Chrome is testing larger cache data to speed up the performance

In the past, Google Chrome put all cached data in the same bucket. When the web page is loaded, the browser will retrieve in advance whether there is a cache in the bucket.

If there is a cache, there is no need to load repeatedly, which can improve the loading efficiency, but all cached data in the same place may cause side-channel attacks.

For security reasons, Google Chrome has used a separate storage bucket for each website in previous versions, which can directly avoid potential side-channel attacks.

But the disadvantage is that this scheme will affect the actual performance of the browser, because duplicate data cannot be shared and must be reloaded, so performance degradation is inevitable.

In order to make up for the performance loss, Google Chrome is currently testing to increase the cache size in the beta version, increasing the original cache limit to 2 times, 2.5 times, or even 3 times.

According to Google engineers, increasing the cache size can prevent part of the cache from being automatically cleaned up, so that duplicate resources can be avoided from reloading through the network.

Of course, it is not clear for now whether this improvement is effective, but as a test, Google will directly enable this feature in the beta, development, and canary versions.

So when the new improvement is approved, it will be directly pushed to the user, and it does not need to be turned on through experimental options or other special settings.

It is worth noting that the disk space occupied by Google Chrome will increase significantly after enabling this function because more cache files will not be automatically cleaned up by Google.

The purpose of not cleaning up is to prevent too many resources from requiring frequent reloading. Obviously, users should not use cleaning software to clean up too frequently.

Appropriate cache files can help improve browser performance and improve loading efficiency, etc. Frequent cleaning of the cache will only waste your time and slow performance.

Via: bleepingcomputer