On Tue, 2011-02-22 at 11:54 -0700, Alex Williamson wrote:
> This series introduces a new weight-balanced binary tree (wbtree) for
> general use. It's largely leveraged from the rbtree, copying it's
> rotate functions, while introducing different rebalance and erase
> functions. This tree is particularly useful for managing memory
> ranges, where it's desirable to have the most likely targets (the
> largest ranges) at the top of each subtree.
>
> Patches 2 & 3 go on to convert the KVM memory slots to a growable
> array and make use of wbtree for efficient managment. Trying to
> exercise the worst case for this data structure, I ran netperf
> TCP_RR on an emulated rtl8139 NIC connected directly to the host
> via a tap. Both qemu-kvm and the netserver on the host were
> pinned to optimal CPUs with taskset. This series resulted in
> a 3% improvement for this test.
Marcelo asked about kernbench with ept=0 on this series. Using a 4-way,
10G guest with pinned vcpus, build in tmpfs, 10x optimal run, I got:
before (stdev) after (stdev) %
--------+-----------------+----------------+-------
Elapsed | 42.809 (0.19) | 42.305 (0.23) | -1.18
User | 115.709 (0.22) | 114.720 (0.31) | -0.85
System | 41.605 (0.14) | 40.924 (0.20) | -1.64
%cpu | 366.9 (1.45) | 367.6 (1.51) | 0.19
context | 7272.3 (68.6) | 7249.5 (97.8) | -0.31
sleeps | 14826.2 (110.6) | 14798.5 (63.0) | -0.19
So, a small but measurable gain. Also, sorry for forgetting to address
Marcelo's comments from the original version of patch 2/3, I'll pick
them up in the next round. Thanks,
Alex
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
