Cloud Parts

Azure Blob Storage

Azure’s object storage, with three blob types and an optional real directory tree for analytics.

What it is
Object storage
Type
Managed service on Azure
Made by
Microsoft
License
Proprietary
Facts checked
5 October 2026
Website

What it is

Blob Storage nests data three levels deep: a storage account holds containers, and containers hold blobs. Unlike most object stores, a blob has one of three types, fixed at creation. Block blobs are ordinary files, uploaded as up to 50,000 blocks and committed in one step, for a maximum of about 190.7 TiB. Append blobs only grow at the end, which suits logs. Page blobs are made of 512-byte pages you can overwrite in place, up to 8 TiB, and they are what Azure VM disks are built on.

Block blobs sit in an access tier: hot, cool, cold or archive. Cool and cold bill for at least 30 and 90 days. Archive is offline: it bills for at least 180 days, and a blob must be rehydrated before you can read it, which can take up to 15 hours. Turning on the hierarchical namespace (Azure Data Lake Storage) adds real directories, so renaming or deleting a folder updates one entry instead of millions of blobs. Spark and Hive jobs that rename their output depend on this.

Why teams pick it

  • Append and page blobs cover logs and random writes that plain objects can’t
  • Four tiers plus lifecycle rules, and a smart tier that moves blobs between hot, cool and cold by usage
  • Hierarchical namespace gives atomic directory renames for analytics engines
  • Also reachable over NFS 3.0 and SFTP when enabled on the account

Watch out for

  • No S3 API: tools need Azure’s own REST API or SDKs
  • Archived blobs take up to 15 hours to rehydrate, and the archive tier only works with LRS, GRS or RA-GRS redundancy, not zone-redundant accounts
  • Early deletion fees apply if you delete or re-tier a blob before the tier’s minimum days
  • The hierarchical namespace can’t be turned off once enabled, and some Blob features don’t support it

Alternatives

What all object storage have in common →
Amazon S3AWSAWS’s object storage, and the de facto API most object stores copy.Apache OpenDALOpen sourceA Rust library, with bindings for Python, Java, Node.js, Go and more, that gives one API (read, write, stat, list, delete) over 50-plus storage backends: S3 and other object stores, local and HDFS file systems, Google Drive and Dropbox, and key-value stores.Apache OzoneOpen sourceA Hadoop-born object store that serves the same buckets over S3 and as a Hadoop file system.Backblaze B2 Cloud StorageCommercialLow-cost object storage on Backblaze’s own hardware, with its own API and an S3-compatible one.CephOpen sourceA distributed storage cluster that serves S3 buckets, block devices and a file system from the same pool of disks.Cloudflare R2CommercialS3-compatible object storage from Cloudflare that charges nothing for data going out.GarageOpen sourceA lightweight S3 object store that keeps copies of your data in several sites linked by ordinary Internet connections.Google Cloud StorageGoogle CloudGoogle Cloud’s object storage, where even the archive class answers in milliseconds.MinIOArchivedOpen sourceA self-hosted object store that speaks the S3 API. The open-source edition is archived and its images are gone from Docker Hub; MinIO now sells AIStor.PGSTY SiloOpen sourceA community-maintained fork of the open-source MinIO server, with builds, images, security fixes and the full web console.RustFSOpen sourceA young Rust reimplementation of MinIO’s design: S3-compatible, erasure-coded, Apache 2.0, and generally available since September 2026.SeaweedFSOpen sourceA distributed blob store with an S3 gateway and a file system layer, built to store and serve billions of small files.TigrisCommercialGlobally distributed S3-compatible object storage that moves data to where it’s read, with no egress fees.Versity S3 GatewayOpen sourceA stateless S3 server that exposes an existing file system, or another object store, through the S3 API.Wasabi Hot Cloud StorageCommercialS3-compatible object storage sold at one flat price per TB, with no egress or request fees.Zenko CloudServerOpen sourceScality’s open-source Node.js S3 server: a light S3 endpoint for development and tests, and a front end that can store objects in other clouds.

How it fits with other technologies

What uses Azure Blob Storage

Create an account and container, upload to cool, then archive · bash

az storage account create --name acmemedia --resource-group media-rg \
  --location westeurope --sku Standard_GRS --kind StorageV2

az storage container create --account-name acmemedia --name photos --auth-mode login

az storage blob upload --account-name acmemedia --container-name photos \
  --name 2026/10/a.jpg --file a.jpg --tier Cool --auth-mode login

# archive is offline: reading it later needs a rehydrate of up to 15 hours
az storage blob set-tier --account-name acmemedia --container-name photos \
  --name 2026/10/a.jpg --tier Archive --auth-mode login

See it in a real stack

No architecture uses Azure Blob Storage yet. Drop it into the playground and wire it up yourself.Open playground →
Where it came from2008–2026

Microsoft announced Windows Azure on 27 October 2008 at its Professional Developers Conference, where chief software architect Ray Ozzie described it as “the bedrock, underneath all of Microsoft’s service offerings”. Its storage service, Windows Azure Storage, went into production that November with three abstractions: Blobs for files, Tables for structured data and Queues for messages. The platform became commercially available on 1 February 2010.

Brad Calder and a large Microsoft team described the design at SOSP in 2011. Where S3 was then eventually consistent, Azure promised strong consistency, because enterprise customers moving line-of-business applications to the cloud asked for it. The system grew out of Cosmos, Bing’s internal storage system, which Azure extended into an append-only “stream layer”. Microsoft used it heavily itself: one Bing pipeline kept around 350 TB of public Facebook and Twitter posts in it to make them searchable within 15 seconds.

Later additions turned it into an archive and a data lake: an offline Archive tier in 2017, a hierarchical namespace for analytics (Data Lake Storage Gen2) in 2019, and automatic tiering in 2026.

Why it’s built this way

Strong consistency from the start
Customers wanted a read to see the latest write, and conditional updates for optimistic concurrency. Azure put a strongly consistent partition layer on top of a simple replicated layer to get both availability and consistency within a cluster.
Append-only storage underneath
Never overwriting committed data “greatly simplified the replication protocol”, made snapshots almost free and eased erasure coding. The price is a garbage collector that reclaims dead space.
Three blob types instead of one
Page blobs, with 512-byte pages you can overwrite in place, let virtual machine disks live in blob storage; append blobs fit logs; block blobs cover ordinary files.
An optional real directory tree
On a flat namespace, renaming a folder of 5,000 blobs took 10,000 REST calls and could fail halfway. Data Lake Storage Gen2 made such renames a single atomic operation for Hadoop and Spark.

Timeline

  1. 2008

    Windows Azure announced at PDC in October; storage enters production in November

  2. 2010

    Commercially available on 1 February, with Blobs, Tables and Queues

  3. 2011

    SOSP paper describes Windows Azure Storage and its strong consistency

  4. 2017

    Archive tier generally available at $0.002 per GB-month (December)

  5. 2019

    Data Lake Storage Gen2 adds a hierarchical namespace; GA on 7 February

  6. 2026

    Smart tier moves blobs between hot, cool and cold automatically (April)

Sources9 · facts checked 5 October 2026

Improve this page on GitHub