The Hammerspace Technology

To deliver Data-as-a-Service, Hammerspace ties together data virtualization, hybrid cloud data management, and Metadata-as-a-Service into a single platform. There are critical pieces of technology necessary to make this work together.

Data virtualization with a Hybrid Cloud Data Control Plane

By managing metadata separately from data, it becomes possible to make unstructured data appear virtually anywhere without copying it.  This data virtualization is key to overcoming the challenge of storage silos, making data appear present across the hybrid cloud through an active-active geo-spanning namespace.

Hammerspace manages the namespace by pooling together heterogeneous storage resources and replicating the full set of metadata to any connected cloud or data center, creating a Hybrid Cloud Data Control Plane.

As workloads access virtualized data, the machine learning engine learns patterns to predictably fetch only the data necessary to support the running job; while live data-instance mobility non-disruptively moves files and objects to where they need to be, even during read/write.

Machine Learning

Global access to billions of files across the hybrid cloud demands a unique approach to data orchestration – a continuous market economy simulation between real data and available infrastructure resources run by machine learning. The model treats storage services as landlords with resources to lease, and data files as tenants who spend limited currency to meet specific needs.

Hammerspace continuously collects performance telemetry from workloads for each file accessed, in the form of metadata. This monitoring provides a rich understanding of how the infrastructure is performing, so Hammerspace can automatically correct for issues before they happen. Real-time decisions for data placement are fully automated by machine learning, balancing performance and cost across the hybrid cloud.

Live Data-Instance Mobility

Live data-instance mobility delivers real-time data availability, defined by policy and automated with Machine Learning. Workloads can run across clouds and data is made available automatically without disruption or downtime.

Hammerspace DSX Data Services support NFS, SMB, and S3; deploying advanced data layout techniques seamlessly moving the instances of data in the namespace, even during active read/write.  WAN optimization keeps things efficient with automatic global deduplication and compression, while data is encrypted end-to-end using military grade and government approved algorithms.


A global service that scales metadata beyond regular file system standards (POSIX, NFS, and SMB) including telemetry for performance and access (IOPS, throughput, latency), user-defined and analytics harvested metadata.

User-defined metadata allows for rapid prototyping of metadata (keywords and tags) as well as pre-declared entries (labels and attributes). Pre-declared entries can define taxonomies of metadata. For example, a tiger in a picture can automatically be part of the animal family when looking for content in image files.

User-defined and harvested metadata can be inherited from directories or managed with per-file granularity. Automatically generated harvested metadata by using cloud services to populate the metadata, easily integrating cloud services for common needs such as data classification or content identification to immediately apply to workloads, even those that don’t run in the cloud.

Hammerspace namespace delivers integration into the metadata by allowing users to view, filter and search metadata in-place while navigating the namespace. Rather than relying on filenames to identify data, user-defined metadata rapidly, accurately, and efficiently enables users to find the data they need.


Hammerspace virtualizes and abstracts data from the infrastructure using machine learning, creating the agility required to provide File Data-as-a-Service to any application or data service.