Performance Benchmarks

Benchmark results for Arc are published on the Basekick blog rather than in these docs, so that every figure stays tied to the hardware, dataset, and Arc version it was measured on.

Published results

Start with the ClickBench summary, which covers methodology, dataset, and how Arc was configured for each run:

• Arc on ClickBench — methodology and headline results

Per-database comparisons:

• vs InfluxDB
• vs TimescaleDB
• vs DuckDB
• vs CrateDB
• vs StarRocks
• vs Elasticsearch

Additional runs:

• Cold-run results — first-query latency against object storage
• Log benchmark

Reproduce locally

The benchmark harness ships in the Arc repository:

git clone https://github.com/basekick-labs/arc.git
cd arc
make bench

Benchmarking an Enterprise cluster

Published numbers are measured on a single node. A clustered deployment adds variables that dominate the result, so size from your own measurements rather than from the blog figures:

• Node roles. Writers, readers, and compactors are benchmarked separately — a reader's query throughput is unrelated to a writer's ingest ceiling. See Clustering.
• Storage topology. Shared object storage and local storage with peer replication have different latency profiles. See Deployment patterns.
• Tiered storage. Queries that reach cold-tier data read it from object storage, a latency hot-tier queries do not pay. See Tiered storage.
• Query governance. Rate limits and row limits cap throughput by design; benchmark with the limits you intend to run. See Query governance.

Next steps

• Getting started — run Arc Enterprise locally
• Configuration — tune for your workload
• Python SDK — client for driving load