Re:Initial Hammer2 filesystem implementation
by binarycrusader on Tuesday August 22, 2017 @12:04AM
(#15069611)
The caching can be considered optional and (in Oracle Solaris) there is a tunable user_reserve_hint_pct:
https://docs.oracle.com/cd/E53394_01/html/E54818/chapterzfs-...
Re:Initial Hammer2 filesystem implementation
by chongli on Monday August 21, 2017 @09:15PM
(#15069017)
Performance is substantially worse than other filesystems with more modestly sized caches, so the memory isn't really available to applications.
Is that in comparison to other filesystems that provide the same level of data integrity as ZFS? Otherwise you're comparing apples and oranges.
Re:Initial Hammer2 filesystem implementation
by tedunangst on Monday August 21, 2017 @09:18PM
(#15069031)
You didn't say the memory use was justified by providing better integrity. You said it was optional.
Re:Initial Hammer2 filesystem implementation
by chongli on Monday August 21, 2017 @09:29PM
(#15069072)
Yes, the memory use is optional. Pick two:
1) data integrity
2) low memory use
3) performance
Re:Initial Hammer2 filesystem implementation
by dozzie on Monday August 21, 2017 @09:56PM
(#15069176)
With ZFS I hear it's "pick one, and you cannot pick `low memory'".
Re:Initial Hammer2 filesystem implementation
by aidenn0 on Monday August 21, 2017 @11:51PM
(#15069559)
You can run ZFS with megabytes of ram, it's just slow, so it is pick 2. That being said, ZFS metadata is much bigger than it needs to be (most infamously dedupe has this issue).
Re:Initial Hammer2 filesystem implementation
by chongli on Tuesday August 22, 2017 @03:44AM
(#15070429)
Right, but to then compare the memory use of ZFS with dedupe turned on against another fs without that feature would be disingenuous since dedupe is optional in ZFS. That was the whole point of my original comment.
Re:Initial Hammer2 filesystem implementation
by mrighele on Tuesday August 22, 2017 @04:17AM
(#15070542)
deduplication doesn't have necessarily to be expensive; HAMMER (the first version, not HAMMER2) has offline deduplication that is usually scheduled to run at night. This allows regular use to be quite performant and with a low memory footprint.
Of course there are tradeoffs, in particular heavy disk usage at certain hours (which can be an issue depending on the workload) and the fact that space is reclaimed only after some time that it has been wasted.
Re:Initial Hammer2 filesystem implementation
by aidenn0 on Tuesday August 22, 2017 @12:41PM
(#15074089)
I did say that ZFS metadata in general, and dedupe especially is much bigger than it needs to be. Even live dedupe can be done much more cheaply than how ZFS does it.
Re:Initial Hammer2 filesystem implementation
by gigatexal on Tuesday August 22, 2017 @12:03AM
(#15069606)
Lol this is FUD. ZFS is easily tunable to not blow up and use all your ram.
Re:Initial Hammer2 filesystem implementation
by benchaney on Tuesday August 22, 2017 @02:09AM
(#15070134)
Does "FUD" just mean "something I disagree with" now? How is this FUD?
Re:Initial Hammer2 filesystem implementation
by gigatexal on Tuesday August 22, 2017 @03:29AM
(#15070384)
Spreading misinformation is what I was taking issue with not just disagreeing and making a stink about it like a troll might.
Re:Initial Hammer2 filesystem implementation
by JdeBP on Tuesday August 22, 2017 @04:06AM
(#15070505)
You're not appreciating nor applying the fear, uncertainty, and doubt parts of "Fear, Uncertainty, and Doubt".
Re:Initial Hammer2 filesystem implementation
by gkya on Tuesday August 22, 2017 @06:25AM
(#15071051)
Talking about downsides of a thing is not FUD nor making stink about it.
Re:Initial Hammer2 filesystem implementation
by snovv_crash on Tuesday August 22, 2017 @01:49AM
(#15070043)
Depends if you want dedup.
I believe the complaint about ZFS and RAM is that the caching cannot be considered optional. Performance is substantially worse than other filesystems with more modestly sized caches, so the memory isn't really available to applications.