| 1 |
On 03/27/2013 05:51 PM, Alan McKinnon wrote: |
| 2 |
> On 27/03/2013 22:41, Michael Mol wrote: |
| 3 |
>> The case for systemd is twofold: |
| 4 |
> |
| 5 |
> ... |
| 6 |
> |
| 7 |
>> 2) Reduce the amount of CPU and RAM consumed when you're talking about |
| 8 |
>> booting tens of thousands of instances simultaneously across your entire |
| 9 |
>> infrastructure, or when your server instance might be spun up and down |
| 10 |
>> six times over the course of a single day. |
| 11 |
> |
| 12 |
> I seems to me that this is rather a niche quite-specialized case (albeit |
| 13 |
> a rather large instance of a niche case). In which case it would be |
| 14 |
> better implemented as Redhat MagicSauce for their cloud environment |
| 15 |
> where it would be exactly tuned to that case's need. |
| 16 |
|
| 17 |
But it's a great deal cheaper to convince volunteers and package |
| 18 |
maintainers to put in the time to build the necessary service files of |
| 19 |
their own accord. Add in the complexity of parallel boot, and you can |
| 20 |
induce upstream to fix their own race-driven bugs rather than have to |
| 21 |
pay for that development directly. |
Archives