The Future of Active Archive – Node-based Solutions for Access, Scalability and Protection
In my near 30 years working with storage and archive solutions, capacities have grown exponentially for many organizations, from Terabytes to Petabytes and in some cases Exabytes. In the past, the only method to work with large capacities was to add more servers, whether this was the application server itself, a file server or NAS.
One of today’s major storage dilemmas is finding and managing data on dispersed servers and making use of the data stored there. Modern highly scalable primary storage solutions are replacing traditional NAS. Scale-up, scale-out node-based architectures allow organizations to increase both performance and capacity – yet maintain a single “pane of glass” view of all data residing there.
Archive solutions also mirrored this, with archive gateway solutions based on open operating systems and NAS architectures, each with their own file structures. In the near future, highly scalable active archive solutions will also use node-based architectures, where the single pane of glass is not only available through a file system interface (SMB or NFS), but also for remote access through cloud interfaces – such as S3.
The technologies used in an active archive environment can vary, depending on frequency and required speed of retrieval. Disk or cloud-based active archives tend to be the most expensive but fastest for small data retrieve activities. Tape or optical are lower cost but slower for small data retrieves – but possibly faster for large data sets.
Using node-based architectures has a number of advantages – besides simplified access. The concept, when applied to libraries, allows any node to address any request to any media using any drive.
A series of nodes can then support a single massive library (or multiple smaller ones) with tens or hundreds of thousands of slots and hundreds of tape or optical drives, without partitioning. Any job can use a single drive or span across multiple drives for enhanced job performance. Redundancy is in-built, if a node or drive were to fail, others will automatically take over, as today’s primary storage solutions do. Mirrored or replicated writes within the library, to a second library or to cloud (or even to tape in the cloud) happen automatically for data resilience.
QStar is currently working on making this concept a reality with our next release of Archive Manager – Global ArchiveSpace™. The product will run on standard Windows or Linux servers and allow organizations to add nodes as they see fit to improve performance and capacity as they grow. Access to the active archive is through standard interfaces such as SMB, NFS and S3.
Finally, as tape is a sequential storage solution, it also resists ransomware attacks that try to encrypt data at rest. Rolling back the file system to a previous state allows all data to be seen as it was before the attack.
QStar Archive Manager is currently trusted by major movie studios, museums, universities, courts and government bodies around the world to preserve critical content for as long as it is required.
*Christopher Bullock’s The Cobbler of Preston in 1716