summaryrefslogtreecommitdiffstats
path: root/chrome/browser/browser_main.h
diff options
context:
space:
mode:
authorziadh@chromium.org <ziadh@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98>2010-08-02 16:34:26 +0000
committerziadh@chromium.org <ziadh@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98>2010-08-02 16:34:26 +0000
commit78a258a24ae6d4b764b5616cc639367fb3a573b7 (patch)
tree87dbb4ae0e2465b29c4829961490274a8f8cd803 /chrome/browser/browser_main.h
parent695092fd7614c257866347f049e28ec327d7a3d3 (diff)
downloadchromium_src-78a258a24ae6d4b764b5616cc639367fb3a573b7.zip
chromium_src-78a258a24ae6d4b764b5616cc639367fb3a573b7.tar.gz
chromium_src-78a258a24ae6d4b764b5616cc639367fb3a573b7.tar.bz2
A/B test for measuring effect of number of proxy connections.
Currently, we limit the number of connections per proxy server to 32 (raised from 15). Firefox uses 8. A low value keeps from overloading the proxy servers; at the same time, some users may hit the limit (since we use the same number for both HTTP and SOCKS proxies). This experiment selects a different number at browser launch (from the set [8, 16, 32, 64]), and attempts to look at certain stats to get a clearer picture of how this affects the user. BUG=44501 r=jar Review URL: http://codereview.chromium.org/3033045 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@54541 0039d316-1c4b-4281-b951-d872f2087c98
Diffstat (limited to 'chrome/browser/browser_main.h')
-rw-r--r--chrome/browser/browser_main.h3
1 files changed, 3 insertions, 0 deletions
diff --git a/chrome/browser/browser_main.h b/chrome/browser/browser_main.h
index f7a5a80..6ef2bbe 100644
--- a/chrome/browser/browser_main.h
+++ b/chrome/browser/browser_main.h
@@ -107,6 +107,9 @@ class BrowserMainParts {
// A/B test for determining a value for unused socket timeout.
void SocketTimeoutFieldTrial();
+ // A/B test for the maximum number of connections per proxy server.
+ void ProxyConnectionsFieldTrial();
+
// A/B test for spdy when --use-spdy not set.
void SpdyFieldTrial();