Dream Machine ( Build your own PC ) Message Board

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Pastimes : Dream Machine ( Build your own PC )

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next

To: Zeuspaul who wrote (822)	5/25/1998 12:58:00 PM
From: LTBH	of 14778

Error Detection and Correction and RAID The 3 common error detection schemes in use today are checksum, parity and ECC. The first two will detect and notify but will not correct. Parity and ECC can be used in any data path. Checksum is commonly used as a final gross check of data after either a data transmission of a block of data over a modem (where a failure causes a retransmission) or after a block of data is read from a tape drive (where a failure usually asks for operator intervention). Checksum is normally a modem implementation for PC use. Parity checking is the addition of an extra bit to (usually) an 8 bit byte. This addition is such that all bytes will end up with odd parity. When a byte is detected that is not odd parity, a parity error occurs usually followed by a hard error. Parity detection will not detect multi bit errors. Drop two bits and the byte stays odd. Parity is available in PC RAM but seldom implemented due to extra cost. ECC goes back to the mainframe tape drive days. Tapes were slow and unreliable devices where a multidrive sort could take 20 hours. Enter ECC (error correction check or code) which would detect and correct single bit errors and detect but NOT correct multibit erros. Both parity and ECC are used internally on mainframes at various functional blocks since you wish to detect/correct the error as close to the point of occurrence as possible. Before you have taken action on the data (anding/oring data, before a branch based on a relative address etc.) On todays PCs, you will find ECC for SDRAM, RAID and high end tape subsystems. Can't really discuss your specific failure because you don't know the cause. However how would the second HD become identical? It probably would end up with a corrupted registry also. Unless you did a destructive (for Win) power down without closing all programs. Would rather say what RAID with ECC implemented will do.1) if this was caused by a single dropped bit to the HD the data would have been correctly written to HD. 2) If from the HD, the correct data would have gone to the CPU. 3) if caused by a physical bad HD you could have continued running with no redundancy or hot swapped a HD for background data reconstruction and continued running. Implementation of ECC is RAIDlevel 3 (I think) and each manufactures do things slightly different. You need to check the manufacturer/model for its exact method of implementation. I would guess that most would use HW but SW is always cheaper. On the fly simply means its fast and transparent to your operation. Hope this helps Networm Kept timing out at end of my reply, hope I haven't forgotten to include everything in this one

Report TOU Violation

Share This Post

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next