Explorer, Chrome

This might be an inform toward we published in October 2013. Because web log, we compared performance of the Salesforce Console in several browsers: web browser 7/8/9, Chrome and Firefox. When you look at the releases considering that the 2013 web log, we now have invested considerable work to boost overall performance associated with Salesforce Console. Specifically starting summertime 2015 release, we improved our overall performance examination abilities because of the next improvements:

• Improved test protection: we currently frequently test console overall performance for several different variants of typical usage situations. For instance, with and without sidebar elements, with and without MRU, major tabs plus subtabs, different types of user activities including starting one loss at the same time versus starting a number of tabs in addition, and tested different organizations including Account, Case, chance, contribute, and Contact.
• Beginning summertime 2015 release we now have set up a system overall performance test automation framework with allowed us to run nightly examinations to cover an increased few test situations on Virtual Machines against six browsers: IE8, IE9, IE10, IE11, Chrome and Firefox. Before summertime 2015 release gaming console performance examinations had been performed by hand or semi-manually where overall performance examinations were automatic with a script nevertheless the collection of memory information had been done by hand.
• The enhanced test coverage and automated tests have actually decreased the full time to breakthrough and quality of performance issues substantially.

Under is an example of circumstances chosen from nightly console overall performance automation operates.

Memory Qualities

To test the memory characteristics of every browser, we simulated various workloads. We simulated a standard workload that constituted open and shutting tabs, then tried an identical workload utilizing sub-tabs, and lastly worry tested a workload that has been according to bursts of activity.

Scenario 1: Simulate a Tab-level Work

Within situation we have a fundamental web page layout with no system sidebar elements. Into the test one Account tab is exposed, and closed after 20 moments. The procedure will be repeated for 100 loops, with an overall total range 100 tabs started and closed. The aim of this test is simulate conversation behavior for an average workload at major tab level. Under is the plot of measured web browser memory impact versus the number of tabs becoming exposed:

As well as be seen from the land, Chrome performed best for this scenario, using least expensive web browser memory footprint one of the browsers. Firefox features a slightly greater memory footprint but presented regular during test run, so is IE11. IE10 shows a moderate increase in memory footprint and IE9 may be the worst one of the 5 browsers, seeing a more steep boost throughout the run.

Situation 2: Simulate a Sub-tab Work

This situation is comparable to situation # 1 overhead except that in the place of main loss, our company is opening Account entity as sub-tabs inside test. The aim of this test should simulate communication behavior for a normal workload at sub-tab loss level.

Here we can also see Chrome and Firefox have most useful performance, followed closely by IE10 and IE11, while IE9 gets the biggest rise in memory footprint.

Scenario 3: Stress Testing Bursts

In this scenario, we start 10 Account tabs at a time, and close them after 20 moments. The action is then repeated for 30 loops, with a total few 300 tabs launched and closed at the end of the test. The goal of this test will be worry test interaction behavior for a bursty workload at main loss degree.

We could see Chrome is once again the champion, where memory impact held steady throughout the test, while IE9/IE10/IE11 all see significant upsurge in browser memory impact. The “bursty” nature of tab orifice (opening several tabs quickly within a rather little while of time) generally seems to trigger Firefox memory impact to fluctuate commonly, which should want to do with the certain garbage collection mechanism followed because of the browser.

Console First View

In addition to browser memory impact, within the 2013 web log we in addition sized “console very first view” timing plus the time it requires to weight a loss after several (for example, 300) of tabs have been exposed and shut. The matching dimensions done for the Winter 2016 launch are as follows:

The 301st tab available time (after opening and closing 300 tabs)

Note that internet browser memory impact growth has actually overwhelmingly already been the principal element affecting gaming console overall performance. In the event that web browser memory footprint expands to a tremendously large value, then your time it will take the browser to complete trash collection may considerably grow, which will impact the responsiveness skilled by the user. In order long as internet browser memory footprint continues to be little, the console performance to expect is fairly fast.

Conclusion

From automatic tests run nightly against a number of scenarios representing typical console consumption, we can see Chrome remains the internet browser of choice. It handles browser memory better than other browsers tested, as well as the overall memory impact continues to be in a relatively low range.

Note that you will find many elements influencing the memory impact readings of every provided test therefore the readings provided preceding aren't straight comparable with the numbers from 2013 web log. Even though whenever tested in the same environment, gaming console overall performance numbers do show an important improvement in more recent releases (summer time 2015 and after) when compared with previous releases.