Amazon EBS
AWS's block storage volumes for EC2, with SSD and HDD volume types and snapshots.
Best for: Block volumes for AWS instances.
Visit Amazon EBS →Low-latency volumes for databases and virtual machines: managed cloud block storage, enterprise block arrays, software-defined and NVMe-oF block storage, and Kubernetes block storage.
AWS's block storage volumes for EC2, with SSD and HDD volume types and snapshots.
Best for: Block volumes for AWS instances.
Visit Amazon EBS →Managed block storage for Azure virtual machines, with several performance tiers.
Best for: Block volumes for Azure VMs.
Visit Azure managed disks →Google Cloud's block storage for Compute Engine, including the configurable Hyperdisk family.
Best for: Block volumes on Google Cloud.
Visit Google Persistent Disk and Hyperdisk →Oracle Cloud's block storage with adjustable performance.
Best for: Oracle databases on OCI.
Visit Oracle Block Volumes →IBM's block storage for virtual servers on IBM Cloud.
Best for: IBM Cloud virtual servers.
Visit IBM Cloud Block Storage →Block storage disks for Alibaba Cloud ECS instances.
Best for: Alibaba Cloud workloads.
Visit Alibaba Cloud disks →Block storage volumes for DigitalOcean Droplets.
Best for: Simple volumes for small deployments.
Visit DigitalOcean Volumes →NVMe and HDD block storage volumes for Vultr instances.
Best for: Low-cost cloud volumes.
Visit Vultr Block Storage →Dell's all-flash, NVMe-based storage appliance family for block and file workloads.
Best for: Mid-range to enterprise block storage.
Visit Dell PowerStore →Dell's high-end NVMe storage arrays for mission-critical block workloads.
Best for: Mission-critical databases.
Visit Dell PowerMax →All-flash block storage arrays from Everpure (formerly Pure Storage).
Best for: Simple all-flash block storage.
Visit Everpure FlashArray →IBM's all-flash storage systems with block storage and virtualisation.
Best for: IBM and storage virtualisation.
Visit IBM FlashSystem →Hitachi Vantara's family of block storage platforms.
Best for: Always-on enterprise applications.
Visit Hitachi Virtual Storage Platform →HPE's storage portfolio, including Alletra block arrays.
Best for: HPE environments.
Visit HPE storage (Alletra) →Dell's software-defined, scale-out block storage that can run alongside compute.
Best for: Scale-out software-defined block storage.
Visit Dell PowerFlex →Software-defined block storage that delivers NVMe over TCP volumes.
Best for: NVMe/TCP block storage.
Visit Lightbits →Software-defined block storage for clouds and virtualisation, running on standard servers.
Best for: Cloud providers and private clouds.
Visit StorPool →Open-source replicated block storage for clusters.
Best for: Replicated block storage.
Visit LINBIT (LINSTOR and DRBD) →Ceph's RADOS Block Device provides block volumes from a Ceph cluster.
Best for: Block volumes for OpenStack and Kubernetes.
Visit Ceph (RBD) →Software-defined block storage built on NVMe over fabrics for Kubernetes and databases.
Best for: NVMe-oF block storage for Kubernetes.
Visit simplyblock →Software-defined block storage with NVMe/TCP and Kubernetes integration.
Best for: NVMe/TCP block storage.
Visit Blockbridge →A distributed block storage system for Kubernetes, hosted by the CNCF.
Best for: Lightweight Kubernetes volumes.
Visit Longhorn →Open-source container-attached storage for Kubernetes.
Best for: Local and replicated volumes.
Visit OpenEBS →A data-services platform for Kubernetes from Everpure (formerly Pure Storage).
Best for: Enterprise Kubernetes storage.
Visit Portworx →Elastic Volume Service: block disks for Huawei Cloud servers.
Best for: Block storage on Huawei Cloud.
Visit Huawei Cloud EVS →NetApp ONTAP software in the public cloud, with block (iSCSI) and file access.
Best for: ONTAP features in AWS, Azure and Google Cloud.
Visit NetApp Cloud Volumes ONTAP →Everpure's block storage software for public clouds, bringing array features to cloud volumes.
Best for: Array-style data services in the cloud.
Visit Everpure Cloud Block Store →Match your requirement to a starting point, then compare vendors.
| Disks for cloud VMs | The cloud provider's own block storage, such as EBS, managed disks or Hyperdisk |
|---|---|
| Mission-critical databases on premises | An enterprise all-flash array over Fibre Channel or NVMe-oF |
| Scale-out block on commodity servers | PowerFlex, Lightbits or StorPool |
| Open-source block storage | Ceph RBD, LINSTOR or Longhorn |
| Stateful Kubernetes workloads | A CSI-based option such as Longhorn, OpenEBS or Portworx |
| Shared files or objects | File or object storage instead |
Neighbouring directories and hubs.
Quick answers.
Block storage presents raw volumes that a server formats and uses like a local disk. It offers low latency and is the standard choice for databases and virtual machine disks.
Block storage is raw volumes, file storage is shared folders, and object storage is objects accessed through APIs. Block gives the lowest latency but is usually attached to one host at a time.
NVMe over Fabrics carries NVMe commands over a network, such as TCP, RDMA or Fibre Channel, so remote volumes can approach the latency of local flash.
Look at IOPS and throughput limits, volume size limits, durability, snapshot features and price per provisioned or used capacity. Match volume types to the workload.
Yes. Storage systems expose volumes to Kubernetes through the Container Storage Interface (CSI). Options range from cloud disks to software such as Longhorn, OpenEBS and Portworx.
Descriptions summarise each vendor's or project's own positioning, checked on 5 October 2026. Product names and offerings change often, so confirm details on the linked site before deciding.