Object storage is the new disk with infinite pools of low cost, durable, geo-redundant capacity, but how do you use it as a replacement for NAS?
For starters, why replace NAS with object storage in the first place? Good question, and thanks for asking. NAS solutions have been used since the release of Purple Rain, and it has a rightful place in the IT storage landscape. Primarily used for file access to unstructured data, NAS has been a workhorse for IT. So why use object storage instead of your NAS solution?
Object storage is a swiss-army knife. As you can see, it can be used for many use cases. Object storage provides limitless scalability and a cloud-like approach with managing storage. With object storage, integrated data management features are already built-in. It can automatically create non-rewritable, non-erasable data to prevent files from being altered or deleted, set predetermined retention dates for data removal, and easily replicate data between the two sites, with the consistency level you choose, all with a click of a button.
Here at ASG, we prefer to use the Cloudian object storage platform. Cloudian’s global data fabric architecture enables enterprises to store, find and protect object and file data seamlessly across sites, both on-premises and in public clouds, within a single, unified platform. The best part though is it’s a private cloud and is significantly cheaper than using a public cloud services like AWS or Azure.
Now for the NAS part. Object storage requires a translation layer, or NAS gateway for file sharing protocols, and that translation layer creates latency and scalability problems, not to mention global file access challenges inherent in all traditional NAS solutions.
The traditional NAS file systems were designed assuming an underlying disk sector-block structure. As a result, they use RAID to protection against disk and sector failures. This works fine until a disk failure or data corruption event happens, and that is when RAID needs to rebuild the data blocks on the failed drive using parity information. This is a simple method that is proven and has been around for decades, albeit with performance degradation issues, lengthy rebuild times, and storage capacity inefficiencies.
What we need is a file system designed and built that can live in an object store, not a traditional file system relying on an underlying disk sector-block subsystem. We also need a file system that can leverage WORM principals and never over-writes data once it is written keeping data immutable. And wouldn’t it be nice if every file change is time-stamped as its own object to provide complete data protection, eliminating the need for separate file backup and replication tools and processes? Whether the file was deleted by a user, corrupted by a malfunction, or infected by WannaCry, this is the perfect fast ransomware fix.
Finally, we need a file system that integrates with the built-in geo-redundancy of object storage. What if you could restore end user access to files within minutes of a complete data center outage? A file system like this would mean the days of building and maintaining expensive, dedicated disaster recovery sites are finally over!
OK, how can we use object storage as an alternative NAS solution? How can we break the shackles of file system hierarchies and avoid ever having to backup data again? Enter Nasuni UniFS (Nasuni short for “NAS Unified” and UniFS short for “Unified File System”). With Nasuni UniFS, no data is ever overwritten. Both existing and new versions of files are written as WORM, which is why UniFS never experiences file corruption events. Nasuni UniFS is the first cloud-native global file system that starts with object storage, and is not an add-on to a legacy built architecture.
Nasuni provides on-premises caching for high performance file data access using traditional NAS protocols, with all file data automatically propagated to Cloudian. All file access occurs through intelligent caching for LAN-like file access performance. Imagine a NAS solution that can store active and inactive data in one global file system that has no capacity limits and avoids ever having to migrate, backup, or tier file data again! Can I hear a hooyah for Cloudian and Nasuni!
Now you understand why my object storage (Cloudian with Nasuni in this case) is smarter than your NAS. With unlimited scale, no file size limits, unlimited retention for files and versions, no limit on inodes or metadata to support billions of files, global namespace to make data accessible in many locations simultaneously, global file locking, and local caching for access performance, what’s not to like?
I love Cloudian & Nasuni.