<!-- mobian-agent-page publisher="dailydev" canonical="https://daily.dev/posts/apache-iceberg-manifest-files-explained-vois550g1" -->

---
title: Apache Iceberg Manifest Files Explained | daily.dev
description: Manifest files are Avro-based components of Apache Iceberg's metadata tree that list data files and their column-level statistics, enabling two-stage query...
canonical: https://daily.dev/posts/apache-iceberg-manifest-files-explained-vois550g1
twitter:card: summary_large_image
twitter:site: @dailydotdev
og:type: website
og:site_name: daily.dev
og:title: Apache Iceberg Manifest Files Explained | daily.dev
og:description: Manifest files are Avro-based components of Apache Iceberg's metadata tree that list data files and their column-level statistics, enabling two-stage query...
og:url: https://daily.dev/posts/apache-iceberg-manifest-files-explained-vois550g1
og:image: https://api.daily.dev/og/posts/vOis550G1.png
og:image:alt: Apache Iceberg Manifest Files Explained
og:image:width: 1200
og:image:height: 630
og:locale: en
---

> ## Documentation Index
> Fetch the complete documentation index at: https://daily.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# Apache Iceberg Manifest Files Explained

**[Dremio Blog](https://daily.dev/sources/dremio)** · 14 min read · 0 upvotes · 0 comments

## Summary

Manifest files are Avro-based components of Apache Iceberg's metadata tree that list data files and their column-level statistics, enabling two-stage query pruning: partition-level skipping via the manifest list, then file-level skipping via column bounds inside surviving manifests. The piece explains what fields a manifest entry contains, why Avro (row-oriented) rather than Parquet is used for metadata, how delete files and deletion vectors also live in manifests, and why manifest count grows over time through commits, deletes, and compaction. It covers the table properties controlling automatic manifest merging, practical SQL queries against Iceberg's metadata tables to diagnose slow query planning, and the distinction between partitioning (which manifests open) and sorting (which files get pruned via column bounds). It closes with a plug for Dremio's native Iceberg support and autonomous table maintenance.

## Full article

daily.dev links to this article rather than hosting it. Read it at the original source: <https://www.dremio.com/blog/apache-iceberg-manifest-files-explained>

## Questions this post answers

### What fields does an Apache Iceberg manifest entry contain?

A manifest entry contains file_path, file_format (PARQUET, ORC, or AVRO), partition tuple, record_count, file_size_in_bytes, column_sizes, value_counts, null_value_counts, nan_value_counts, lower_bounds, and upper_bounds. The bounds are keyed by column id rather than column name, which is why renaming a column in Iceberg does not invalidate existing statistics or require rewriting metadata.

_Engineers tuning Iceberg scan performance can dig into schema-level breakdowns like this one on daily.dev._

### Why are Apache Iceberg manifest files stored as Avro instead of Parquet?

Manifests use Avro because it is row-oriented, matching how manifests are actually read: every field of every entry, start to finish, each time the manifest is opened. Columnar layout would add decode overhead with no benefit since no query wants only some columns of a manifest. Avro also embeds its schema in the file, letting older manifests stay readable as the format adds new fields.

_Anyone deciding on metadata formats for a data lakehouse can compare tradeoffs like these on daily.dev._

### Why does an Apache Iceberg table's manifest count keep growing even as file count stays stable?

Manifest count grows because Iceberg never mutates existing metadata, it only appends and later releases: every commit adds at least one manifest, deletes add a manifest recording the deletion without removing the underlying data file, and compaction writes new manifests for rewritten files before old ones become reachable for expiry. A streaming job committing every minute can produce 1,440 manifests a day before any merging runs.

_Teams debugging slow Iceberg query planning can track down root causes like this on daily.dev._

## Similar posts on daily.dev

- [What Is Inside an Iceberg Table?](https://daily.dev/posts/what-is-inside-an-iceberg-table--p5dgradez) · DuckDB · 2 upvotes · 0 comments
- [Apache Iceberg Architecture Deep Dive](https://daily.dev/posts/apache-iceberg-architecture-deep-dive-1znwuwiy4) · BigData Boutique blog · 5 upvotes · 0 comments
- [Maintaining Apache Iceberg Tables: Compaction, Snapshots, Metadata and Orphan Files](https://daily.dev/posts/maintaining-apache-iceberg-tables-compaction-snapshots-metadata-and-orphan-files-vk7yjrfyp) · ITNEXT · 0 upvotes · 0 comments
- [Apache Iceberg vs Apache Parquet Explained](https://daily.dev/posts/apache-iceberg-vs-apache-parquet-explained-9kadpsw5t) · Dremio Blog · 1 upvotes · 0 comments
- [Parquet and Iceberg: An Overview — CommandPrompt Inc.](https://daily.dev/posts/parquet-and-iceberg-an-overview-commandprompt-inc--pv4vposch) · Planet PostgreSQL · 0 upvotes · 0 comments

---

Tags: [#data-engineering](https://daily.dev/tags/data-engineering), [#apache-iceberg](https://daily.dev/tags/apache-iceberg)

[View this post on daily.dev](https://daily.dev/posts/apache-iceberg-manifest-files-explained-vois550g1)

```json
{"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]}
{"@context":"https://schema.org","@type":"TechArticle","headline":"Apache Iceberg Manifest Files Explained","url":"https://daily.dev/posts/apache-iceberg-manifest-files-explained-vois550g1","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/apache-iceberg-manifest-files-explained-vois550g1"},"datePublished":"2026-09-10T17:25:25.451Z","dateModified":"2026-10-15T02:22:14.036Z","description":"Manifest files are Avro-based components of Apache Iceberg's metadata tree that list data files and their column-level statistics, enabling two-stage query...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/cfc8f3990bd5da6b957eeb3062ab72e5?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/cfc8f3990bd5da6b957eeb3062ab72e5?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Dremio Blog","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Dremio Blog","logo":"https://media.daily.dev/image/upload/s--MGKANojy--/f_auto,q_auto/v1786870389/logos/dremio?_a=BAMAMicg0","url":"https://daily.dev/sources/dremio"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/apache-iceberg-manifest-files-explained-vois550g1","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"data-engineering,apache-iceberg","timeRequired":"PT14M"}
{"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Dremio Blog","item":"https://daily.dev/sources/dremio"},{"@type":"ListItem","position":3,"name":"Apache Iceberg Manifest Files Explained"}]}
{"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/apache-iceberg-manifest-files-explained-vois550g1#faq","mainEntity":[{"@type":"Question","name":"What fields does an Apache Iceberg manifest entry contain?","acceptedAnswer":{"@type":"Answer","text":"A manifest entry contains file_path, file_format (PARQUET, ORC, or AVRO), partition tuple, record_count, file_size_in_bytes, column_sizes, value_counts, null_value_counts, nan_value_counts, lower_bounds, and upper_bounds. The bounds are keyed by column id rather than column name, which is why renaming a column in Iceberg does not invalidate existing statistics or require rewriting metadata. Engineers tuning Iceberg scan performance can dig into schema-level breakdowns like this one on daily.dev."}},{"@type":"Question","name":"Why are Apache Iceberg manifest files stored as Avro instead of Parquet?","acceptedAnswer":{"@type":"Answer","text":"Manifests use Avro because it is row-oriented, matching how manifests are actually read: every field of every entry, start to finish, each time the manifest is opened. Columnar layout would add decode overhead with no benefit since no query wants only some columns of a manifest. Avro also embeds its schema in the file, letting older manifests stay readable as the format adds new fields. Anyone deciding on metadata formats for a data lakehouse can compare tradeoffs like these on daily.dev."}},{"@type":"Question","name":"Why does an Apache Iceberg table's manifest count keep growing even as file count stays stable?","acceptedAnswer":{"@type":"Answer","text":"Manifest count grows because Iceberg never mutates existing metadata, it only appends and later releases: every commit adds at least one manifest, deletes add a manifest recording the deletion without removing the underlying data file, and compaction writes new manifests for rewritten files before old ones become reachable for expiry. A streaming job committing every minute can produce 1,440 manifests a day before any merging runs. Teams debugging slow Iceberg query planning can track down root causes like this on daily.dev."}}]}
```

