Become a fan of Slashdot on Facebook

 


Forgot your password?
Close
typodupeerror
×
Chromium IT Technology

Chromium Cleans Up Its Act -- and Daily DNS Root Server Queries Drop by 60 Billion (theregister.com) 35

Posted by msmash from the things-improving dept.
The Google-sponsored Chromium project has cleaned up its act, and the result is a marked decline in queries to DNS root servers. From a report: As The Register reported in August 2020, Chromium-based browsers generate a lot of DNS traffic as they try to determine if input into their omnibox is a domain name or a search query. Verisign engineers Matthew Thomas and Duane Wessels examined the resulting traffic and reached the conclusion that it accounted for up to 60 billion DNS queries every day. Wessels has since penned a new post that went unreported when it appeared on January 7 -- the day after the US Capitol riot -- but was today resurfaced by APNIC, the Regional Internet Registry for the Asia-Pacific region. In the post he says the Chromium team redesigned its code to stop junk DNS requests, and released the update in Chromium 87. The result? "Before the software release, the root server system saw peaks of ~143 billion queries per day," he wrote. "Traffic volumes have since decreased to ~84 billion queries a day. This represents more than a 41 per cent reduction of total query volume."
This discussion has been archived. No new comments can be posted.

Chromium Cleans Up Its Act -- and Daily DNS Root Server Queries Drop by 60 Billion More

Chromium Cleans Up Its Act -- and Daily DNS Root Server Queries Drop by 60 Billion

Comments Filter:

  • Nice (Score:5, Interesting)

    by nospam007 ( 722110 ) * on Thursday February 04, 2021 @02:45PM (#61028170)

    It made me recall that HP color printer that somebody brought from home to the office (to print pretty Excel charts) and that immediately began to check the worldwide intranet for all the 1800 printers on 5 continents and checked if they might be out of paper or ink every couple of seconds and thus brought the whole system to a halt.

    • Subnet, bitches!

      • Re: (Score:2)

        by skids ( 119237 )

        Subnets are dying these days. Non-broadcast-multiple-access is the emerging paradigm (in various forms and the PHB acronyms of course change monthly). Just kill all the crappy multicast bullshit that thinks there's such a thing as a "local segment", and use broker services to put back whatever the users just cannot live without, with proper access controls to isolate users and groups.

        Eventually we will beat it into vendors to stop using multicast discovery protocols and work with router vendors to consult

  • Missing Version Info (Score:3, Insightful)

    by Anonymous Coward on Thursday February 04, 2021 @03:16PM (#61028350)

    Which versions of the various browsers that use Chromium have this update?
    Wouldn't that be really important to include?

    • Re: (Score:2, Informative)

      by nadass ( 3963991 )
      RTFA:

      The figure below highlights the significant decline of query volume to the root server system immediately after the Chromium 87 release. Before the software release, the root server system saw peaks of ~143 billion queries per day. Traffic volumes have since decreased to ~84 billion queries a day. This represents more than a 41% reduction of total query volume.

      • Re: (Score:1)

        by Anonymous Coward

        RTFA:

        The figure below highlights the significant decline of query volume to the root server system immediately after the Chromium 87 release. Before the software release, the root server system saw peaks of ~143 billion queries per day. Traffic volumes have since decreased to ~84 billion queries a day. This represents more than a 41% reduction of total query volume.

        RTF Parent Post:

        Which versions of the various browsers that use Chromium have this update?

        • RTF Grandparent's post:

          Chromium 87 release

          Check your about page, or do you think Edge for Windows has already had 88 individual releases since adopting Chromium code?

      • That info was in the summary, not even the article. Although this summary , like many others, is sadly mostly cut and paste from the article.

    • Several browsers using the Blink rendering engine, expandes to now be called a "browser engine" since the UI is also rendered by it etc.

      Edge, Chrome, Brave, and Chromium are all different browsers that use *Blink*. These browsers do not use Chromium, they use Blink.

      Chromium is just one browser that uses Blink.
      Chromium had this issue. Blink didn't have the issue, so it doesn't affect other browsers.

      • Several browsers using the Blink rendering engine, expandes to now be called a "browser engine" since the UI is also rendered by it etc.

        Edge, Chrome, Brave, and Chromium are all different browsers that use *Blink*. These browsers do not use Chromium, they use Blink.

        Chromium is just one browser that uses Blink.

        Incorrect. Those all use Chromium, which is both the name of the open source version of Chrome, but also the name of the wider toolkit around Blink that includes the HTTP stack, WebRTC, extensions and much more. Blink is only HTML rendering, and no project I know of uses it alone without the rest of the Chromium stack.

        • To compare with the WebKit, it is also both the name of the Safari beta test, and the name of the whole project as a toolkit. Blink is equivalent to WebCore in WebKit.

        • Interesting. Is that a recent change? The security list calls it Blink>WebRTC, the Chromium components list has webrtc as part of Blink, and the bugs go to bugs.chromium.org/p/chromium/issues/list?q=component%3ABlink%3EWebRTC That last bit decides to component: Blink > WebRTC Of course there is also WebRTC.org, which isn't part of Chromium.org. I'm told that administratively, everyone with a webrtc.org email address is in the Blink team at Google - they work for the Blink boss. Unless that has c

          • WebRTC has parts both in blink (DOM APIs) and out of blink (video decoding, RTC network transfer, etc).

  • How many DNS lookups could be saved by hardcoding the pornhub.com IP address?

    • Would this not bypass load balancing and result in degraded performance?

      And I just realized I typed "degraded performance" in reference to a porn site.

      • Comment removed based on user account deletion

      • Umm I was kidding, you know that right? But I guess you could hardcode a list of IPs and have the browser randomly select one when it connects. If many browsers did that it should help distribute the load.

  • Now if they'd just fix their gmail and calendar web clients...

    Both of them, open in a firefox tab on Ubuntu, are dispatching five-ish or more times per second - even if the window with their tab is closed.

    (What ARE they doing, for a company that makes billions selling information it mines from its users?)

    They also cache enough stuff locally and don't flush it so that, if left running for a day or two on a machine with a "small" (4G!) amount of physical RAM, they shove everything else out to the paging store in a day or so. When I come back the next day the machine is comatose for minutes, while it brings back enough working set to even notice the mouse and keyboard.

    • If it's Firefox specific you can submit a bug report to Firefox's bug reporting site https://bugzilla.mozilla.org/ [mozilla.org]

      When I don't use my internet I set my firewall to block everything. It's basically like pulling the network cable to the internet router. If you don't have that feature then just turn off the internet router.

      Oh boy. 4 gb ram. That's claustrophobicly little ram. How old is your computer? I haven't had that little ram in ages. I've got 12 gb and even that is not enough. My next build will have at

Slashdot Top Deals

Don't tell me how hard you work. Tell me how much you get done. -- James J. Ling

Close