DEA-1TT5 RAID and Data Protection 3

Question 1

What is 'erasure coding' and how does it compare to RAID?

Accepted Answer

It distributes data and redundancy across many nodes using mathematical algorithms, offering more flexible fault tolerance than traditional RAID

Answer

Erasure coding breaks data into fragments and adds calculated redundancy pieces distributed across nodes, allowing configurable tolerance for multiple node/disk failures — common in object storage and scale-out arrays.

Question 2

In a RAID 50 configuration, data is striped across multiple RAID 5 sets — what advantage does this provide?

Accepted Answer

It improves performance and fault tolerance by combining striping across multiple redundant parity groups

Answer

RAID 50 (RAID 5+0) stripes data across two or more RAID 5 sets, improving I/O performance and allowing one drive failure per RAID 5 group while maintaining data availability.

Question 3

What is the purpose of a 'patrol read' (or disk scrubbing) in a RAID array?

Accepted Answer

To proactively scan all disks for errors and correct them before they cause a drive failure

Answer

Patrol read continuously scans all drives in a RAID set for media errors (UREs) and corrects them using parity, preventing silent data corruption before a disk actually fails.

Question 4

Which RAID level is best suited for a write-intensive database that requires high availability and fast recovery?

Accepted Answer

RAID 10

Answer

RAID 10 provides the best write performance (no parity computation) and fastest rebuild times (copy from mirror), making it ideal for high-transaction, write-intensive database workloads.

Question 5

What is an 'URE' (Unrecoverable Read Error) and why is it significant in RAID arrays?

Accepted Answer

A disk read error that cannot be corrected by error correction codes, which can cause data loss during RAID rebuilds

Answer

A URE occurs when a disk encounters a sector it cannot read after all retry and ECC attempts; during a RAID 5 rebuild, a single URE on any remaining disk means the rebuilding data for that stripe is unrecoverable.

DEA-1TT5 - Information Storage and Management Exam Practice Test

DEA-1TT5 - Information Storage and Management Exam Practice Test

DEA-1TT5 RAID and Data Protection 3