1 |
On Mar 10, 2012 8:33 PM, "Alex Schuster" <wonko@×××××××××.org> wrote: |
2 |
> |
3 |
> Hi there! |
4 |
> |
5 |
> Is there an advantage in putting the portage tree on an extra partition? |
6 |
> |
7 |
> Currently, I'm using reiserfs, because I read that it is efficient when |
8 |
> using many small files. On the other hand I also heard that it tends to |
9 |
> get slower with every emerge --sync. |
10 |
> |
11 |
> Space is no longer an argument in these days, at least for my desktop |
12 |
> machine. But I would like to optimize for speed -- emerge -DputnVj |
13 |
> @world takes quite a while to calculate, I assume this is because so many |
14 |
> ebuild files have to be accessed. |
15 |
> |
16 |
> Any tips on this? Does it make sense to use a special file system just |
17 |
> for the portage tree? What would be best? Would it help to re-create this |
18 |
> file system from time to time in case it gets slower with every sync? Or |
19 |
> wouldn't I notice a difference if I just used a big ext4 partition for |
20 |
> all portage related stuff? |
21 |
> |
22 |
> Anyone using a compressed RAM file system for that? :) |
23 |
> |
24 |
|
25 |
This had been my burning question when I was deploying the company's |
26 |
production server, and forced me to do some research: |
27 |
|
28 |
* reiserfs is amazingly fast for reads, but suffers on simultaneous writes |
29 |
* reiserfs does not have inode limits |
30 |
* reiserfs' notail affects performance greatly depending on the nature of |
31 |
the system: I/O-bound (use notail) or CPU-bound (don't use notail) |
32 |
* reiserfs, if mounted without notail, is very space-efficient |
33 |
|
34 |
So, I end up with the following mix: |
35 |
|
36 |
* ext2 for /boot |
37 |
* reiserfs for /usr/portage and /var/tmp (RAM is at premium; can't use |
38 |
tmpfs) |
39 |
* ext4 for everything else |
40 |
|
41 |
This cocktail has been serving me well. I don't need advanced filesystems |
42 |
like ZFS, XFS, or btrfs, because my servers are virtualized, and the |
43 |
advanced features (e.g., snapshot) is handled by the underlying hypervisor |
44 |
(XenServer) and SAN Storage (we use NetApp). |
45 |
|
46 |
Rgds, |