I think what MS and LinuxPPC are doing is great. For a long time now, the default install of many operating systems- Red Hat 6.0 included - has been very insecure. For instance, I believe you might be able to remotely attach to a default installation's X server and watch users enter passwords!
I'd like Red Hat to try to make their next release be secure by default - no Internet services turned on - and still have X working properly (maybe using Unix domain sockets?).
Another point: the point of those evidence rules is to keep anyone from being unfairly suprised. Yet the evidence being introduced is software that Microsoft published, and is therefore already aware of! Nothing unfair about that.
I agree. Case in point: a linux-oriented magazine recently ran an editorial slamming the Mindcraft test mercilessly, using hot-button words like "scam", "fiasco", "sham", "suckers", "propaganda", and "mind control", among others.
What was he trying to do, incite a riot? And why did he not want to admit that there might be performance problems in Linux?
IMHO it's more productive to turn the other cheek, and work on fixing our own faults. That's why my page on the Mindcraft benchmarks is relentlessly objective and focused on helping everyone, rather than cutting anyone down.
Check your road rage at the door, folks. One thing that helps me is, when I write an angry email, I let it sit for a while, and then when I'm calmer, I rewrite it until I'm sure everything in it is reasonably fair and true. It's so easy to write stuff quickly that isn't really accurate.
The Mindcraft benchmark, while flawed, did show real performance problems in Apache on Linux.
The worst performance problem was probably fixed by 2.2.7.
Another performance problem is scheduler overhead (the current design has trouble handling 200 running processes). A fix is in the works.
Dean Gaudet thinks mod_mmap_static should be able to give a significant performance boost.
Apache 1.x's process-per-client architecture probably prevents it from achieving the highest possible performance on static files.
The proper response to the Mindcraft challenge is to accept, but only after fixing the bottlenecks in the Apache/Linux combination that prevent it from achieving good SPECWeb96 scores. We shouldn't aim to beat IIS on NT; instead, we should aim to make Apache on Linux as good as it can possibly be. After all, SWS on Solaris beats IIS on NT on the same hardware; no need to set our sights too low:-)
See http://www.kegel.com/mindcraft_redux.html for support for the views stipulated above and continuing updates on the linux kernel team's response to the Mindcraft challenge.
Slashdot Mirror
I'd like Red Hat to try to make their next release be secure by default - no Internet services turned on - and still have X working properly (maybe using Unix domain sockets?).
Another point: the point of those evidence
rules is to keep anyone from being unfairly
suprised. Yet the evidence being introduced
is software that Microsoft published, and
is therefore already aware of! Nothing
unfair about that.
What was he trying to do, incite a riot? And why did he not want to admit that there might be performance problems in Linux?
IMHO it's more productive to turn the other cheek, and work on fixing our own faults. That's why my page on the Mindcraft benchmarks is relentlessly objective and focused on helping everyone, rather than cutting anyone down.
Check your road rage at the door, folks. One thing that helps me is, when I write an angry email, I let it sit for a while, and then when I'm calmer, I rewrite it until I'm sure everything in it is reasonably fair and true. It's so easy to write stuff quickly that isn't really accurate.
So far, I have some info on how Apache was configured. Kernel is next.
- The Mindcraft benchmark, while flawed, did show real performance problems in Apache on Linux.
- The worst performance problem was probably fixed by 2.2.7.
- Another performance problem is scheduler overhead (the current design has trouble handling 200 running processes). A fix is in the works.
- Dean Gaudet thinks mod_mmap_static should be able to give a significant performance boost.
- Apache 1.x's process-per-client architecture probably prevents it from achieving the highest possible performance on static files.
The proper response to the Mindcraft challenge is to accept, but only after fixing the bottlenecks in the Apache/Linux combination that prevent it from achieving good SPECWeb96 scores. We shouldn't aim to beat IIS on NT; instead, we should aim to make Apache on Linux as good as it can possibly be. After all, SWS on Solaris beats IIS on NT on the same hardware; no need to set our sights too lowSee http://www.kegel.com/mindcraft_redux.html for support for the views stipulated above and continuing updates on the linux kernel team's response to the Mindcraft challenge.