NooBaa: Frictionless Storage for Unstructured Data

NooBaa, Inc. announced the worldwide availability its frictionless storage solution and provided demonstrations at VMworld 2016.

The company allows enterprises with storage capacity challenges to create a secure, high-performance storage solution across any compute resource available, including heterogeneous, cloud, and even shared hosts. Data placement across heterogeneous capacity resources is optimized through a novel control plane architecture using machine learning technologies.

The firm’s Community Edition is free to download and install. It supports unlimited provisioned capacity and a maximum of 20TB of files stored.

Until now, customers deploying scale-out storage faced daunting implementation issues, including high costs, lengthy evaluation and deployment times, vendor lock-in, complex administration features, and configuration rigidity. In contrast, company’s storage software supports unprecedented infrastructure flexibility. Customers can utilize resources that are heterogeneous, shared, widely distributed, and adapt to changing cloud strategies by blending public and private resources. The firm delivers this experience in a software-only solution that deploys in under 15 minutes, reducing the time and effort required for deploying storage. This frictionless experience also applies to ongoing management of the archive, where the ability to take advantage of different hardware technologies eliminates lock-in, technology risk, and administrative cost.

“Unstructured data already consumes most of the capacity in an enterprise datacenter, and the emerging workloads like Splunk, Hadoop, and IOT are accelerating the problem. To reduce the cost and risk of acquiring storage, we really need to break the traditional rigidity and lock-in imposed by traditional systems,” says Yuval Dimnik, co-founder and CEO, NooBaa. “We realized that the way to support any capacity resource located anywhere on the network, was to approach it as an orchestration problem, using big data concepts like machine learning to manage data placement. This is what we’ve delivered.“

“Over the last decade we’ve seen evolutionary developments in things like object storage, but we really haven’t seen anything that substantially changes the customer experience,” says Henry Baltazar, research director, 451 Research. “For customers to reduce their costs in large-scale archival, hardware independence, hybrid flexibility, and reducing administrative tasks over the life of the archive are where the money is going to be saved.“

The software delivers the low-TCO attributes of public-cloud storage in a locally managed service. His architecture reduces capex, opex, setup headaches, and lifecycle costs compared to existing file storage.

Benefits of the software include:

• Storage software to support unlimited resource heterogeneity, including shared host resources

• Storage software using machine learning to optimize for resilience, performance, economics, and locality

• Hybrid architecture supporting any blending of public and private capacity resources

• Non-disruptively accommodates introduction of new hardware, media, and networking technologies; and autonomously takes advantage of new capabilities

• Installation in 15 minutes, in many cases with zero up-front cost.

Key features of the software include:

• Support for any virtual or physical Linux or Windows storage resource

• Machine learning optimization for resilience, performance, and economics

• Enterprise capabilities for global dedupe, compression, encryption, audit, and DR

• Installation in 15 minutes, using VMware ESX to host the NooBaa core control-plane
  Access through an Amazon S3 compatible interface.