Skip to main content
In this lesson we cover AWS Storage Gateway — a hybrid cloud storage service that connects on‑premises environments to AWS cloud storage. Storage Gateway enables on‑premises applications to use cloud storage for backup, migration, disaster recovery, and capacity extension while preserving existing protocols like iSCSI, NFS, and SMB. Key benefits:
  • Seamless hybrid integration between data center and AWS.
  • Support for block, file, and tape workflows.
  • Cost‑effective backup and archival using S3 and Glacier classes.
  • Can be deployed as a virtual appliance or a physical appliance on‑premises.

Common use cases

  • Extend on‑premises storage capacity by offloading data to Amazon S3.
  • Migrate data gradually to AWS by replicating on‑premises datasets.
  • Provide low‑cost, durable backups and archives using S3, S3 Glacier, and S3 Glacier Deep Archive.
  • Enable disaster recovery by restoring snapshots to EC2/EBS when needed.
  • Replace physical tape infrastructure with a Virtual Tape Library (VTL) stored in S3/Glacier.

Deployment model

Storage Gateway runs as an on‑premises appliance (VM or hardware). It presents storage to local servers and asynchronously copies or stores data in AWS. Choose the gateway type that matches your application protocol (block vs file vs tape).

Gateway types (at a glance)

For detailed guidance on when to choose each type, see the sections below.

Volume Gateway

Volume Gateway presents iSCSI block volumes to on‑premises servers. It comes in two modes — Stored and Cached — each designed for different needs.

Volume — Stored mode

In Stored mode, the gateway keeps the full copy of each volume locally on the gateway appliance and asynchronously backs up changes to AWS as EBS snapshots (stored via AWS snapshot storage). Key points:
  • Primary copy resides on local disks attached to the gateway.
  • AWS stores backups as EBS snapshots (persisted in S3 via AWS snapshot storage).
  • Useful for local performance with cloud backups for disaster recovery and restore to EC2.
  • Does not increase on‑prem storage capacity — you are limited by the gateway’s local drives.
A presentation slide titled "Storage Gateway – Volume Stored" with five colored boxes listing features. It notes that data is stored locally on‑prem, replicated asynchronously to AWS S3 for backup and disaster recovery, and that this doesn't increase datacenter storage capacity.
Exam tip: In Volume Stored mode, the primary copy of data resides on the on‑premises gateway; AWS holds asynchronous backups (EBS snapshots backed by S3).

Volume — Cached mode

In Cached mode, the authoritative copy of your data is stored in AWS (snapshots/objects), and the on‑premises gateway keeps a local cache of frequently accessed data (hot blocks) to provide low latency. Key points:
  • Primary data is in AWS; local cache improves performance for active blocks.
  • Acts as a datacenter extension — effectively expands available storage capacity without adding local drives.
  • Servers continue to use iSCSI volumes as usual while most capacity is provided by the cloud.
A diagram titled "Storage Gateway – Volume Cached" showing on-premises servers connecting over iSCSI to a local AWS Storage Gateway with cached storage. The gateway links to the AWS cloud gateway which stores data in an S3 bucket.
A slide titled "Storage Gateway – Volume Cached" showing three colored boxes that state: data is stored on S3; only frequently accessed on‑prem data is cached; and cached mode acts as a datacenter extension to increase storage capacity. The slide is copyrighted by KodeKloud.

File Gateway

File Gateway exposes a file interface (NFS or SMB) to applications while storing file contents as Amazon S3 objects. The gateway maps file paths and filenames to S3 object keys so applications can continue using standard file protocols while benefiting from S3 scalability and durability. Behavior and benefits:
  • Applications access file shares via NFS or SMB as usual.
  • Files are persisted to Amazon S3 as objects; the gateway maps /<path>/<file> to an S3 object key.
  • A local cache stores frequently accessed files for improved read performance.
  • Ideal for lift‑and‑shift file workloads, content repositories, and backup targets that need file semantics.
A diagram titled "Storage Gateway – File" showing on‑premises servers mounting a Storage Gateway via NFS/SMB. The on‑prem gateway forwards data to the AWS Cloud gateway which stores files in an S3 bucket (example: /media/pic1.jpg).

Tape Gateway

Tape Gateway emulates a tape library (VTL) and virtual tape drives so legacy backup applications can continue to use existing tape workflows. Backups written to virtual tapes are stored in Amazon S3 and can be archived to S3 Glacier classes for long‑term retention. How it works:
  • Presents virtual tape drives and a VTL over iSCSI to backup servers.
  • Virtual tapes are stored in S3 (VTL). Less‑frequent tapes can be moved to a Virtual Tape Shelf (VTS) backed by S3 Glacier or S3 Glacier Deep Archive.
  • Virtual tape sizes commonly range from 100 GB to 5 TB.
  • Enables retirement of physical tape hardware while keeping familiar backup software.
A diagram of AWS Storage Gateway — Tape showing on-premises servers connecting via iSCSI to a local storage gateway and tape drive. The gateway links to the AWS Cloud where a virtual tape library (VTL) and virtual tape shelf (VTS) are backed by S3 buckets.
A presentation slide titled "Storage Gateway – Tape" with three numbered panels. It explains that the gateway emulates a tape library, stores data in AWS (VTL on S3 and tape shelf to Glacier), and provides virtual tapes sized 100 GB–5 TB.

Comparing File and Tape Gateways

  • File Gateway: Presents file shares over NFS/SMB; stores files as S3 objects and caches frequently accessed files locally. Best for applications that require file semantics and S3 durability.
  • Tape Gateway: Emulates tape libraries over iSCSI for backup applications; stores virtual tapes in S3 (VTL) and archives to S3 Glacier storage classes (VTS). Best for replacing physical tape infrastructure while keeping backup software unchanged.

Quick decision guide

  • Use Volume Stored when you need a local primary copy of block volumes with cloud backups (EBS snapshots).
  • Use Volume Cached when you want to extend on‑prem capacity while placing the primary data in AWS and caching hot blocks locally.
  • Use File Gateway to expose S3 as NFS/SMB file shares for applications that expect file semantics.
  • Use Tape Gateway to migrate tape backup workflows to AWS without changing backup software.
Choose the gateway type based on the protocol your applications expect (iSCSI for block/tape, NFS/SMB for file) and whether you require the primary data on‑prem or in the cloud.

Watch Video