Kubernetes Cluster Nodes and Pods Specifications for Scheduling Algorithms

A collection of real-world Kubernetes specifications from a multi-microservice architecture, designed for testing scheduling algorithms. The data includes YAML and JSON definitions for nodes and pods, partitioned into a homogeneous 8-node cluster and a heterogeneous 42-node production cluster. The repository was authored by Parth Shah and last updated on 2026-05-18.

Use Cases

• Testing custom scheduling algorithms based on authentic placement constraints and resource requests.
• Evaluating bin-packing efficiency using the detailed CPU and memory specifications for nodes and pods.
• Analyzing cluster resource optimization against the heterogeneous topologies of the 42-node production cluster.
• Benchmarking scheduler performance using the uniform node topology of the 8-node baseline cluster.

Strengths

• Includes two distinct cluster profiles: a homogeneous 8-node baseline and a heterogeneous 42-node production cluster.
• The production cluster subset aggregates approximately 324 CPUs and 1327 GB of memory across its 42 nodes.
• Data is sourced from an organization operating a diverse multi-microservice architecture, capturing real-world constraints.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last updated 2026-05-18; freshness should be verified.

Provenance

Source

Harvard Dataverse

Collection Method

Sourced from an organization operating a multi-microservice architecture.

Freshness

2026-05-18