From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-ia0-f171.google.com (mail-ia0-f171.google.com [209.85.210.171]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (Client CN "smtp.gmail.com", Issuer "Google Internet Authority" (verified OK)) by huchra.bufferbloat.net (Postfix) with ESMTPS id DAA6121F12C for ; Thu, 13 Dec 2012 08:28:08 -0800 (PST) Received: by mail-ia0-f171.google.com with SMTP id k27so2122859iad.16 for ; Thu, 13 Dec 2012 08:28:08 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:x-originating-ip:date:message-id:subject:from:to :content-type:x-gm-message-state; bh=+cJ54pfLyMa/L2Q0EPtJ8HYoailxjy8p9/p7Da7NhsA=; b=ghuhty9EXPERoNO+5IcFJHW4wC5Ad2NsEPuT9dHe9r0CfGDnxdU9IdIMY/ZzW4KCPm V1vKZ9KszJlmwt07mOIRa1En96c8gP1A82NX/PF6DBUcvz5CZ7URSLGBJwLo3NNavtBD eAWCo3bn/Y9RfUMgY4wW/oyod6iUA+PfV5hCxtUzXuVanZrbv915OWOZH212iEXjFebY hk9Fl2Hjpu0MyFQOU9W+v4jb3y/lXfuOOGkGr/1tp1jJ15dQOgRxSUcjkMEgvO6sVwMi hBcBOenVRlx+z0Wbqe4W0k3CO80h/0+0r+W8wyzAdCm3Txn+Vm88+WkDOFgIOdMvKaVr QUIA== MIME-Version: 1.0 Received: by 10.50.194.131 with SMTP id hw3mr2135249igc.71.1355416087754; Thu, 13 Dec 2012 08:28:07 -0800 (PST) Received: by 10.64.68.137 with HTTP; Thu, 13 Dec 2012 08:28:07 -0800 (PST) X-Originating-IP: [198.28.69.5] Date: Thu, 13 Dec 2012 17:28:07 +0100 Message-ID: From: Maciej Soltysiak To: cerowrt-devel@lists.bufferbloat.net Content-Type: multipart/alternative; boundary=14dae9340f796ef93c04d0be66a7 X-Gm-Message-State: ALoCoQmEXVNLP+4efv8fJhVinHBCkH/qEeCXhWqh8u1ORn72l1QToyYNd58rouuK+NSsDQml4M4b Subject: [Cerowrt-devel] How to address IT Pros to the issue? X-BeenThere: cerowrt-devel@lists.bufferbloat.net X-Mailman-Version: 2.1.13 Precedence: list List-Id: Development issues regarding the cerowrt test router project List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 13 Dec 2012 16:28:09 -0000 --14dae9340f796ef93c04d0be66a7 Content-Type: text/plain; charset=UTF-8 Hello, Real Example #1: "We have a 5Mbps CIFS partition and XXX was saturating it so no one could print, use office applications (e.g. Word would hang trying to get printer parameters across the network), network shares, anything that uses CIFS basically. We asked the network profiling guys to investigate and they came back with XXX as the culprit." Basically here they had an application that saturated the whole 5 Mbps of their Class of Service - The XXX application was made the root-cause, not how the queues were utilized. Real Example #2: Introduction of XXX application to "Reduce bandwidth utilization, typically by 60-95 percent, across all TCP-based applications. Free up bandwidth for other applications, like VoIP or Citrix, so they can perform better." Here the problems for VoIP and Citrix are perceived in bandwidth, so instead of working to reduce latency, they implement TCP mangling gizmos to compress/coalesce/whatever traffic to reduce bandwidth used. Probably the good result they are having is not due to bandwidth use reduced but simply from the amount of packets on wire and less to manage on the droptail queues. These 2 examples are real world IT Pros talking. Enterprise might be a huge pain to debloat! Regards, Maciej --14dae9340f796ef93c04d0be66a7 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
Hello,
=C2=A0
Real Example #1:
"We have a 5Mbps CIFS partition and XXX was= saturating it so no one could print, use office applications (e.g. Word wo= uld hang trying to get printer parameters across the network), network shar= es, anything that uses CIFS basically. We asked the network profiling guys = to investigate and they came back with XXX as the culprit."
=C2=A0
Basically here they had an application that satu= rated the whole 5 Mbps of their Class of Service - The XXX application was = made the root-cause, not how the queues were utilized.=
=C2=A0
Real Example #2:
Introduction of XXX application to "Reduce bandwidth utilization, typically= by 60-95 percent, across all TCP-based applications. = Free up bandwidth for other applications, like VoIP = or Citrix, so they can perform better."
=C2=A0
Here the problems for VoIP and Citrix are percei= ved in bandwidth, so instead of working to reduce latency, they implement T= CP mangling=C2=A0gizmos to compress/coalesce/whatever traffic to reduce ban= dwidth used. Probably the good result they are having is not due to bandwid= th use reduced but simply from the amount of packets on wire and less to ma= nage on the droptail queues.
=C2=A0
These 2 examples are real world IT Pros talking.= Enterprise might be a huge pain to debloat!
=C2=A0
Regards,
Maciej
=C2=A0
--14dae9340f796ef93c04d0be66a7--