Adds a histogram to measure the difference between the bandwidth estimate at 2 seconds and 20 ...

Adds a histogram to measure the difference between the bandwidth estimate at 2 seconds and 20 seconds into a WebRTC call.

BUG=425925
R=asvitkine@chromium.org

Committed: https://crrev.com/7d6cf9d53051f186c64c4cc4661c3b241ebd71f0
Cr-Commit-Position: refs/heads/master@{#302672}

[Alexei Svitkine (slow), 2014-11-04 15:33:03]

https://codereview.chromium.org/682813004/diff/1/tools/metrics/histograms/his...
File tools/metrics/histograms/histograms.xml (right):

https://codereview.chromium.org/682813004/diff/1/tools/metrics/histograms/his...
tools/metrics/histograms/histograms.xml:39714: +    The difference between the
bandwidth estimate at 2 seconds and 20 seconds
Histograms generally can't have negative values.

So assuming the difference can go in either direction, a single histogram
wouldn't be enough to capture it.

[Stefan, 2014-11-04 15:34:59]

https://codereview.chromium.org/682813004/diff/1/tools/metrics/histograms/his...
File tools/metrics/histograms/histograms.xml (right):

https://codereview.chromium.org/682813004/diff/1/tools/metrics/histograms/his...
tools/metrics/histograms/histograms.xml:39714: +    The difference between the
bandwidth estimate at 2 seconds and 20 seconds
On 2014/11/04 15:33:03, Alexei Svitkine wrote:
> Histograms generally can't have negative values.
> 
> So assuming the difference can go in either direction, a single histogram
> wouldn't be enough to capture it.

The negative direction is the most interesting part, so in that case I can put a
threshold on it (<= 0) and take the absolute value. Will update this
description.

[Stefan, 2014-11-04 15:43:10]

On 2014/11/04 15:34:59, Stefan wrote:
>
https://codereview.chromium.org/682813004/diff/1/tools/metrics/histograms/his...
> File tools/metrics/histograms/histograms.xml (right):
> 
>
https://codereview.chromium.org/682813004/diff/1/tools/metrics/histograms/his...
> tools/metrics/histograms/histograms.xml:39714: +    The difference between the
> bandwidth estimate at 2 seconds and 20 seconds
> On 2014/11/04 15:33:03, Alexei Svitkine wrote:
> > Histograms generally can't have negative values.
> > 
> > So assuming the difference can go in either direction, a single histogram
> > wouldn't be enough to capture it.
> 
> The negative direction is the most interesting part, so in that case I can put
a
> threshold on it (<= 0) and take the absolute value. Will update this
> description.

ptal

[Alexei Svitkine (slow), 2014-11-04 18:14:25]

lgtm

[Stefan, 2014-11-04 19:39:35]

The CQ bit was checked by holmer@chromium.org

[commit-bot: I haz the power, 2014-11-04 19:40:16]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/682813004/40001

[commit-bot: I haz the power, 2014-11-04 21:22:23]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2014-11-04 21:23:25]

Patchset 3 (id:??) landed as
https://crrev.com/7d6cf9d53051f186c64c4cc4661c3b241ebd71f0
Cr-Commit-Position: refs/heads/master@{#302672}