1 |
-----BEGIN PGP SIGNED MESSAGE----- |
2 |
Hash: SHA1 |
3 |
|
4 |
|
5 |
http://gentoo-wiki.com/HOWTO_Migrate_to_NPTL |
6 |
http://gentoo-wiki.com/HOWTO_Gentoo_2004.2_for_linux_2.6_and_NPTL |
7 |
http://gentoo-wiki.com/HOWTO_Install_Gentoo_-_The_Gentoo_Developers_Method_with_NPTL_and_2.6_from_Stage1 |
8 |
|
9 |
radu herinean wrote: |
10 |
|
11 |
>I was just following the conversation but I am convinced now. |
12 |
>For the record, I am running two "pretty regular" web-oriented servers: |
13 |
>web, db, email, lists, dns, some low band streaming, no fileserving and |
14 |
>no nat or any type of routing. |
15 |
> |
16 |
>I've used nptl before, but I gave up because I had some problems running |
17 |
>turck-mmcache on a nptl-only apache. |
18 |
>Since mmcache is obsolete and not used on the server anymore I see no |
19 |
>problem switching back to a nptl-only environment. |
20 |
> |
21 |
>Although I think I know all the steps I am supposed to do, I would |
22 |
>really appreciate it if some would point out the "migration" steps, just |
23 |
>in case I don't miss some important stage. |
24 |
> |
25 |
>Thank you guys. |
26 |
> |
27 |
>-----Original Message----- |
28 |
>From: J. Ryan Earl [mailto:ryan@×××××××××××××××.com] |
29 |
>Sent: 10 august 2005 18:44 |
30 |
>To: gentoo-server@l.g.o |
31 |
>Subject: Re: [gentoo-server] nptl |
32 |
> |
33 |
> |
34 |
>Billy Holmes wrote: |
35 |
> |
36 |
>>This makes it easier and faster for the kernel to context switch |
37 |
>>between processes (and threads within a process). |
38 |
>> |
39 |
>>A prime example of this is a test from the LKML, where Ulrich Drepper |
40 |
>>wrote he was able to start and stop 100,000 threads in about 2 |
41 |
>>seconds. In further messages, Ingo Molnar confirms this and also |
42 |
>>points out that before these changes, it would have taken upwards to |
43 |
>>15 minutes for the same thing. |
44 |
>> |
45 |
>>http://kerneltrap.org/node/422 |
46 |
> |
47 |
> |
48 |
>Absolutely true. NPTL is drastically more scalable than linuxthreads. |
49 |
>For instance, spawn 1000 worker threads in a JVM and the scheduling |
50 |
>overhead alone will cause the system to become unresponsive across the |
51 |
>board. Use NPTL and you can get 10000 threads all running no problem. |
52 |
> |
53 |
>-ryan |
54 |
|
55 |
|
56 |
-----BEGIN PGP SIGNATURE----- |
57 |
Version: GnuPG v1.4.1 (GNU/Linux) |
58 |
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org |
59 |
|
60 |
iD8DBQFC+tITqjyCvrpPZ1sRAo2oAJ4uw9vGXr+L1KKDrtuT6btNOCPqvACggbK5 |
61 |
AM/LWgajPUFr01XYeIsM+y8= |
62 |
=Pmrm |
63 |
-----END PGP SIGNATURE----- |
64 |
|
65 |
-- |
66 |
gentoo-server@g.o mailing list |