As part of the Covad Blogger"s program, Andy setup several industry bloggers with a Covad T1. Greg, Phoneboy, and Ken Camp, have all covered their experiences with using a Covad Internet T1 connection. On a related note, Rich Tehrani had an interview with Covad"s Jeff Ahlquist, Vice President of Corporate Development and Strategy that is definitely worth checking out.Anyyway, last week, I had my T1 installed and configured. One of the first things I did was perform a bandwidth speed test. I went to http://www.speakeasy.net/speedtest/ and ran an online bandwidth test. First the download test ran and I got 1,450Kbps, which is pretty close to a standard-rated 1.544Mbps (1544Kbps) T1 connection. Then the upload (upstream) test ran and to my shock I was only getting 444Kbps. This isn"t nearly as fast as it should be. I tried another speed test website and got similar results.
I fired off an email to Andy explaining my benchmark results and asked him if I received a "true" T1. He responded that it was a true T1 and that there was no way my upstream should be that slow. He suggested I contact Covad technical support, but before I did that, I decide to play around with my Linksys WRT54G router.
Just by chance I happened to click the Applications & Gaming menu in the Linksys Web-based admin and noticed a QoS tab. I clicked on the QoS tab and noticed I had QoS turned on and configured to give my now defunct Cisco ATA-186 "high priority". I set my ATA-186 to "high priority" over a year ago when I was having VoIP quality issues with Vonage. I forgot I had QoS turned on, but surely having QoS enabled wouldn"t affect my upstream bandwidth, would it? The Linksys router should be smart enough to see that there is no Vonage device with the MAC address specified currently receiving/transmitting packets, and therefore the router should give all the bandwidth to any application that wants it. Seems logical, right?
Wrong! I turned off QoS and then ran another bandwidth test and voila, I had symmetric speeds of about 1,450Kbps. Actually, the upstream was slightly faster than the downstream on my first test with 1,468Kbps vs. 1,450Kbps. I joked with Andy, "Well, it is a cheap residential WRT54G router, so I guess I can’t expect commercial-grade QoS with ToS bits, etc." Here"s a screenshot of the Linksys QoS setup screen showing how I had the Vonage ATA device set to high priority. (click for Larger Image)
Moral of the story - If you want maximum performance leave QoS turned off if using a Linksys router. Seems counter-intuitive doesn"t it?
However, if I still had Vonage and used a bandwidth hog such as P2P client (eMule, Bittorent, etc.) I would need to leave QoS turned on so the voice packets get priority. It would appear as a side effect I would lose roughly 1,000Kbps of upstream bandwidth, which really sucks. Maybe Linksys has new firmware for the WRT54G, which resolves this issue. I"m running v3.03.6 currently.
In any case, I did one other test to check the latency of the Covad T1, which is actually more important than raw bandwidth if you"re doing VoIP or videoconferencing. As a benchmark test, I pinged Yahoo.com and TMC"s T3 router. Here are the results of an approximately 10min ping test:
TMC T3 Router
Packets: Sent = 1075, Received = 1075, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 10ms, Maximum = 921ms, Average = 100ms
Yahoo!
Ping statistics for 216.109.112.135:
Packets: Sent = 1017, Received = 1016, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 13ms, Maximum = 37ms, Average = 18ms
Pretty impressive latency numbers, especially when pinging Yahoo.com since it"s maximum latency was 37ms and it"s average was an amazing 18ms. It would appear that Covad sits very close to where Yahoo! connects to the Internet. Obviously, since Covad is a carrier they sit near the main fat pipes. Yahoo gets massive traffic, so I"m sure they connect just a few short hops from the fat pipes. I"m sure both have multiple connections across the country and of course Covad has Tier-1 peering agreements, which guarantees that cross-country traffic is super fast.
Overall, I"m very happy with my Covad T1 experience. The installer was friendly and courteous. The latency is excellent and the bandwidth, in particular the upstream is much better than my DSL. I asked the Covad installer about Covad offering fiber in the near future, since Verizon is quickly deploying their FiOS service and AT&T is going gangbusters with their U-verse service. Both of these competitors can provide Triple Play packages (voice, video, data).
He mentioned he was aware of Verizon FiOS and that Covad currently doesn"t offer a fiber solution, but hinted that FiOS was helping to push Covad to provide fiber in the very near future. I think AT&T and Verizon are starting to drive some of the Tier-2 carriers to provide fiber to the home (FTTH) or provide a hybrid-fiber/copper approach like AT&T"s U-verse is taking. And that my friends is great news for residential customers which are clamoring for more bandwidth and more services at an inexpensive price point.
TrackBack (0) | Comments (1) | Tag with del.icio.us | VoIP & Gadget Blog | Permalink: Covad T1 Speed and Latency Test Plus Overview
Tags: bandwidth, Covad, latency, Linksys, QoS, speed test, T1, Vonage, WRT54GCopyright VoIP & Gadgets Blog
>>Source Link
>>Blog: VoIP & Gadgets Blog
>>Publish Date: 6/21/2007 7:12:58 AM
>>Keywords: covad test
Related Posts
>>Known-Issues and workarounds for Orcas (Beta1): Inheritance #
Issue #1: Inheritance of Test Classes is not supported when inheritance spans classes defined outside of the active test project WorkaroundPlace all the classes which will use inheritance in the sa
>>Coded Web tests (and Web test plug-ins) should not block the thread #
If you are using Coded web tests (or Web test plug-ins with a declarative Web test) and running the Web test in a load test, you should avoid doing anything in your code that will cause the running th
>>New Unit Testing Features in Orcas (Part 1) #
This is a part of a series of postings on the new unit testing features appearing in Orcas. All these features will be available in Orcas beta 2. I know that MSTest isn’t yet the darling of the TDD w
>>How to use a Hosts file to test a site that uses host headers on an Intranet (and Internet) #
This can be used to test your site changes without publishing to a remote server. http://support.microsoft.com/kb/228760/en-us Share this post: email it! | bookmark it! | digg it! | reddit! | kic
>>Ben on MbUnit"s Row Test #
Ben Hall has a short and sweet post on MbUnit's row test, it includes some sample code for you to try out. Share this post: email it! | bookmark it! | digg it! | reddit! | kick it! | live it
>>MbUnit test duration and using row tests with VB2003 #
Two great posts from the MbUnit community.Julian on using MbUnit row tests with VB 2003.Vadim on setting test duration with MbUnit. Share this post: email it! | bookmark it! | digg it! | reddit! |
>>Power Tool Task update #
Buck posted a new cool update of the Power Tools for Visual Studio 2005 (Team Test and Team System) and Team Foundation Server. Just a little note: "Unlike the test container feature, the test categor
>>Update to "How to run tests in a build without test metadata files and test lists (.vsmdi files)": Test Categories #
Our good friend Buck Hodges (TFS TeamBuild) has worked with some internal enthusiasts to produce a great power tool to enable more ways of selecting tests to run with a build. You can read about it o
>>WCAT 6.3 web performance and scalability test tool released #
The NT Performance team just publically released the latest version of WCAT (Web Capacity Analysis Tool), the web performance test tool that the IIS team and NT Performance team use to conduct interna
>>WCAT 6.3 web performance and scalability test tool released #
The NT Performance team just publically released the latest version of WCAT (Web Capacity Analysis Tool), the web performance test tool that the IIS team and NT Performance team use to conduct interna