All Versions of the True incremental Backup System use a disk cache. The disk cache is used as a staging area to synthetically consolidate backups and expedite transfer to the tape library(s) if deployed. Server side disk caching allows backups to run in parallel. The disconnected operations between client backups and the subsequent synthetic consolidation, permits the backup system to run significantly more parallel True incremental network backup operations. TiBS is designed to minimize the impact of backup operations on both networks and client systems. Learn more about TiBS enterprise disk storage solutions...
TiBS Automated Tape Library Interface software module is compatible with virtually all leading tape library manufacturers, robotic libraries, and tape technologies. Teradactyl maintains a broad base of large installation clients using robotic tape libraries in addition to disk storage sub-systems. TiBS supports these tape libraries and tape library manufacturers...
The True incremental Backup System provides enterprise backup solutions for a broad range of heterogeneous systems. Our patented multi-level and multi-volume synthetic backup consolidation technology can even be applied to mass storage running network file systems such as OpenAFS and NFS. Learn about TiBS supported operating and network file systems...
Teradactyl uses a significant variety of backup terms which may be unfamiliar to you or used differently by less knowledgeable vendors. One of the most obvious errors is the common misuse of Incremental and Differential terms. We went to the National Institute of Standards and Technology (NIST) and other reputable sources to confirm the accuracy in our use of the terminology. Teradactyl has compiled a list of backup definitions for reference by anyone interested or involved in the backup storage market. Learn more about industry backup terms and definitions...
The True incremental Backup System Automated Tape Library Interface can manage multiple media types in the same robotic library. TiBS can provide a migration path between tape technologies. Tape pools can also be established to put certain types of backup volumes or clients on a given type of tape media or technology.
The Full Version, Lite Version, and Basic Version of the True incremental Backup System all have optional support for robotic tape library systems. Both the Lite and Full versions can support multiple robotic tape libraries. This provides facilities with the ability to utilize older hardware and tape technologies even as they migrate to modern high speed robots.
The Spring 2009 Storage Magazine, SearchStorage Purchasing Intentions published that in the final quarter of 2008, a survey conducted by Fleishman-Hillard Research concluded that mid to large size organizations are now returning to tape from disk only enterprise backup systems. Approximately 60% of the 200 sites surveyed will return tape back into their storage sub-systems. With tape capacity currently at 800 GB uncompressed and native speeds of 240 MBps, Ultrium LTO 4 technology has provided a cost effective and green solution for retention of backup data from mass storage systems.
Teradactyl has continued to advocate the use of Tape in addition to disk in enterprise network backup systems. The combination of technology can provide the ability to balance cost, energy efficiency, offsite requirements, and data retrieval times. Clients can maintain select systems, backup levels, or all backup volumes on both disk and tape to mitigate the risks or exercise the benefits of either technology. Learn more about data storage options with TiBS...
The True incremental Backup System (TiBS) Automated Tape Library Interface (ATLI) can be configured to store True incremental, Partial Cumulative Incremental, Cumulative Incremental, Differential, and Full Backup Volumes on tape in addition to, or instead of, disk. With TiBS, client systems by default begin by transferring data to backup server disk caches. The ATLI enables the backup system administrators to maintain all or part of the backup system data on virtually any robotic tape library. ATLI storage is licensed at a fixed cost per backup server regardless of capacity, number of slots, or the number of tape drives. Heterogeneous enterprise backup to disk can be achieved at far lower cost with Teradactyl and the True incremental Backup System.
The True incremental Backup System (TIBS) requires a disk cache for several purposes. The disk cache is used as a staging area to consolidate backups and expedite transfer to the tape library(s). Data is streamed from the backup cache to tape resulting in faster write times and increased tape compression rates. This approach allows a greater amount of data to be stored on each tape, decreasing tape costs. Server side disk caching allows backups to run in parallel to the backup server.
Network backup windows are compressed with only True incremental backup changes needing to be transferred across the network. Changes are transferred to servers not network tape devices eliminating the bottleneck in processing. Restores from tape devices can run even while the backup servers are ingesting data.
The disconnected operations between client backups and the subsequent synthetic consolidation and tape writes, permits the backup administrator to run significantly more parallel backup operations than the number of drives in the robotic library. With adequate space in the backup server disk cache or the use of the TiBS Disk Library Interface (DLI), network backups can continue even if there are hardware problems with a tape library or drive.
Reliability of both disk and tape for network backup systems has been the subject of heated debate. Teradactyl believes the combined use of both technologies permits users to leverage the benefits of each approach while mitigating the risks associated with either one. Veterans of the industry will generally agree that there have been significant improvements in reliability of tape devices, robotics, and media. Use of tape in enterprise network backup systems have also seen significant improvements in performance metrics including read/write speeds, tape load times, network capacity, and storage capacity. The use of disk caching by Teradactyl and other leading backup software vendors has improved tape compression and enhanced tape drive availability.
Data integrity issues in distributed processing environments continue to exist, not only in backup systems, but across the networks and primary storage in general. Many backup approaches will still stream data from primary storage directly to tape. Today this approach is commonly used for NAS and SAN filer backups.
TiBS uses a file checksum analysis to track data as it moves through the backup system to ensure transfers are correct. This process for each file begins with the computation of a checksum for the data as it is backed up from the target system. The backup server uses this checksum to verify correct network data transfer. During any backup, checksum data is stored within the backup volume to verify data on subsequent use. In rare instances, we have used this technique to diagnose and fix subtle network hardware failures where a specific block of data was corrupted during network transfer. Each time the data is read by the backup server, it recalculates and compares the checksum. This includes reuse of data as part of the server's backup consolidation process, restore processing, and optional tape and disk library verification processing. Our checksum analysis detects hardware failures such as corrupted disk cache volumes, bad tapes, and failing tape devices. In many instances, our built in redundancy allows the failed backup volume to be rebuilt from other backup volumes on the server. This ongoing verification ensures that your backup hardware is performing properly and can detect problems in advance of an actual restore request. During a restore, the data is again verified for correct network transfer to the restore client.
TiBS requires bar-coded tapes and barcode readers for all tape libraries attached to TiBS backup servers. Our efficient Scalable File Index stores the location of every file on every tape ever generated by TiBS on the backup server. We currently support billions of files. With TiBS there is no need to read the wrong tapes to find your data.
For increased data redundancy or for faster offsite capabilities, tape media pools can be configured for mirroring. When tape mirroring is enabled, each backup volume is written to a pair of tapes simultaneously. TiBS does not need to copy data from one set of tapes to another for offsite requirements. For additional data integrity assurance, mirrored tapes can be verified by the TiBS backup server before being sent offsite. A duplicate copy remains onsite for ongoing synthetic backup processing and for restore.
TiBS also provides efficient mechanisms for the recycling of tapes and barcode information. Many institutions we support began with a desired level backup strategy, new tape library, and small mountain of blank tapes. Over time they have decided to change level backup strategies, experienced hardware and media failures, and sometimes even run out of tapes before the next shipment arrives. These are all serious issues organizations should take into account. Not only do you want the ability to barcode tapes, you also need to have flexibility to change your level backup approach, or to remove tapes from the system due to bad media, damage by failing tape drives, or potentially even a lost cartridge. You may also wish to break mirrors to free tapes to continue backup until the next shipment arrives or to recycle one half of the mirror after a given tape retention period. TiBS provides you with flexibility to manage and not just track your media.
The maintenance and cleaning of tape drives in robotic tape libraries is critical for data integrity. TiBS includes support for automated tape cleaning and tape library management. TiBS detects tape drive cleaning status, automatically loads and unloads tape cleaners, and tracks tape cleaner usage.
TiBS can automatically place tapes that are scheduled for offsite into import/export slots and request new blank tapes as needed. TiBS can provide concise reports for which cartridges are needed from off-site locations and which cartridges should be released to off-site storage. Policy changes such as deciding to break mirrors at a given backup level after a set period of time are immediately translated into tape inventory control systems for onsite return requests and subsequent cartridge recycling.
Detailed search functions allow sites to quickly locate specific backup volumes, partitions, directories, sub-directories, and individual files for direct or redirected single pass restores. TiBS can easily restore files from incremental, cumulative incremental, differential, partial cumulative incremental, and full backup volumes. All versions of the files available in the backup system are reported, providing organizations the ability to pinpoint and select specifically the information they need. Parallel processing of restore operations is limited primarily by bandwidth, client ingestion speeds, and backup system hardware. Restore of information can be from multiple media sources including a mix of tape technologies and near line storage.