Exponential growth of unstructured data and the advent of the distributed enterprise push the boundaries of traditional network attached storage (NAS) arrays. The need for more scalable and less expensive storage capacity that is centralized and globally accessible is driving many enterprises to both on-premises and public cloud-delivered object storage solutions.
However, there are some challenges when it comes to supporting enterprise file data with an object storage infrastructure. One key pain point is the fact that object storage is slower-performing than NAS architectures, and the public cloud-delivered model adds additional latency. This flies in the face of today’s performance hungry mission-critical workload set. Additionally, because object storage architectures use different access protocols than NAS architectures, a timely, expensive and risky application rewrite is typically required for compatibility. Furthermore, unpredictable and often substantial egress fees are often accumulated as data is migrated back on premises from the cloud. Finally, many solutions don’t offer enterprise grade collaboration and data protection features including backup, versioning and file locking.
Global file collaboration solutions have emerged and address a number of these pain points. These solutions typically centralize data storage to object storage, and then use on-premises edge devices as a cache that feeds data to production workloads. Effectively, users work from the faster-performing edge devices, and changes are copied to the object storage repository via snapshots and distributed across the enterprise’s other edge devices. This approach provides access to low-cost and scalable object storage capacity while reducing latency.
When evaluating global file collaboration solutions for compatibility with enterprise-grade requirements, storage managers should bear in mind a few key capabilities, including:
- Compatibility with common file protocols, including NFS and SMB, to enable file data to be stored in object architecture.
- The ability to be agnostic in terms of the underlying object store that is supported, whether an on-premises solution such as Dell EMC’s Elastic Cloud Storage (ECS) appliance or Amazon Web Service’s S3 service. Large enterprises in particular might find it more cost effective to own and operate their own object storage infrastructure.
- Centralized and programmable management via a RESTful application programming interface (API). Notably, the ability to automatically reclassify data to the lowest-cost storage tier according to its usage levels – and to analyze cost and usage metrics – are key to maximizing return on investment (ROI).
- Accelerated communication of data changes between object storage and edge device resources. High-performance local area network (LAN) access, and the ability to process volume-level snapshots in chunks and in parallel, can help.
- Multi-file synchronization and redundant global file lock capabilities to avoid version conflicts and to ensure data availability.
- The ability to apply snapshots and geo-redundancy to further enhance disaster recovery by providing more recent versions of files to restore, while also facilitating business continuity.
Register now to join experts from Storage Switzerland and Nasuni for the on demand webinar discussion of how to overcome the shortcomings of legacy NAS and stand-alone object storage solutions when it comes to serving modern file data needs.