prakashsurya

5 ms·

1.
OpenZFS: Notes on ZIL Transactions(prakashsurya.com)
2 points by prakashsurya 9y ago|0 comments
2.
OpenZFS: Artificial Disk Latency Using Zinject(blog.delphix.com)
1 points by prakashsurya 11y ago|0 comments
3.
by prakashsurya 12y ago
Well, if the exact same semantics were required (i.e. always evict the oldest buffer), then it's really O(M*n) where M is the number of bytes to evict, and n being the number of sublists; each sublist would have to be checked for each
4.
by prakashsurya 12y ago
Currently, the number of sublists created is equal to the number of cores on the system. While it'd be fairly trivial to extend the code to make it easy for an admin to define the number of sublists, I don't have any reason to bel
5.
OpenZFS: Reducing ARC Lock Contention(blog.delphix.com)
65 points by prakashsurya 12y ago|7 comments
6.
OpenZFS Device Removal(blog.delphix.com)
60 points by prakashsurya 12y ago|6 comments
7.
by prakashsurya 12y ago
Could not have said it better myself. It's disappointing to see a post with such unfounded nonsense rated so high on HN.
8.
by prakashsurya 12y ago
yup, that's why I added "by default". a user won't get that functionality without explicitly asking for it.
9.
by prakashsurya 12y ago
Error detection can always be used, but error correction may or may not be available (it depends on the type of block). Metadata blocks are redundant even on a single drive pool; so if you just have a partial failure (e.g. overwrite a m
10.
by prakashsurya 12y ago
No problem. It's really a shame to have such a high bar for using dedup. It either fits a given workload extremely well, or can be extremely detrimental. There's been talk in the developer community about ways to address the usabi
11.
by prakashsurya 12y ago
There is work in progress to add the ability to remove a device from a pool. So, hopefully in the not too distant future, that work will land in illumos and migrate to all the downstream implementations (e.g. linux, mac, and freebsd).
12.
by prakashsurya 12y ago
No, that's flat out not true. I've seen that metric thrown around when talking about the "dedup" feature of ZFS, but honestly, don't use dedup unless you know what you're doing. It's way to easy for things
13.
by prakashsurya 12y ago
Yes, I run ZFS on Linux on my single drive laptop currently. ZFS doesn't _need_ multiple drives to work well, but it is capable of using them if they're available. One can also run a HW raid solution underneath ZFS, where the RAID
14.
by prakashsurya 12y ago
FWIW, if there's anybody interested in learning about the ZFS code base, we'd love help porting patches from ZoL into Illumos and vice versa. That's a good way to get a new developer integrated with the code and process surro
15.
by prakashsurya 12y ago
I can't recall the exact detail from memory, but I believe #1 has to do with the fact that zfs creates/clones/snapshots/etc are done in "syncing" context. Thus, each command has to wait for a full pool sync to
16.
by prakashsurya 13y ago
Is dtrace available on Linux? I know there's been work towards that goal, but I haven't payed much attention to it recently.