Quote:
Originally Posted by citivolus
While you are at it you may want to consider the use of a solid state hard drive, to reduce your mechanical failure chances to zero. Electrolytic capacitors in the power supply would likely be your primary source of failure in such a configuration, so depending on your capacity overhead, you could likely go several years without replacing components.
|
Hi Ric,
Though solid state drives (SSDs) can provide wonderful solution in many applications,
one needs to be aware of the inherent limitations of FLASH memory, which are
commonly used in many modern SSD's. FLASH sectors only have a limited number of
write cycles (typically in the order of 100,000) before the sector fails. Firmware
integrated within the disk controller or a FLASH-aware file system can mitigate
the wear by distributing writes to disk across the disk, what is commonly referred
to as 'wear-leveling'.
Nevertheless, when using SSD's that employ FLASH memory in a mission critical
application, one needs to perform some profiling of the number of write cycles
that occur when the target OS and applications are installed and running.
Applications that perform large numbers of writes can result in SSD drive
mean-time before failure (MTBF) rates that are lower than conventional magnetic
drives.
In other words, without profiling the application's disk requirements, using
SSD's may result in a situation where the failure rate can be made worse.
Best Regards
Gary Kopff
Mt. Kuring-Gai