Chrome (Chromium) is the most popular viewer today, but it is known to be a big consumer of memory. However, the new edition will come with a number of consumption-optimizing optimizations.
Today, Chrome is based on measured statistics on websites on every second computer. His closest competitor, Internet Explorer, has a half representation. In spite of all the advantages, Chrome has a long memory consumption problem and many users do not like it.
In practice, the testing of news sites is done by loading the page, clicking the first link, waiting for full load, and returning. Then he clicks on the next link and repeats this activity several times. In the case of a multimedia content site, the video will start, run for a few seconds, and then select the next item to watch. This activity is repeated several times.
The benchmarks created in this way can be repeated in the test environment during development to track the impact of optimizations on different versions of the browser. The developer has published his tests, so anyone can repeat them.
Based on memory allocation, it is possible to return to the code to see if the individual structures behave as expected or there is a possibility of optimization. We used this tool to detect low efficiency in many data types, writes developers.
There is a direct link between garbage collector intensity, latency and memory consumption. For example, the use of a garbage collector may be reduced in such a way that the application receives more memory - it is not necessary to clean it so often and the user does not notice delays. For devices with a small amount of memory (such as cheap phones with 512 MB RAM), however, excessive spreading would result in an out-of-memory killere and the inaction of a greedy application.
Developers have heuristics that dynamically decides what strategy they choose when using the garbage collector. After the cleaning is completed, the statistics of the living objects are executed and the environment determines when the next cleaning is planned. If it decides to save the memory, it will begin to plan the collector run more often.
The scripts are being processed in the background when the page is loaded. Using memory visualization, developers have discovered that this process leaves the tagged zone even after the compilation has finished. A new measure has been taken to cause the zone to be canceled immediately after the compiler is complete. This leads to overall lower memory consumption and, above all, smaller tops.
Developers promise not to complete this area and plan further zone optimizations. It will focus on devices that have between 512 MB and 1 GB of memory.
EUR 9 / Mo
EUR 135 / Mo