.

    Remove the biggest bottleneck

    Load Balancing Diagram

    Centralized load balancers have no place in decentralized architectures, with ZeroLB we can provide intelligent and portable load balancing on K8s and VMs.

    Increase performance by 2-4x

    Load Balancing Diagram

    By removing an extra hop in the network we can decrease network latency by 2x on request lifecycles and 4x on full roundtrips. Being slow is not an option.

    Reduce annual TCO by 6-7 figures

    Load Balancing Diagram

    Without a centralized load balancer in front of every service in our organization, we can reduce costs up to 7 figures per year while decreasing complexity.

    Case Study

    MS3 Selects Kong Mesh to Power its Next-Generation Integration Platform

    Yahoo! Japan - Looking Beyond the API Platform

    Verifone Trusts Kong for its Global, Omnichannel Payments Solution

    Papa John’s Accelerates Time to Market 3X with Kong

    Problem

    “Our customers want to provide the best possible experience for their customers, and being agile and able to rapidly improve their software is critical to that,” said Aaron Weikle, CEO of MS3. With more of its businesses moving digital, MS3 needed a simple way to address cross-cutting concerns such as encryption, identity and tracing to accelerate the building of new cloud native applications using containers and Kubernetes.

    Solution

    MS3 selected Kong Mesh to form the backbone of its new Enterprise Integration Platform. While other service meshes provided part of the solution with handling Kubernetes workloads, they fell short on their support for the virtual machines that form the bulk of MS3’s enterprise customers workloads.

    Products

    Being able to inject end-to-end security and tracing without introducing performance issues is huge for our customers. With Kong Mesh and Kong, we know that our customers will get a flexible, high-performance solution for all their needs, now and in the future.Aaron Weikle, CEO at MS3

    Problem

    Innovation is at the core of Yahoo! Japan’s DNA. With its shift to microservices and an API-centric architecture, Yahoo! Japan needed more than just a gateway – it needed a true, high-performance platform built for centralizing and managing APIs across a modern architecture.

    Solution

    “We moved rapidly towards microservices over the past few years, and we needed to centralize our APIs,” said Kanaderu Fukuda. With Kong as a single point for proxying and routing traffic across all its API endpoints, Yahoo! Japan was able to centralize authentication and authorization functionalities.

    Products

    With Kong, we knew that we would get something that would work perfectly with our technology today and set us up for where we want to go. Kong allows us to deploy on-premise, easily integrates with all the technologies we use across the company and gives us the flexibility we want to avoid vendor lock-in.Kanaderu Fukuda, Senior Manager at Yahoo! Japan

    Problem

    Verifon needed to streamline a large-scale, complex global payments system without disrupting services for global customer and merchant base. This posed a challenge for the global architecture team at Verifone since ecommerce and physical payments are traditionally supported by completely different systems.

    Solution

    Initially, the team at Verifone liked Kong because of its lightweight API gateway and the company’s focus on microservices. “As we learned more, Kong really stood out for being able to support lots of use cases.” said Hans van Leeuwen, Lead Architect at Verifone. “Kong’s flexible plugin architecture meant we could pick and choose the functionality we needed for each use case.”

    Products

    With the number of new services we’re adding and creating from decoupling legacy applications into microservices, manually managing our infrastructure would be nearly impossible. Using Kong with Kubernetes allowed us to make decentralization a reality.Jason Walker, Senior Enterprise Architect at Cargill

    Problem

    In the fast-evolving retail and restaurant marketplace, consumers are discovering and engaging with restaurants in a variety of formats and on a variety of devices. With integration to aggregators and other emerging channels representing a critical business opportunity for Papa John’s, refining the API strategy became a priority for the company’s technology team.

    Solution

    Papa John’s chose Kong Enterprise as the service connectivity platform, including developer portal and integrations to CI/CD pipeline. Commerce Platform Team is at the center of the technology strategy for tapping into innovation wherever it resides, with integrations to various store, digital, and partner channels.

    Products

    As these integrations scaled up, the addition of a robust API Gateway solution like Kong to our technology stack was a no-brainer.Sarika Attal, Senior Director at Papa John’s

    Deploy ZeroLB architectures

    Kong Mesh can automatically inject a modern service mesh across every service, API, and application to provide smart decentralized load balancing.

    • Round Robin, Least Connect, Ring Hash, Maglev, Random
    • Traffic splitting, weights, canary, blue/green, feature flagging.
    • Support for multi-clusters and multi-cloud.
    • Injects on Kubernetes, VMs and bare metal.
    • Zone aware load balancing to reduce egress costs.
    • Portable across every cloud and datacenter.

    Self-healing and intelligent

    With ZeroLB, load balancing not only gets faster, but it also provides automatic self-healing and zone failover to maximize uptime.

    • Dynamic health checks and circuit breakers
    • Zone failover across both containers and VMs
    • Cross zone load balancing to lift and shift workloads
    • Dyamically configured via Envoy proxy

    Native load balancing telemetry

    Load balancing our workloads is not enough, being able to determine where and when error occurs is critical to determine the health of our applications.

    • 70+ observability charts out of the box
    • Service topology map
    • Native tracing via Zipkin, Jaeger or Datadog
    • Logging and metrics collection to any 3rd party

    Related Use Cases

    Zero-Trust Security

    Provide a platform to deliver zero trust security.

    Learn More →

    Service Mesh

    Turn connectivity into electricity with Kong Mesh

    Learn More →