"The right tool for the right job" holds just as true in the storage space as any other space. In fact, that is the main reason storage companies have products that focus in on various tiers of storage from tier-1 OLTP (Online Transaction Processing) down into the archival or tier-"3/4" spaces. It is at the lower end of these tiers, however, that companies can fail to see the value of using the right tier of storage for their archival requirements. In fact, I have worked with many organizations that attempt to use very inexpensive JBOD (Just a Bunch of Disks) for their archival data needs.
When you really look into the data that your organization is archiving, there is usually a good reason for the business to retain that data, otherwise why would it be archived at all? Placing that data on a system designed and built from the ground up to be an archival repository is probably a really good idea. However many problems exist when storing massive amounts of archival data on JBOD storage systems including the following:
- Scalability. JBOD is typically limited to a single box, or a shelf of drives, that cannot be expanded outside the enclosure. Once full, the system will require fork-lift upgrades, and painful data migration processes to newer systems or larger drives. Further, the number of useable TBs in a single footprint is very small and it is very difficult to forecast how large your archival needs will grow but you can almost bet they will grow in the multiple TBs. Maybe the largest problem is that deduplication and compression features do not exist in JBOD systems which control data growth and storage utilization. Without these features, it forces companies to purchase even more storage capacity.
- Reliability. JBODs use aging RAID5/6 protection schemas to maintain data integrity. Raid 5/6 is becoming more impractical as disk drive sizes grow and rebuild times to recover failed disk drives increase. Further, JBODs provide no or limited self-healing capabilities to detect errors that can occur in disk drives over time and automatically correct them.
- Recoverability. Low disk reliability ratings may force companies to regularly backup archival data stored on JBODS to tape as a protection mechanism. While companies could theoretically replicate the data to another site, that is not always an option since there is usually no replication features supported on JBOD storage.
- Availability. JBODs often contain single points of failure in the hardware since they are typically single controller configurations, or at best active/passive. This lack of an immediate failover to alternate system(s) or internal controller pairs can cause host disconnection and potential data loss during writes.
- Manageability. Growing beyond one JBOD box will require end-users to manage multiple systems thru multiple separate interfaces that increases the amount of effort, time and ultimately cost required to maintain each unique system. Further, since backup of the archival data is required, the backup window is likely to increase since most archival systems are dealing with millions of files, which take an eternity to backup Additionally if the JBOD system is SAN connected, there would be even more management tasks required, including Zoning, LUN allocation, LUN provisioning, and LUN Masking etc.
- Compliance, Regulatory Validation and Encryption. JBOD contains no native WORM (Write Once Read Many) abilities at all so there are no reasonable way to lock down your files to enable your company to comply with regulatory requirements. For instance, JBOD will offer no retention features to speak of to lock down your data, via file type, users or groups in an LDAP directory. If data encryption is required to meet an internal governance, or external regulation, a 3rd party product would need to be selected and procured in order to facilitate encryption in a JBOD archive.
With all the failings of JBOD storage when used as an archive, it is obvious that archival systems such as the Permabit Enterprise Archive, are reflective of the type of robust architecture that enterprise companies should expect for their archival needs (Based on the rich feature-sets like deduplication, continuous online storage expansion, no fork lift upgrades etc). Permabit puts all of JBOD's failings aside by scaling into the PB range, providing compression, deduplication, WORM and encryption natively, using a storage grid architecture that provides resiliency and high performance and offering replication to eliminate the need for backups.
Identifying the right tool for the right job is a prerequisite for any task and it is no different with archival data. While on the surface JBOD may seem like the right answer for archival data because JBOD appears inexpensive, its hidden costs quickly become apparent. Products like Permabit Enterprise Archive take those hidden costs out of the archival equation while providing companies with a platform that meets the needs of their archival data now and well into the future.