While there are significant benefits, such as lower power consumption, to being able to store more data on fewer drives, there are new challenges for backup operations as well. W. Curtis Preston, backup expert and founder of Truth in IT, discusses some of those challenges in this video, noting that the risk of a failure isn't eliminated just because storage systems have more hard drive capacity.
"As the data gets bigger, the drives get bigger, and as drives get bigger, they don't necessarily get more reliable," Preston said.
Preston cited a statistic that suggested that RAID rebuilds require about 24 hours to 36 hours per terabyte. For systems that have multi-terabyte hard drive capacity, that is a significant amount of time. "That's three to five days to rebuild one failed SATA [Serial Advanced Technology Attachment] drive," he said of the multi-terabyte units. During that time, a RAID 5 system is susceptible to data loss, and while they offer more protection, even RAID 6 systems are not without risk, he added.
"The odds -- and actual happenings -- of people having second drive failures in RAID 6 arrays are well documented," Preston said. "You're beating on it, and your performance [suffers]. Meanwhile, you're at risk for a week to two weeks. … Scary."
Preston also said that traditional backups can't keep up with the demands of today's workloads for hard drive capacity, especially because single servers carry a lot of data, files and virtual machines.
Traditional backup systems back up entire files if just one block changes, full backups are still conducted of databases and applications, plus restores require moving "everything back from A to B," according to Preston.