sakkmesterke - Fotolia
In a clear pitch at current trends in hybrid- and multi-cloud storage, the product will allow customers to access unstructured data in NAS storage anywhere – on-premises or in the cloud.
Key use cases targeted are those where users need access from a wide variety of locations, in particular where they may need to access disparate data for machine learning, AI, the internet of things (IoT) and analytics workloads, for example.
Hammerspace appears, in function at least, a little like the file virtualisation products of around a decade ago, but very clearly built for the scale, landscape and use cases of the cloud era.
The product virtualises data in any cloud or on-premises storage by separating data payload from its metadata and using that metadata to represent the file in a global namespace.
According to executive chairman Dave Flynn: “Hammerspace is data centric, the emphasis is not on storage. It looks like all you want to be there, is there. It appears there, but is not there yet”, with the data payload moved to support workloads based on the file’s activity profile.
Hammerspace will be deployed in the cloud as a service, and on customer premises as a software-based appliance.
The product will “make traditional file systems truly cloud-native”, according to Flynn.
Customers will have the option of initially being able to deploy Hammerspace, “to sit as an observer of existing data estate; as a non-disruptive on-ramp,” said Flynn. “But the true power comes when it is adopted as part of the Hammerspace.”
In version 1, Hammerspace will virtualise files only, although according to Flynn the technology is already object storage-capable.
Hammerspace gets its name from comic and animation fans’ name for the storage space where characters mysteriously find weapons, equipment, for example, and in particular a female Japanese manga character who unfailingly produced a big hammer at opportune moments.
Read more about multi-cloud storage
- We look at how to build a multi-cloud storage strategy and benefits such as performance, availability and features, as well as potential limitations such as data mobility.
- While IT providers are fond of suggesting moving workloads between clouds is as easy as dragging and dropping apps between environments, the reality can be far more complex.