Modern Active Archives Leverage Tape-based Object Storage Controlled by Intelligent Software
Data volumes are growing beyond file systems, as they can only manage a limited number of files. Due to the uncontrolled growth of unstructured data, object and cloud-based storage technology continue to gain in popularity. This data is generated ever faster – but most of it is rarely or never read after a short processing phase. Nevertheless, the data must be stored for the long term, either because of legal requirements or because it may be relevant again at a later point in time.
This poses enormous challenges for storage infrastructures and active archives. Object-based storage systems meet the requirements imposed by data volumes and retention periods. In the on-premises sector, these are mostly hard disk-based storage systems. Newer systems now also use solid-state disks (SSD). However, when it comes to archiving data volumes in the high PB range, these systems reach their limits. One aspect is the economic factor, both in terms of the hardware acquisition costs and the costs for its operation. Moreover, in the HDD sector, capacity increases are only possible to a limited extent. The predicted data growth far exceeds the technically possible increase in HDD capacities.
So, to meet the challenges and bridge this gap, tape-based object storage is becoming more and more popular. To store enormous amounts of data, tape offers unbeatable cost-efficiency. In addition, tapes can provide virtually unlimited capacity. Their use is very energy efficient: for the simple reason that tape media that is not being accessed does not consume power. Last, but not least, the offline capability of the storage media and WORM protection are convincing in the face of increasing threats from cybercrime.
For tape technology to be seamlessly integrated into an existing storage infrastructure as an S3 storage class, it must be integrated via the standardized S3 REST API. Then all the advantages of tape can be used without restriction in the object storage area – and in particular for use as an Active Archive for object data. In the meantime, there are various software-defined storage solutions with which tape libraries can be used as object storage and as target storage for object data backup and archiving.
Ultimately, the aim is to create a uniform, object-based storage system that enables different storage technologies to be combined and used according to their respective advantages. The larger the data volumes, the more the need for active archives based on object storage technology with standardized interfaces that can be homogeneously integrated into existing infrastructures.
So when it comes to storing the large amounts of object data in the most optimized way possible in an Active Archive, such a system will have to use combinations of SSD, HDD and tape media supported by intelligent software-based tiering functions.
This means that a unified object storage system with a standardized S3 interface that transparently combines multiple storage classes (e.g. SSD, HDD, Tape) provides all capabilities an active archive requests. The decisive factor is the controlling software which uses intelligent tiering algorithms for moving objects between the storage classes. The objective of such a system is “the right data in the right place at the right time”.