I've spent the past 10 years working in the data analytics & engineering space, architecting nation-state cyber defence systems, consulting for global financial institutions, and leading developer relations & experience teams at early stage startups.
Al Brown
Product Marketing Engineer at ClickHouse
Follow Al Brown
Blog posts by Al Brown
Showing 6 of 14 articles
Resources by Al Brown
Showing 6 of 13 resources
What is columnar storage?
Al Brown • Last updated: May 8, 2026
Columnar storage is an on-disk data layout where each column is stored contiguously. It cuts I/O for analytical queries and enables 5–10× compression.
Why columnar databases are fast
Al Brown • Last updated: May 8, 2026
Columnar databases are 10-1000× faster than row stores on analytical queries because two principles compound: efficient execution (storage layout, vectorisation) and smart pruning (data skipping, late materialisation).
Columnar storage formats: Parquet, ORC, and Arrow explained
Al Brown • Last updated: May 8, 2026
How Apache Parquet, ORC, and Arrow lay out columnar data, and why they coexist instead of converging on one format.
Row-oriented vs column-oriented databases: a head-to-head comparison
Al Brown • Last updated: May 8, 2026
Row stores win point lookups and per-row updates; column stores win wide aggregations and compress 5–10× tighter. A measured comparison with a decision rule.
What is vectorised query execution?
Al Brown • Last updated: May 8, 2026
Vectorised query execution processes batches of values per operator call instead of one row at a time. It is the dominant execution model for analytical databases.
Best columnar databases in 2026
Al Brown • Last updated: May 8, 2026
Nine columnar databases in 2026, ranked using measured benchmark latency and compression rather than marketing copy.