Edge Processor was designed to help customers process, filter, and route data closer to the source, reducing cost, improving data quality, and strengthening security. But until now, deploying Edge Processor at scale required :
Manually provisioning servers
Running installation scripts from the UI
Managing short-lived bootstrap tokens
Configuring external load balancers
Scaling capacity by hand
For customers running hundreds or thousands of edge nodes, this simply doesn’t scale effectively. Modern platforms expect infrastructure to be declarative, automated, and elastic. With this release, Edge Processor becomes a more scalable and an elastic solution in Kubernetes environments.
Customers can now:
Deploy Edge Processor as a Splunk-supported Docker container
Install and manage deployments using official Helm charts
Scale automatically using Kubernetes Horizontal Pod Autoscaling (HPA)
Integrate seamlessly with CI/CD pipelines
This isn’t just containerization; it’s a complete rethink of how edge infrastructure should operate in modern environments. In Kubernetes-native environments, workloads are dynamic. Traffic spikes, pipelines change, and data volumes fluctuate. With Kubernetes-driven Edge Processor:
Capacity scales up or down automatically based on CPU, memory, or Edge Processor metrics
Traffic is distributed using native Kubernetes services and load balancers
Scale-down events drain queues gracefully to avoid data loss
No manual node provisioning required
The result is an ingest layer that adapts in real time, without constant tuning or operational overhead.
Deployment of Edge Processor in Kubernetes will introduce automation. Automation is no longer optional; it’s become table stakes. That’s why when Kubernetes-native Edge Processor is paired with a customers CI/CD infrastructure it enables full lifecycle automation, including:
Creating and managing Edge Processors
Integrating deployment into infrastructure-as-code workflows
Together, containers, Helm, and Kubernetes make EP fully automatable, unlocking reproducible environments and safer rollouts across dev, test, and production.
User Interface Changes
Introducing the UI changes to support the introduction of Edge Processor on Kubernetes.
Unified Install and Deployment Experience
We upgraded the EP install and uninstall experience to support containerized deployments, specifically Docker and Kubernetes (via Helm), alongside the existing standalone model. These new deployment scripts are surfaced through a single, familiar, unified sidebar in the UI.
Users can now select between standalone, Docker and Kubernetes. This keeps the experience consistent while making it easier for customers to adopt containerized EPs without relearning how to deploy or manage them.
Host-‑Based Infrastructure View
The new Hosts view enables high‑level monitoring of how customer infrastructure is provisioned. EP instances are now grouped by machine hostnames, allowing customers to quickly understand where instances are running and how they’re deployed across their environment.
The hosts view displays host metrics, instance health, and counts by instance type, giving a condensed but meaningful overview of their deployments. Before there was a single instances tab which combined displaying metrics by hostname and instances.
In the new UI, we have a hosts tab and an instances tab.
The hosts tab relates entirely to hosts, and displays host level metrics and all hosts, along with the instances they contain and their health.
Figure 1 In this screenshot we can see that there are 3 hosts along with the number and type of instances on the host.
And the instances tab is a flat list of all the instances across all hosts in that EP group, and allows sorting by health, instance type or hostname to provide as much flexibility as possible for infrastructure visibility and monitoring.
Figure 2 This screenshot shows the instance tab. In it we can see the instance types on the host.
Instance type is an evergreen deployment concept that stays consistent from installation through monitoring. Customers select an instance type at install time and can later view, filter, and group instances using the same terminology across the UI.
Figure 3 We can also view instances on a host.
Kubernetes-native Edge Processor unlocks a new operating model for ingest:
Faster deployments — minutes, not days
Elastic scaling — match capacity to demand
Lower operational risk — automation over manual steps
Modern DevOps workflows — CI/CD, GitOps, IaC
Kubernetes-native Edge Processor is more than a deployment enhancement; it fundamentally transforms how your data pipeline operates. By bringing declarative infrastructure, elastic scaling, and automated lifecycle management directly into the Edge Processor layer, we eliminate the manual friction that has historically slowed ingest modernization. Edge Processor now scales with data volume in real time, adapts to changing pipeline demands, and integrates seamlessly into CI/CD-driven environments. The result is a smarter, more resilient ingest tier — one that optimizes data before it ever reaches indexers, reduces operational overhead, and gives platform teams precise visibility and control. As data pipelines grow more distributed and dynamic, Edge Processor becomes not just part of the pipeline, but the intelligent control points that power it forward.
... View more