Page 5
OS + Application
Protected Server/VM
1. Change Clock Tracking (forever)
11:00am Full Backup
11:15am Full Backup
11:30am Full Backup
1 0 1 0 1 1 1 0 1 0 0 1 0 0 1 1 1 1 0 1 0 1
1 0 0
1 1 0
0 1 0
2. Deduplication
3. Inline
Compression
BLOCK-LEVEL DEDUPLICATION & INLINE COMPRESSION SAVINGS
ReadyRECOVER employs a unique technique to deliver storage efficiency that saves space on disk and reduces bandwidth
consumption when replicating backups offsite. These efficiencies are delivered by three complementary techniques:
1.
Change Tracking (Forever):
Only newly changed blocks ever need to be sent from the source system (backup client)
to the ReadydATA onsite or the ReadydATA offsite
2.
Block-level Deduplication:
when each full backup image is written, common blocks from previous backups are not
stored twice. Because only unique blocks are written, many independent full backup images can be stored, while mini-
mizing capacity consumption.
3.
Inline Compression:
As data is written to ReadydATA, it is compressed and checksummed in real-time.
deduplication of backup data requires end-to-end data integrity at a block level because many backups may reference the
same block. in ReadyRECOVER, inline checksumming is implemented to guarantee every block is correct when restoring. if
a block is corrupt, it is silently repaired using checksums and object-based RAid technology built into ReadydATA.
•
Reliable change tracking:
The storageCraft backup agent (called shadowProtect), can reliably track incremental
sector changes on servers and PCs. storageCraft started as a storage driver vendor in 2004. it has since created
backup software with deep Vss integration and sector level change tracking using a microsoft-certified driver in the
windows i/O stack. Every piece of data written (saved) on a PC or server must pass through the storageCraft driver,
allowing it to track every change.
•
Data integrity forever:
ReadydATA employs a next-generation file system that checksums all data inline. when new
backup data is written, it is checksummed at the block level. when that data is read (for recovery), the checksum is
verified. if the checksum indicates an issue, ReadydATA uses self-healing RAid technology (object based RAid) to find a
copy of the same data elsewhere on disk and repair the broken copy.
• All storage devices that use RAid keep redundant data (extra copies) on disk. however, most of these devices do not
checksum data and are unable to tell if the data being read back is valid. Additionally, they are unable to heal corrupted data.