Re:Oracle built a 1,060 node Raspberry Pi cluster
by ComputerGuru on Thursday November 21, 2019 @03:17PM
(#21599040)
ZFS is 100% software raid.
[ Parent
| on HN ]
Re:Oracle built a 1,060 node Raspberry Pi cluster
by whalesalad on Thursday November 21, 2019 @03:19PM
(#21599064)
Sorta. At the end of the day, yes, the disks are being managed by your CPU and not a dedicated raid controller... but it's a completely different ballgame than something like mdadm.
Re:Oracle built a 1,060 node Raspberry Pi cluster
by ownagefool on Thursday November 21, 2019 @04:05PM
(#21599610)
Not "sorta", it's software raid and thus, like software raid, it's portable to any system that'll run the software and JBOD the drives.
Most would argue RAIDZ is better, but many of these reasons are functionality based over and about just redundancy, i.e. Copy on Write, et all.
Re:Oracle built a 1,060 node Raspberry Pi cluster
by GhettoMaestro on Thursday November 21, 2019 @07:03PM
(#21601153)
Friends don't let friends create RaidZ pools.
Bite the bullet of the cost of N*2 mirrors.
Re:Oracle built a 1,060 node Raspberry Pi cluster
by whalesalad on Thursday November 21, 2019 @04:23PM
(#21599790)
I say sorta because in most circles when one is speaking of 'software RAID' they are referring to a very specific configuration and are explicitly not referring to ZFS.
Re:Oracle built a 1,060 node Raspberry Pi cluster
by colejohnson66 on Thursday November 21, 2019 @03:55PM
(#21599497)
ZFS doesn’t support hardware RAID controllers?
Re:Oracle built a 1,060 node Raspberry Pi cluster
by whalesalad on Thursday November 21, 2019 @04:22PM
(#21599775)
It supports it, but ZFS itself is operating on the host CPU, not the RAID controller. Additionally, ZFS prefers that the RAID controller is disabled/defeated and operating in JBOD mode where disks are passed-thru to the OS without any caching.
Re:Oracle built a 1,060 node Raspberry Pi cluster
by ComputerGuru on Thursday November 21, 2019 @10:34PM
(#21602192)
Not just caching. The problem is that non-JBOD HBAs also swallow error codes and mask retries, meaning ZFS is potentially unaware of some read errors and is not in control of the latency as the controller firmware has its own timeouts and retry counters. Moreover, most will not pass through SMART or else fake it.