1 |
Kai Krakow <hurikhan77@×××××.com> writes: |
2 |
|
3 |
> Am Wed, 20 Jan 2016 01:46:29 +0100 |
4 |
> schrieb lee <lee@××××××××.de>: |
5 |
> |
6 |
>> >> Overcommitting disk space sounds like a very bad idea. |
7 |
>> >> Overcommitting memory is not possible with xen. |
8 |
>> > |
9 |
>> > Overcommitting diskspace isn't such a bad idea, considering most |
10 |
>> > installs never utilize all the available diskspace. |
11 |
>> |
12 |
>> When they do not use it anyway, there is no reason to give it to them |
13 |
>> in the first place. And when they do use it, how do the VMs handle |
14 |
>> the problem that they have plenty disk space available, from their |
15 |
>> point of view, while the host which they don't know about doesn't |
16 |
>> allow them to use it? |
17 |
>> |
18 |
>> Besides, overcommitting disk space means to intentionally create a |
19 |
>> setup which involves that the host can run out of disk space easily. |
20 |
>> That is not something I would want to create for a host which is |
21 |
>> required to function reliably. |
22 |
>> |
23 |
>> And how much do you need to worry about the security of the VMs when |
24 |
>> you build in a way for the users to bring the whole machine, or at |
25 |
>> least random VMs, down by using the disk space which has been |
26 |
>> assigned to them? The users are somewhat likely to do that even |
27 |
>> unintentionally, the more the more you overcommit. |
28 |
> |
29 |
> Overcommitting storage is for setups where it's easy to add storage |
30 |
> pools when needed, like virtual SAN. You just monitor available space |
31 |
> and when it falls below a threshold, just add more to the storage pool |
32 |
> whose filesystem will grow. |
33 |
> |
34 |
> You just overcommit to whatever storage requirments you may ever need |
35 |
> combined over all VMs but you initially only buy what you need to start |
36 |
> with including short term expected growth. |
37 |
> |
38 |
> Then start with clones/snapshots from the same VM image (SANs provide |
39 |
> that so you actually do not have to care about snapshot dependencies |
40 |
> within your virtualization software). |
41 |
> |
42 |
> SANs usually also provide deduplication and compression, so at any |
43 |
> point you can coalesce the images back into smaller storage |
44 |
> requirements. |
45 |
> |
46 |
> A sane virtualization solution also provides RAM deduplication and |
47 |
> compaction so that you can overcommit RAM the same way as storage. Of |
48 |
> course it will at some point borrow RAM from swap space. Usually you |
49 |
> will then just migrate one VM to some other hardware - even while it is |
50 |
> running. If connected to a SAN this means: You don't have to move the |
51 |
> VM images itself. The migration is almost instant: The old VM host acts |
52 |
> as some sort of virtualized swap file holding the complete RAM, the new |
53 |
> host just "swaps in" needed RAM blocks over network and migrates the |
54 |
> rest during idle time in the background. This can even be automated by |
55 |
> monitoring the resources and let the VM manager decide and act. |
56 |
> |
57 |
> The Linux kernel lately gained support for all this so you could |
58 |
> probably even home-brew it. |
59 |
|
60 |
Ok, that makes sense when you have more or less unlimited resources to |
61 |
pay for all the hardware you need for this. I wonder how much money |
62 |
you'd have to put out to even get started with a setup like this ... |