Quoting Eric W. Biederman ([email protected]):
> "Serge E. Hallyn" <[email protected]> writes:
>
> > This patchset introduces a per-process utsname namespace. These can
> > be used by openvz, vserver, and application migration to virtualize and
> > isolate utsname info (i.e. hostname). More resources will follow, until
> > hopefully most or all vserver and openvz functionality can be implemented
> > by controlling resource namespaces from userspace.
> >
> > Previous utsname submissions placed a pointer to the utsname namespace
> > straight in the task_struct. This patchset (and the last one) moves
> > it and the filesystem namespace pointer into struct nsproxy, which is
> > shared by processes sharing all namespaces. The intent is to keep
> > the taskstruct smaller as the number of namespaces grows.
>
>
> Previously you mentioned:
> > BTW - a first set of comparison results showed nsproxy to have better
> > dbench and tbench throughput, and worse kernbench performance. Which
> > may make sense given that nsproxy results in lower memory usage but
> > likely increased cache misses due to extra pointer dereference.
>
> Is this still true? Or did our final reference counting tweak fix
> the kernbench numbers?
>
> I just want to be certain that we don't add an optimization,
> that reduces performance.
Here are the numbers with the basic patchsets. But I guess I should
do another round with adding 7 more void*'s to represent additional
namespaces.
(intervals are for 95% CI, tests were each run 15 times)
| with nsproxy | without nsproxy |
kernbench | 68.90 +/- 0.21 | 69.06 +/- 0.22 |
dbench | 386.0 +/- 26.6 | 388.4 +/- 21.0 |
tbench | 391.6 +/- 8.00 | 389.4 +/- 10.95 |
reaim with nsproxy
1 115600.000000 5512.441557
3 246985.712000 9375.780582
5 272309.092000 8029.833742
7 290020.000000 7288.367116
9 298591.580000 5557.531915
11 nan nan
13 nan nan
15 nan nan
reaim without nsproxy
1 110160.000000 5728.697311
3 246985.712000 9375.780582
5 262204.197333 11138.510652
7 288660.000000 6880.898412
9 300631.580000 4351.926692
11 nan nan
13 nan nan
15 nan nan
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
[Index of Archives]
[Kernel Newbies]
[Netfilter]
[Bugtraq]
[Photo]
[Stuff]
[Gimp]
[Yosemite News]
[MIPS Linux]
[ARM Linux]
[Linux Security]
[Linux RAID]
[Video 4 Linux]
[Linux for the blind]
[Linux Resources]
