作者: New Human 時間: 2011-8-23 15:11 標題: Linux with ZFS
提示: 作者被禁止或刪除 內容自動屏蔽
作者: little_keung 時間: 2011-8-23 15:42
ZFS_on_Linux
但係, ZFS 有咩 feature 真係想擺嚮 Linux 到呢 ?
作者: New Human 時間: 2011-8-23 15:46
提示: 作者被禁止或刪除 內容自動屏蔽
作者: New Human 時間: 2011-8-23 15:48
提示: 作者被禁止或刪除 內容自動屏蔽
作者: liangsuilong 時間: 2011-8-23 21:19
好似 ZFS On Linux 已經死左咯。
宜家都係睇 BTRFS 幾時搞掂。
作者: amoslee 時間: 2011-8-24 01:14
小弟正想試試NexentaStor,未知c hing有何心得?
作者: rayman 時間: 2011-8-24 13:31
提示: 作者被禁止或刪除 內容自動屏蔽
作者: New Human 時間: 2011-8-24 13:38
提示: 作者被禁止或刪除 內容自動屏蔽
作者: killerpub 時間: 2011-8-25 00:37
Gentoo 用 btrfs bleeding 噴血 好順
Debian 好似磨薑咁

same partition, multiple roots
作者: killerpub 時間: 2011-8-25 00:40
Ubuntu 都係磨薑

作者: liangsuilong 時間: 2011-8-25 03:19
Gentoo 用 btrfs bleeding 噴血 好順
Debian 好似磨薑咁
same partition, multiple roots ...
killerpub 發表於 2011-8-25 00:37
Btrfs 仲有好多嘢未搞得店,唔係 Fedora 都唔會 postpone btrfs default 去到 Fedora 17 喇。
前冇幾耐仲係話 Btrfs 行 Database 同 Virtualization 都仲係話有好多問題,會比 EXT4 慢幾條街,唔知好左喇。
作者: rayman 時間: 2011-8-25 03:52
提示: 作者被禁止或刪除 內容自動屏蔽
作者: icarus-c 時間: 2011-8-25 13:05
回復 7# rayman
OpenSolaris 收左皮比人fork左做 OpenIndiana
作者: killerpub 時間: 2011-8-27 00:43
Btrfs 仲有好多嘢未搞得店,唔係 Fedora 都唔會 postpone btrfs default 去到 Fedora 17 喇。
前冇幾耐 ...
liangsuilong 發表於 2011-8-25 04:19
一直用緊 btrfs, 因為一個 partition 可以好多個 root path
gentoo 好順無事, debian 有些古怪,明顯 laggy。Ubuntu 唔在講。
作者: killerpub 時間: 2011-8-27 00:44
USB bootable Linux 都係 btrfs, multiple distros with compression on flash drives
幾時有咁好康既野?
作者: icarus-c 時間: 2011-8-27 15:05
我都有諗過試玩btrfs on root 但係諗到如果炒左 無咩livecd 夠update搞btrfs :funk:
作者: computer1975 時間: 2011-8-29 18:05
ZFS個license不兼容GPL不能直接放到kernel
還有本身ZFS會開發FUSE,聽聞好似都似摺,想玩ZFS轉用FreeBSD
作者: nohuy 時間: 2011-8-29 20:42
ZFS個license不兼容GPL不能直接放到kernel
還有本身ZFS會開發FUSE,聽聞好似都似摺,想玩ZFS轉用FreeBSD ...
computer1975 發表於 2011-8-29 18:05
FreeNAS已經支援了
作者: computer1975 時間: 2011-8-29 20:54
沒記錯FreeNAS是base on FreeBSD
作者: lockheed 時間: 2011-8-29 22:04
using Nexenstor
Stable and easy to use.
作者: amoslee 時間: 2011-8-29 22:14
I am using FreeNas 8, just for some simply usages.
Would like to try Nexenstor in a few days, c hing do you have any 心得 ?
THX!
作者: New Human 時間: 2011-8-29 22:44
提示: 作者被禁止或刪除 內容自動屏蔽
作者: killerpub 時間: 2011-8-31 22:58
Btrfs 仲有好多嘢未搞得店,唔係 Fedora 都唔會 postpone btrfs default 去到 Fedora 17 喇。
前冇幾耐 ...
liangsuilong 發表於 2011-8-25 04:19
可以用 feed reader subscribe btrfs
印像中 linux fuse zfs 口碑幾好
始終 btrfs 唔同目前既 linux ext filesystems
原生 raid (bye bye mdadm), compression, ssd support, subvolumes
zfs 梗係好啦,有 dedupe, encryption, etc..
hardware 要求呢?
作者: killerpub 時間: 2011-8-31 23:00
since
Filesystem created: Sat Oct 23 23:36:23 2010
無數強制 power off
btrfs 都沒死

作者: frank7lan 時間: 2011-9-1 12:14
有道理 !
作者: frank7lan 時間: 2011-9-1 12:17
我一般在公司既 production server 都係用 XFS 同 EXT3 On CentOS
最近測試轉左d EXT3 上 EXT4
作者: icarus-c 時間: 2011-9-1 12:41
回復 26# frank7lan
老細要求定係你自己咁勇敢
作者: lazyfai 時間: 2011-9-1 14:21
公司用 CentOS 本身已經好勇

作者: killerpub 時間: 2011-9-1 22:57
本帖最後由 killerpub 於 2011-9-1 23:59 編輯
http://www.sysresccd.org/Download
或者 自制 usb debian
filesystem 本身 support raid , compression, ssd, 唔駛 mdadm 好
作者: icarus-c 時間: 2011-9-1 23:49
本帖最後由 icarus-c 於 2011-9-1 23:52 編輯
回復 29# killerpub
我對佢compression最有興趣
小弟家貧 無錢買hard disk 用notebook hdd慢到死 之前squash一squash read快左好多 不過squashfs唔方便
作者: killerpub 時間: 2011-9-3 00:47
回復 30# icarus-c
要去細嘗一下 btrfs wiki 啦
一支手指用 subvolume 可以好多 root
即係好多個 Linux lives in a single partition
cpu 夠應付 compress 就行了
投資 effort 在 Linux ZFS, 個人上唔值得。
至於 FreeBSD 都係得個追字,即係夢。
作者: t1066 時間: 2011-9-3 11:27
回復 31# killerpub
正在使用 FreeBSD + zfs v28 的路過
作者: t1066 時間: 2011-9-3 11:29
本帖最後由 t1066 於 2011-9-3 11:34 編輯
基本上,若不使用dedup或encrypt等功能的話,cpu的要求並不太高,當然ram是越多越好。
作者: computer1975 時間: 2011-9-3 20:57
我一般在公司既 production server 都係用 XFS 同 EXT3 On CentOS
最近測試轉左d EXT3 上 EXT4 ...
frank7lan 發表於 2011-9-1 12:17
終於找到有ching用sgi XFS
作者: lazyfai 時間: 2011-9-4 13:19
我都用咗好多年啦。不過之前炒過一次大大鑊就記得要成日backup了。
作者: ipsec 時間: 2011-9-7 22:11
睇過一些FreeNAS+ZFS 的報道、最好每1TB disk space 要 1GB RAM.
作者: t1066 時間: 2011-9-9 12:38
由於zfs可以用SSD作為cache drive,所以未必需要1G ram對1T data,就好像以下:
https://rcbi.rochester.edu/users ... nd_OpenSolaris.html
不過要注意的是,reboot後cache drive會重新運作,失去之前的cache。所以比較適合長期運行的server.
作者: ipsec 時間: 2011-9-9 12:48
本帖最後由 ipsec 於 2011-9-9 12:53 編輯
回復 37# t1066
I think the quoted example is a extremely case, the writer use 13 x 2TB hard disks on his test lab which is not a common use for SOHO/Home user.
http://doc.freenas.org/index.php/Hardware_Requirements
"RAM
The best way to get the most out of your FreeNAS™ hardware is to install as much RAM as possible. If your RAM is limited, consider using UFS until you can afford better hardware. ZFS typically requires a minimum of 6 GB of RAM in order to provide good performance; in practical terms (what you can actually install), this means that the minimum is really 8 GB. The more RAM, the better the performance, and the Forums provide anecdotal evidence from users on how much performance is gained by adding more RAM. For systems with large disk capacity (greater than 6 TB), a general rule of thumb is 1GB of RAM for every 1TB of storage. "
"Storage Disks and Controllers
.....For example, moving the ZIL (ZFS Intent Log) to a dedicated SSD only helps performance if you have synchronous writes, like a database server. SSD cache devices only help if your working set is larger than system RAM, but small enough that a significant percentage of it will fit on the SSD. "
作者: t1066 時間: 2011-9-10 00:53
回復 38# ipsec
I would refer to the following article:
http://constantin.glez.de/blog/2 ... esystem-performance
#4: Use SSDs to Improve Read Performance
If you can't add any more RAM to your server (or if your purchasing department won't allow you), the next best way to increase read performance is to add solid state disks (aka flash memory) as a level 2 ARC cache (L2ARC) to your system.
You can easily configure them with the zpool(1M) command, read the "Cache devices" section of its man-page.
SSDs can deliver two orders of magnitude better IOPS than traditional harddisks, and they're much cheaper on a per-GB basis than RAM.
They form an excellent layer of cache between the ZFS RAM-based ARC and the actual disk storage.
You don't need to observe any reliability requirements when configuring L2ARC devices: If they fail, no data is lost because it can always be retrieved from disk.
This means that L2ARC devices can be cheap, but before you start putting USB sticks into your server, you should make sure they deliver a good performance benefit over your rotating disks :).
SSDs come in various sizes: From drop-in-replacements for existing SATA disks in the range of 32GB to the Oracle Sun F20 PCI card with 96GB of flash and built-in SAS controllers (which is one of the secrets behind Oracle Exadata V2's breakthrough performance), to the mighty fast Oracle Sun F5100 flash array (which is the secret behind Oracle's current TPC-C and other world records) with a whopping 1.96TB of pure flash memory and over a million IOPS. Nice!
And since the dedup table is stored in the ZFS ARC and consequently spills off into the L2ARC if available, using SSDs as cache devices will also benefit deduplication performance.
As you can see, an SSD or a good USB drive could be used as poor man's RAM. Moreover, in a Home/SOHO setting, I would guess a typical working set will be less than 100G which can easily fit into a 128G SSD.
作者: ipsec 時間: 2011-9-10 01:48
本帖最後由 ipsec 於 2011-9-10 11:10 編輯
回復 39# t1066
From the same article, the writer mentioned "Add Enough RAM" & "Add More RAM" on top of the #1 & #2 places about read/write improvement and "Use of SSD for data caching only if you can't add any more RAM to your server.".
The writer also mentioned "Having enough RAM will benefit all of your reads, no matter if they're random or sequential, just because they'll be easier for ZFS to find on your disks, so make sure you have at least n/1000 + 1 GB of RAM, where n is the number of GB in your storage pool.".
It depends the storage size requirement for the installation, whether they should go to take the benefit of SSD as a second level cache. For the HOME/SOHO users, it is not difficult to build a PC/Server with 16G or even more than 32G RAM memory now day. According to the writer's rule, "n/1000 + 1 GB of RAM", that means the PC/Server can build a 15TB or 31TB more disk storage pool to be used. Some of them may add a hardware RAID card solution instead of building software RAID-Z environment.
As an installation have to use SSD as cache, their disk storage requirements are most likely greater than 31TB. Such installation, the user would like to concern on the data integrity/security/backup/recover support rather than the read/write performance by purchasing ORACLE/IBM/EMC SAN solution in order to make their invaluable data safety. Of course, data read/write performance is a value add factor.
作者: t1066 時間: 2011-9-10 12:08
回復 40# ipsec
While I generally agree with what you said, there are a few points that I want to make:
1. The purpose of using cache drive is foremost to increase the IOPs of the system. That is why we are seeing technologies like Intel smart response, hybrid drive Seagate momentus XT or LSI cachecade.
2. Dedup. When using Dedup, the requirement for RAM jumps significantly. It would be prudent to have spare capacity in the form of SSD to fall back to. Otherwise, the performance of the pool will plummet.
3. It is possible to use ZFS with some old computer parts. Such systems would only have 1-2G of RAM. They could benefit from an SSD or a fast USB flash drive as cache drive.
4. It is not recommended to use ZFS over hardware RAID. Not only do you lose one important feature of ZFS: detect and correct data error, your pool would not be portable.
作者: ipsec 時間: 2011-9-10 13:06
本帖最後由 ipsec 於 2011-9-10 14:34 編輯
回復 ipsec
While I generally agree with what you said, there are a few points that I want to make ...
t1066 發表於 2011-9-10 12:08
Thanks for your response.
From your pt. 1&2, in my point of view, the technology of Intel smart response, hybrid drive..., came out are targeted to low end PC/Server market. That's why we can buy it from PC market. If there are enough of system memory for the taking advantage of ZFS with a small disk storage required, why does HOME/SOHO users going with SSD as cache for the price added?
3. For the old computer parts, it is less of CPU power to take the ZFS others
feature, such as data integrity (SHA-2) throughout the file system tree and
data encryption. The ext3 filesystem is more suitable for old computer parts.
4. For the hardware RAID on ZFS, how about to build two physical disks(set) in hardware RAID-0/1 and configure five RAID-0/1 (as a example) sets in ZFS pool. I haven't tested this approach, have you?
Happy discuss,
cheer
作者: t1066 時間: 2011-9-11 11:44
回復 42# ipsec
Thanks for your feedback.
1&2. I have to agree with you that in many cases, adding an SSD would not be necessary. However, in some cases, having 32G RAM + 120G SSD (think of is as cheaper but slower RAM instead of more expensive but smaller HDD) would be beneficial, e.g. lower latency of the system.
3. Unless you are using Dedup, you should use fletcher4 instead of sha256 for checksum. Also, ZFS offer self healing of data. So it would be down to your needs.
4. I would just list a few objections for using hardware RAID with ZFS: increasing system complexity, you have to add spare drives to each of the RAID subsystem instead of adding them to the whole pool, hardware dependency.
Cheers.
作者: ipsec 時間: 2011-9-11 20:46
回復 ipsec
4. I would just list a few objections for using hardware RAID with ZFS: increasing system ...
t1066 發表於 2011-9-11 11:44
Agree. It is a general question of the hardware RAID pros and cons.

