The Problem with NFS to Object Storage Gateways

NFS compatibility is considered table stakes for object storage products in today’s market. How and why did this happen? Network Attached Storage (NAS) and the NFS protocol have been and is the architecture of choice for unstructured data. Despite all the benefits of object storage, the data is on NFS. All roads to object storage have to pass through NFS. The challenge is how smooth is that road. IT professionals should know that NFS support is more than just a checkbox item. How the object storage vendor implements that support is critical to a successful object storage implementation.

NFS is ubiquitous. Every laptop, workstation, and server has an IP connection and every modern operating system has a built-in NFS client. Using a NAS system is as simple as configuring a share and mounting that share.

Object storage systems, on the other hand, typically speak the S3 API. Although object storage has been out for quite some time and is the predominant way that cloud providers like Amazon Web services store their data, it’s nowhere near as ubiquitous as NFS, especially in the private data center. All applications know how to write to a file system, and NFS appears as a file system. Many, if not most, applications do not yet know how to write to an S3 device – although this is starting to change.

The reason why object storage systems are not a product waiting for a problem is that NAS systems do have a whole host of problems that object storage systems solve, such as difficulty scaling, an inability to find data stored a long time ago, and significant costs from backing up data that the enterprise does not use anymore. Object storage systems can solve all of these problems, but that doesn’t change the fact that the applications creating all of this unstructured data didn’t yet know how to write to the S3 API.

Enter NFS gateways. All major storage vendors started developing a gateway product that would accept files via NFS and eventually store them as S3 objects. These products started the process of bridging the gap between legacy applications and object storage systems. Unfortunately, many implementations of these products have a number of challenges, especially when it comes to scalability.

Join us for our on demand webinar, “What Your Object Storage Vendor Isn’t Telling You About NFS Support”. We discuss what object storage is, why it is better at solving modern unstructured data challenges than NFS, and why IT professionals need to pay careful attention to how their object storage vendors implement NFS compatibility.

Watch On Demand

W. Curtis Preston (aka Mr. Backup) is an expert in backup & recovery systems; a space he has been working in since 1993. He has written three books on the subject, Backup & Recovery, Using SANs and NAS, and Unix Backup & Recovery. Mr. Preston is a writer and has spoken at hundreds of seminars and conferences around the world. Preston’s mission is to arm today’s IT managers with truly unbiased information about today’s storage industry and its products.

Tagged with: , , , , ,
Posted in Blog
One comment on “The Problem with NFS to Object Storage Gateways
  1. Tim Wessels says:

    Well, the use of NFS with an an object store should be defined in terms of its functionality. If your applications need real-time, file locking, global namespace access to data in your object store using NFS or SMB protocols then look at solutions from Panzura and HyperStore Connect for Files from Cloudian.

    If you want to use your object store as an NFS mount point for ingesting or retrieving data, you can do that with most object-based storage software, but you will still have to deal with protocol conversions, which tends to slow things down. SwiftStack includes its own NFS gateway with its OpenStack Swift software.

    If your applications use SMB and NFS shares on Windows and Linux servers, you may have fallen prey to too much old and seldom accessed data residing on your primary DAS, SAN and NAS storage servers. In this situation, Caringo FileFly and Komprise can use policies or plans to move old data from these servers to an object store, which will allow the data to be retrieved if the application user ever needs it again.

    NFS and SMB will be with us until the applications that use them are replaced with modern applications that use a RESTful API, like S3, to communicate with an object store. Until that day arrives, there are different ways to integrate the use of NFS with an object store based on the type of functionality you require.

Comments are closed.

Enter your email address to follow this blog and receive notifications of new posts by email.

Join 21,851 other followers

Blog Stats
  • 1,164,727 views
%d bloggers like this: