Ahmad was kind enough to present on the topic of RAID storage technology.
RAID, an acronym for Redundant Array of Independent Disks. It is an approach to implement replication and data storage across multiple physical drives. The implementation can be done via software or hardware. There are range of schemes, we discussed RAID 0-5. The implications for efficiency and redundancy. For computationally intensive raid modes (for example raid 5), software implementation will have significant overhead as compared to hardware implementation.
There was a discussion of the redundancy built into RAID 1 using Hamming code parity check and this moved to a discussion of hash functions and the purpose of checksum and md5sum as tools to confirm the reliability copied data. This allows for the total failure of one drive or repair of corrupted data using the parity check. In addition RAID 10 (1+0 and 0+1) and the features of each for dealing with redundancy were looked at.
There were brief asides on the emacs undo-tree package and using top to find virtual memory culprits.
FLOSS Weekly
Recent Comments