EMC DATA DOMAIN DATA INVULNERABILITY ARCHITECTURE
Data Domain works on this architecture to provide high availability of data in backup.In this architecture main focus is to prevent data corruption.There are various algorithm and process to ensure it.The main goal is to protect Data Domain File system from corruption.
The Data Domain operating system (DD OS) is purpose-built for
data invulnerability. There are four critical areas of focus:
• End-to-end verification-In this process data integrity is maintained.End-to-end verification means reading data after it is written and comparing it to what was sent to disk, proving that it is reachable through the file system to disk, and proving the data has not been corrupted.
• Fault avoidance and containment:Maintain Data Integrity.This is achieved by below steps.
1-New data never overwrites good data:
2-Fewer complex data structures:
As long as the system can keep track of the head of the log, new writes will not touch old data. This design simplicity greatly reduces the chances of software errors that could lead to data corruption
3.NVRAM for fast, safe restart
4.Persistent RAM protection
Data in RAM is protected by a persistent RAM Implementation (PRAM).Data Domain system move data from RAM to disk in a process called vaulting.
5.No partial stripe writes:
Data Domain System never update just one block in a stipe.Following no-overrite good data policy,all new writes go to new RAID stripes are written in their entirety.
• Continuous fault detection and healing:
RAID 6: Double disk failure protection, read error correction:
On-the-fly error detection and correction:
On the every read from disk the system first verifies that the block read from disk is the block excepted.It then use the checksum to verify integrity.
Scrub to insure data doesn’t go bad
Re-verify the integrity of all data .Scrubbing process intelligently find and repair defect on the disk before they can become a problem.
File system recoverability
File system recoverability:
To re-construct lost or corrupted file system metadata and also file system check tools that can bring an ailing system safely back online quickly.
Self-describing data format to ensure metadata recoverability:
All data is stored along with metadata that describes it. If a metadata structure is somehow
corrupted, there are two levels of recoverability.
FS check, if needed, is fast:
Data Domain file system never overwrites old data and doesn’t have block maps and
reference counts to rebuild, it only has to verify where the head of the log is to safely
bring the system back online to restore critical data.
====================================================
Data Domain works on this architecture to provide high availability of data in backup.In this architecture main focus is to prevent data corruption.There are various algorithm and process to ensure it.The main goal is to protect Data Domain File system from corruption.
The Data Domain operating system (DD OS) is purpose-built for
data invulnerability. There are four critical areas of focus:
• End-to-end verification-In this process data integrity is maintained.End-to-end verification means reading data after it is written and comparing it to what was sent to disk, proving that it is reachable through the file system to disk, and proving the data has not been corrupted.
• Fault avoidance and containment:Maintain Data Integrity.This is achieved by below steps.
1-New data never overwrites good data:
2-Fewer complex data structures:
As long as the system can keep track of the head of the log, new writes will not touch old data. This design simplicity greatly reduces the chances of software errors that could lead to data corruption
3.NVRAM for fast, safe restart
4.Persistent RAM protection
Data in RAM is protected by a persistent RAM Implementation (PRAM).Data Domain system move data from RAM to disk in a process called vaulting.
5.No partial stripe writes:
Data Domain System never update just one block in a stipe.Following no-overrite good data policy,all new writes go to new RAID stripes are written in their entirety.
• Continuous fault detection and healing:
RAID 6: Double disk failure protection, read error correction:
- Each shelf include a global spare drive which automatically replace a failed drive ,which automatically replace a failed drive anywhere in Data Domain System.
- When a hot swappable failed drive is replaced by EMC,it become new global spare.
On-the-fly error detection and correction:
On the every read from disk the system first verifies that the block read from disk is the block excepted.It then use the checksum to verify integrity.
Scrub to insure data doesn’t go bad
Re-verify the integrity of all data .Scrubbing process intelligently find and repair defect on the disk before they can become a problem.
File system recoverability
File system recoverability:
To re-construct lost or corrupted file system metadata and also file system check tools that can bring an ailing system safely back online quickly.
Self-describing data format to ensure metadata recoverability:
All data is stored along with metadata that describes it. If a metadata structure is somehow
corrupted, there are two levels of recoverability.
- A snapshot is kept of the file system metadata every several hours; recoverability can rely on this point-in-time copy.
- The data can be scanned on disk and the metadata structure can be rebuilt.
These capabilities enable recoverability even if there is a worst case corruption of the file system or its metadata.
FS check, if needed, is fast:
Data Domain file system never overwrites old data and doesn’t have block maps and
reference counts to rebuild, it only has to verify where the head of the log is to safely
bring the system back online to restore critical data.
====================================================
No comments:
Post a Comment