Everyone hates it when their computer crashes reboots randomly or shows the dreaded blue screen of death for the average user though this isn’t really the end of the world, but some systems like cloud servers or supercomputers rely on being operational for 24 hours a day and a fatal crash could mean the loss of data, service outages, or cost company millions of dollars but for some IT professionals this may seem like a catastrophic server outage is actually the end of the world the likelihood of it happening.
Thanks to technologies like ECC memory which can actually be quite slim, ECC stands for error correcting code and the principle behind it using mathematical methods to check data for errors quickly has actually been around since about 150BC when Jewish scribes developed a process in which they looked at the number of words in a page or line to quickly determine if there were transcription errors, ECC memory works much the same way as it can quickly check and correct for the most common types of data corruption which often are the cause of crashes and data loss.
Okay so why is this so important well the more that we use cloud services and virtual machines for massive calculations and computing means servers are not only important to large corporations but also to the average consumer, take your Dropbox or iCloud account for instance the servers which store your precious data rely on ECC memory to prevent memory failures which may otherwise leave your data inaccessible or worse lost altogether, many things like electrical magnetic or even cosmic interference can cause a single bed whose value is either 1 or 0 to spontaneously flip to the opposite state as you may know each byte in your memories made up of 8 bits which in the case of your computer’s native software language might represent a letter or a number, if a single bit is switched to its opposite position the result might be garbled data and if this byte is critical to a systems operation it can cause a crash or a failure.
ECC memory corrects for this by first performing what is called parity checking which involves storing an extra parity bit that represents the sum of all the ones in an 8-bit group, the result of this is either an odd or even number one or zero if the sum doesn’t match up with the parity bit the next time the RAM accesses the data then the system will know that at least one part of the bit must have been corrupted than in this case ECC Ram will refer back to code it generated using a special algorithm when the good data was first stored in memory and used this code to correct the error restoring the original 8-bit block of data.
So that’s cool right but if you’re asking should I buy ECC memory for my next build and you aren’t making some kind of server where a zero downtime is absolutely necessary then the answer is likely no ECC memory comes at a higher cost with worse performance due to the greater processing required and isn’t even compatible with most consumer motherboards not to mention if you’re going for that flashy RGB dream rig you might be disappointed to note that ECC memory PCBs are usually green and typically lack a heatsink here in certain scenarios and stuff which might make the next dream build of yours look a little less special to the untrained eye.
Alright, guys, that’s the end of the blog, thanks for reading the whole way through if you enjoyed this blog please share it with someone who would be interested and leave a comment, Thanks for reading guys.
Talking Talks 