HyperFile: Frequently Asked Questions

CLOUDIAN HYPERFILE

Q: How does HyperFile differ from traditional NAS?

Traditional NAS is built on file systems that have practical limits on capacity. As they near that capacity, their performance can degrade significantly. The usual solution is to add new NAS systems, but this leads to “NAS sprawl,” and an ever-increasing level of complexity that imposes significant penalties in management, money and manpower.

HyperFile grows in a much more elegant way, using the HyperStore architecture to provide a unified, hyper-scalable solution.

Q: How is HyperFile different from HyperStore?

HyperFile is a NAS controller that delivers SMB(CIFS) and NSF file services, employing Cloudian HyperStore object storage as its underlying storage layer. This is analogous to a traditional NAS system where a controller employs SATA drive shelves as its storage layer.

Q: How do file storage and object storage differ?

In file storage, files placed in folders, and nested to form a set path. In this way, files are organized into a hierarchy, with directories and sub-directories. Each file also has a limited set of metadata associated with it, such as the file name, the date it was created, and the date it was last modified.

In object storage, unstructured data is combined with metadata tags and a unique identifier. These “objects” are stored in a flat address space which can grow without limit.

Q: Why is object storage less costly than traditional storage?

Object storage is less costly for two reasons. One, is that it’s built on industry-standard servers that benefit from enormous economies of scale. They are, virtually by definition, the lowest-cost hardware platform. Second, object storage employs the simplest possible architecture: a shared-nothing, peer-to-peer design. In this layout, all nodes (or servers) are the same. The same hardware and software are replicated and joined together to form a cluster. There is only one body of software to maintain, and one just underlying hardware platform.

HyperFile leverages the economy and scalability of object storage to create less costly, more easily managed file platform.

Q: What makes the system reliable?

In object storage, data is striped across nodes. Data protection is user defined and can withstand the failure of multiple nodes or even an entire data center, depending on the configuration.

Q: How will HyperFile benefit businesses the most?

Scalability and cost. HyperFile can grow from terabytes to petabytes without disruption. Capacity costs less than 1 cent per GB/month, or about one-third the cost of traditional enterprise NAS storage.

Applications with either large files or numerous files, in video surveillance, media, and engineering for example, will see the substantial benefit.  Healthcare, legal, and financial industries will gain from the WORM functionality plus the scalability.

Q: Growing a NAS deployment demands time and effort from IT. How hard is it to expand HyperFile?

When you need more space, just add nodes and allow the system to incorporate it into the same multi-protocol namespace. This happens automatically, without the need for intervention from IT. Additional nodes can also provide a higher level of data durability.

Q: Will HyperFile mean a change to the way files are accessed?

No. Like any enterprise NAS, users access files with the usual protocols, SMB(CIFS)/NFS. The system is also POSIX-compliant.

Q: What down-time penalties are imposed by a migration to HyperFile?

None. Integrated data migration tools help you execute non-disruptive data migration from conventional NAS to Cloudian HyperFile. All migration occurs as a background task, allowing continuous file access during the process.

Q: How does HyperFile keep my data secure?

HyperFile includes all the data security features present in enterprise NAS solutions. These include WORM (write once, read many) capability, snapshots, and disk quotas. The solution provides client access control through LDAP and Active Directory user authentication. AES 256-bit encryption provides security for data at rest.

Q: How fast is HyperFile?

There are three common ways to view file performance: throughput (GB/s), IOPS (operations per second), and latency (time to complete a single operation).

HyperFile will scale on throughput and IOPS as nodes are added. On those metrics, the environment can be configured for various workloads.

On latency, performance is determined by the workload.  The HyperFile NAS controller includes on-board cache, up to 24TB. Data from cache will be served quickly, with latencies equivalent to traditional NAS devices. Data being retrieved from HyperStore will incur longer latencies.

Q: What are the ideal use cases for HyperFile?

HyperFile best suited for large files in applications that are not sensitive to latency. Examples would be media files, documents, images, and engineering files. Also remote access, such as storage-as-a-service, is ideal. The environment is not intended for database or transaction processing applications.