Sponsored by SearchStorage.com
The attributes of big data applications are often described as the four Vs: volume, velocity, variety and variability. Taken singly, any one of those traits would pose a serious challenge to traditional storage systems; when combined, they force a rethinking of the very storage architectures we rely on. Processing thousands, or millions, of small files comprising structured or unstructured data from disparate sources would strain a typical NFS/CIFS-based file storage system, but new techniques such as object storage and distributed architectures can cut the task down to size, literally breaking it into more digestible chunks that can be processed in parallel. This buyer’s checklist describes the key capabilities and features a storage system will need to handle big data analytics. The goal is, of course, for the four Vs to add up to a fifth: value. Access >>>
Table of contents
- Storage for big data applications
- Big data storage architectures
- IF-THEN decisions for big data storage
Premium Content for Free.
More Premium Content Accessible For Free
A lesson in flash caching
Solid-state storage is proliferating as a replacement for hard disk drives, where it offers a quick shift into the fast lane of storage processing...
Storage performance management: Ways to maximize your environment
Making your storage perform to the very best of its ability is an age-old problem with a long list of ways to approach it. But how should you start ...
Hypervisor vendors up the ante with new storage features
There's a struggle between doing storage the "old way" and the new demands that virtual servers put on networked storage. Storage vendors have ...