Abhimanyubajaj·Oct 27EKS App Slow, CPU Fine? Scale on Splunk Latency with KEDAThis lab fixes the autoscaling gap CPU HPA misses: users are waiting, but CPU is barely moving.
Mahernaija·Oct 22Autoscaling LLM Inference on Kubernetes with KEDA: Scaling on Queue Depth, Not GPU UtilizationGPU utilization stays near 100% long before an LLM server is saturated, so the autoscaler needs a signal that tracks waiting requests.
Moyo Oyegunle·Oct 14Replacing oc idle: Scale-to-zero on OpenShift without dropping the first requestIdling is how you stop paying for workloads nobody is using: dev and test namespaces and internal tools. Scaling them to zero frees node…
Anish Bhat K·Oct 5KEDA Hands-on: Autoscaling Kubernetes Workloads Using AWS SQSIn my last blog, I covered Kubernetes Event-driven Autoscaling (KEDA), its objectives, use cases, and why it can be useful for event-driven…
Mostafa Soliman·Oct 4The race to zero: KEDA, per-event scaling, and the case for external pushI like scale-to-zero.
Anish Bhat K·Sep 29KEDA: Kubernetes Event-Driven Autoscaling ExplainedWhen we deploy applications on Kubernetes, one of the common requirements is autoscaling.
Rohan Mathure·Sep 20KEDA as Lambda, Part 2: Where the Cold Start Actually GoesI timestamped every stage of a KEDA cold start. Scheduling took 17 milliseconds; polling, image pulls and container start took the rest.
Rohan Mathure·Sep 17KEDA-as-Lambda: Building Serverless Workloads on KubernetesRepo: rohan-mathure/keda-as-lambda Author: Rohan Mathure Reading time: ~12 min
nutcas3·Sep 14Why We Replaced Kubernetes HPA with KEDA — and Saved Our Event-Driven Platform from Silent OverloadHow a high-throughput event-driven architecture was failing to scale at the exact moments it needed to — and how event-driven autoscaling…
InAWS in Plain EnglishbySayan Moitra·Sep 14When Kubernetes Scales Down Mid-Stream: Why Graceful Shutdown Is Not EnoughA practical Kubernetes, Istio, and KEDA experiment in protecting long-running streamed responses with application-level replay.
