OBJECT STORAGE

WHY OBJECT STORAGE?

Object storage is still a relatively new alternative to file or block storage. So — what is it and why should you consider making the switch if you haven’t already?

Object storage casts off the size constraints baked into file-based storage approaches. There is no clunky addressing hierarchy to cap growth and flexibility in classifying data — which makes object storage ideal for managing large amounts of unstructured data. There is simply no better way to safeguard and access large amounts of data — especially video, imaging data, health records and big data output. Users can store raw data before and after analysis, and files can be updated with flexible meta-data tags for easy sorting in ways that defy older file systems.


Scalability
Object storage is not limited by the scalability constraints of block or file storage. File storage can ramp up tremendously — but there is a top level cap on the number of files that you can realistically store. Many of our customers are seeing 40-60% growth in unstructured data storage per year — so these limits are of practical concern to more and more companies.


Raw performance
First generation object storage solutions had some performance shortcomings, but that is no longer true. High sequential throughput performance provides an ideal solution for streaming even the most massive files.


Customizable metadata
File system metadata is pretty well hamstrung and inflexible. Date of creation, time of most recent update and ownership information tracks readily — but there is no way to create and update custom metadata. Object storage provides a ready way to classify and sort a virtually unlimited amount of metadata. For example, media files such as images or videos could be located and tracked by overall subject, major visual elements, photographer or videographer, and much more.


Scale-out architecture
RoWork’s Object Storage solution offers an excellent platform for starting small and building out. Storage capacity can easily scale to meet the demands of your organization, making it very easy to manage from the start.


Parallel streaming
Object Storage allows your users and locations to stream large files in parallel over whatever data pipes you have available — maximizing the performance of your bandwidth.


Superior data protection
The industry standard for preventing data loss has been to go with RAID groups of storage drives — sometimes with the added benefit of data replication. The downside to this is that users are locked in so that different data types and sensitivities all have to use the same data protection approach. Object Storage sidesteps that limitation by providing a more flexible approach called erasure coding. Multiple storage drives and data nodes are used to ensure protection. Erasure coding and flexible replication work in tandem to provide a more reliable and customizable approach based on your needs.


Support for industry standards
RoWork Data Centers’ object storage solution uses Amazon’s Simple Storage Service (S3) interface — the industry standard for object storage — but with significant cost and support advantages over Amazon’s product offerings.


Lower costs
Choosing a data storage strategy has always started with cost analysis. RoWork Data Centers offers unbeatable value both in direct costs per byte of data stored — but also in the ease and cost of management, maintenance, and expansion going forward. The scalability of the solution eliminates the need to carry unneeded excess capacity. You can start with what you need and build out. Moreover, you have no ceiling on the amount of data you can store in the same namespace, excellent data protection, and flexible policy tiering — all of which ensure that managing all of your data is more efficient and cost-effective.