Enterprise and On-Premises S3-Compatible Storage Systems

Jun 28, 2026

Behind the cloud services, there is a large market of enterprise storage platforms that expose an S3-compatible endpoint inside the data center. These are the systems that hold petabytes for media companies, banks, research labs, and government, and they all speak S3 so that standard tools can read and write to them. This page lists the enterprise and on-premises S3-compatible platforms and how to connect each one.

This is one category in our complete list of S3-compatible storage providers. All of these connect to Blober through the generic S3-Compatible connector: you point it at the system's S3 endpoint (often a data VIP or load-balanced address inside your network) with your keys. Most on-premises systems prefer path-style addressing. The endpoint notes below come from each vendor's own documentation, cross-checked against current S3 client references^[1]. The endpoint setup notes explain path-style and virtual-hosted addressing.

NetApp StorageGRID

NetApp StorageGRID is a widely deployed enterprise object store with a mature S3 implementation, used for backup targets, archives, and data lakes.

Endpoint format: your StorageGRID gateway or load-balancer address (your own deployment).
Addressing: both styles are supported; path-style is common on-prem.
Notes: StorageGRID supports S3 features like versioning, object lock, and lifecycle, which makes it a strong compliance and ransomware-recovery target^[2].

NetApp ONTAP S3

NetApp ONTAP, the operating system behind FAS and AFF arrays and Cloud Volumes ONTAP, includes an S3 object server.

Endpoint format: the S3 server address you configure on the ONTAP system.
Addressing: path-style is the safe default.
Notes: beginning with ONTAP 9.8, you can enable an ONTAP S3 object storage server in an ONTAP cluster and manage it with System Manager or the ONTAP CLI^[3].

Dell ECS and ObjectScale

Dell ECS (Elastic Cloud Storage) and its containerized successor ObjectScale are enterprise object platforms with a full S3 API.

Endpoint format: your ECS or ObjectScale data node or load-balancer address.
Addressing: both styles; path-style is common on-prem.
Notes: designed for multi-site, geo-distributed deployments with active-active access^[4].

Cloudian HyperStore

Cloudian HyperStore is an S3-native enterprise object store sold as software or as an appliance, known for close S3 API fidelity.

Endpoint format: your HyperStore S3 endpoint (data VIP or load balancer).
Addressing: both styles supported.
Notes: Cloudian markets very high S3 API compatibility, including object lock for immutability^[5].

Pure Storage FlashBlade

Pure Storage FlashBlade is a high-performance, all-flash platform with an S3-compatible object store.

Endpoint format: https://<s3-data-vip> (the FlashBlade S3 data VIP).
Addressing: path-style works everywhere; virtual-hosted needs DNS so that bucket.<endpoint> resolves to the data VIP.
Notes: supports ListObjectsV2, multipart with AWS-compatible ETags, versioning, and advanced checksums on recent Purity releases^[6].

Hitachi Content Platform (HCP)

Hitachi Content Platform is an enterprise object store for long-term retention and compliance, with an S3-compatible API.

Endpoint format: https://<your-hcp-host>.
Addressing: path-style is the safe default.
Notes: HCP supports namespace access through REST, the Hitachi API for Amazon S3, WebDAV, CIFS, and NFS; in the S3-compatible API, namespaces are called buckets^[7].

Spectra Logic BlackPearl

Spectra Logic BlackPearl is an on-premises S3-compatible gateway that tiers data to disk, tape, and public clouds under one namespace.

Endpoint format: https://<your-blackpearl-host>.
Addressing: path-style.
Notes: popular for media archives and backup where tape economics matter^[8].

Scality RING and ARTESCA

Scality RING is a petabyte-scale software-defined object store; ARTESCA is its lighter, cloud-native sibling. Both expose S3.

Endpoint format: your RING or ARTESCA S3 endpoint (load balancer or connector address).
Addressing: both styles supported.
Notes: Scality also maintains the open-source Zenko CloudServer, so the S3 lineage runs deep^[9].

Quantum ActiveScale

Quantum ActiveScale (formerly Western Digital ActiveScale) is an enterprise object system tuned for large-capacity archives and cold data.

Endpoint format: your ActiveScale S3 endpoint (load-balanced address).
Addressing: path-style is common.
Notes: often paired with tape for very long-term retention^[10].

DataCore Swarm (formerly Caringo Swarm)

DataCore Swarm is an object storage platform with an S3-compatible gateway, used for archives and content repositories.

Endpoint format: your Swarm S3 gateway address.
Addressing: path-style is common.
Notes: focuses on long-term data protection and large media libraries^[11].

Nutanix Objects

Nutanix Objects is the S3-compatible object service in the Nutanix hyperconverged platform.

Endpoint format: the Objects store endpoint you configure in Prism.
Addressing: both styles supported.
Notes: integrates object storage into existing Nutanix clusters^[12].

VAST Data

VAST Data is an all-flash, scale-out platform that exposes S3 alongside file protocols.

Endpoint format: your VAST S3 endpoint (virtual IP or load balancer).
Addressing: path-style is the safe default.
Notes: aimed at high-throughput analytics and AI pipelines^[13].

WekaFS (Weka)

Weka is a high-performance data platform with an S3 protocol front end over its parallel file system.

Endpoint format: your Weka S3 endpoint (cluster address).
Addressing: path-style is common.
Notes: built for GPU and HPC workloads that also need object access^[14].

Zadara

Zadara is fully managed, enterprise-grade S3-compatible storage with on-prem, hybrid, and cloud deployment options.

Endpoint format: https://<vsa-id>.zadarazios.com (for example https://vsa-00000001-public-zadara-cloud-01.zadarazios.com).
Addressing: virtual-hosted, with us-east-1 as the default region.
Notes: fetch the endpoint and region from the Zadara Object Storage management interface^[15].

A Note on IBM Cloud Object Storage and MinIO On-Premises

Two systems span the cloud and on-prem worlds. IBM Cloud Object Storage (covered on the cloud providers page) is also sold as on-premises software descended from Cleversafe. MinIO (covered on the self-hosted page) is deployed in many enterprises as a production object tier, not just for testing. Both connect to Blober the same way: point the generic S3-Compatible connector at the endpoint.

Frequently Asked Questions

Can Blober connect to an on-premises storage array? Yes, as long as the array exposes an S3 endpoint your machine can reach. Use the system's data VIP or load-balanced address with your keys, and pick path-style addressing if browsing fails.

Do enterprise systems support S3 object lock and versioning? Many do (StorageGRID, Cloudian, Dell ECS, and others), which is why they are used as immutable backup and compliance targets. Confirm the specific features with your vendor.

Why do these prefer path-style addressing? Virtual-hosted addressing needs DNS so that every bucket.<endpoint> name resolves to the storage. On-prem systems often skip that, so path-style (bucket in the URL path) is the reliable default.

Can I move data from an on-prem array to the cloud? Yes. Blober copies directly between your on-prem S3 endpoint and cloud S3 endpoints that support the normal object operations it uses, or to non-S3 targets like Azure Blob, without staging a full copy on your disk.

Get Blober

Connect Blober to your enterprise S3 endpoint by URL and move data between on-prem storage and the cloud directly, without filling your local disk.

Download Blober at blober.io