powuh
join:2008-05-12
New York, NY
edit:
May 27th, @09:24AM
| reply to houkouonchi
Re: extremely high capacity speedtest
thanks guys. guess my routing to cali sucks 
TCP/Web100 Network Diagnostic Tool v5.4.12
click START to begin
Connected to: nitro.ucsc.edu -- Using IPv4 address
Checking for Middleboxes . . . . . . . . . . . . . . . . . . Done
checking for firewalls . . . . . . . . . . . . . . . . . . . Done
running 10s outbound test (client-to-server [C2S]) . . . . . 2.18Mb/s
running 10s inbound test (server-to-client [S2C]) . . . . . . 8.84Mb/s
The slowest link in the end-to-end path is a a 622 Mbps OC-12 subnet
------
TCP/Web100 Network Diagnostic Tool v5.4.12
click START to begin
Connected to: web100.rit.edu -- Using IPv4 address
Checking for Middleboxes . . . . . . . . . . . . . . . . . . Done
checking for firewalls . . . . . . . . . . . . . . . . . . . Done
running 10s outbound test (client-to-server [C2S]) . . . . . 594.26Mb/s
running 10s inbound test (server-to-client [S2C]) . . . . . . 114.67Mb/s
Server unable to determine bottleneck link type.
[S2C]: Packet queuing detected |
|
houkouonchi
join:2002-07-22
Corona, CA
clubs:
 ·Pacific Bell - SBC
 ·Charter Pipeline
·DSL EXTREME
| said by powuh  :thanks guys. guess my routing to cali sucks TCP/Web100 Network Diagnostic Tool v5.4.12 click START to begin Connected to: nitro.ucsc.edu -- Using IPv4 address Checking for Middleboxes . . . . . . . . . . . . . . . . . . Done checking for firewalls . . . . . . . . . . . . . . . . . . . Done running 10s outbound test (client-to-server [C2S]) . . . . . 2.18Mb/s running 10s inbound test (server-to-client [S2C]) . . . . . . 8.84Mb/s The slowest link in the end-to-end path is a a 622 Mbps OC-12 subnet Yeah that does kind of suck... Here is what I got from socal -> new york off my server on a 100 meg switch:
Testing network path for configuration and performance problems -- Using IPv4 address
Checking for Middleboxes . . . . . . . . . . . . . . . . . . Done
checking for firewalls . . . . . . . . . . . . . . . . . . . Done
running 10s outbound test (client to server) . . . . . 80.23 Mb/s
running 10s inbound test (server to client) . . . . . . 32.06 Mb/s
Server unable to determine bottleneck link type.
Information: Other network traffic is congesting the link
Information [S2C]: Packet queuing detected: 24.77% (local buffers)
Server 'web100.rit.edu' is not behind a firewall. [Connection to the ephemeral port was successful]
Client is not behind a firewall. [Connection to the ephemeral port was successful]
Packet size is preserved End-to-End
Server IP addresses are preserved End-to-End
Client IP addresses are preserved End-to-End
Although I am also running a web100 kernel on my box which helps with higher speeds at long distances but notice that my upload is faster than my download (like yours was to the NY server). Here is what I get to nitro.ucsc.edu:
Testing network path for configuration and performance problems -- Using IPv4 address
Checking for Middleboxes . . . . . . . . . . . . . . . . . . Done
checking for firewalls . . . . . . . . . . . . . . . . . . . Done
running 10s outbound test (client to server) . . . . . 78.25 Mb/s
running 10s inbound test (server to client) . . . . . . 81.54 Mb/s
The slowest link in the end-to-end path is a 100 Mbps Full duplex Fast Ethernet subnet
Alarm: Duplex Mismatch condition detected. Switch=Full and Host=Half
Server 'nitro.ucsc.edu' is not behind a firewall. [Connection to the ephemeral port was successful]
Client is not behind a firewall. [Connection to the ephemeral port was successful]
Packet size is preserved End-to-End
Server IP addresses are preserved End-to-End
Client IP addresses are preserved End-to-End
said by powuh :TCP/Web100 Network Diagnostic Tool v5.4.12 click START to begin Connected to: web100.rit.edu -- Using IPv4 address Checking for Middleboxes . . . . . . . . . . . . . . . . . . Done checking for firewalls . . . . . . . . . . . . . . . . . . . Done running 10s outbound test (client-to-server [C2S]) . . . . . 594.26Mb/s running 10s inbound test (server-to-client [S2C]) . . . . . . 114.67Mb/s Server unable to determine bottleneck link type. [S2C]: Packet queuing detected 600 meg upload, nice. As I figured this test gave you your best results (I am guessing this is the one on NY?). The download is kind of week but that could be a limitation of the server since its at a college and their downstream is probably more saturated than their upstream. When I got a 400 meg speed test it was upload and my download was like 160 meg (this was to nitro.ucsc.edu).
You might wanna try the test again at off-peak hours (if it wasn't when you ran it).
--
Chugging along on 2x 6016/768k DSL Extreme DSL lines and one 6016/768 ATT DSL DIrect line as well as one 10mb/1mb Charter cable line for a combined total of just over 26 meg download and 3 meg up (after overhead). yay! |
|
powuh
join:2008-05-12
New York, NY
| @houkouonchi:
for some reason, on speedtest.net and most other speedtests - either download and upload will be EXTREMELY high, and the other one will be significantly lower. (like speedtest.net gives me 300mbps down, 100mbps up, etc) - always one or the other. no idea why |
|
houkouonchi
join:2002-07-22
Corona, CA
clubs:
·Pacific Bell - SBC
·Charter Pipeline
·DSL EXTREME
edit:
May 29th, @02:16PM
| said by powuh :@houkouonchi: for some reason, on speedtest.net and most other speedtests - either download and upload will be EXTREMELY high, and the other one will be significantly lower. (like speedtest.net gives me 300mbps down, 100mbps up, etc) - always one or the other. no idea why well I don't really trust a flash applet for speeds at that level. Web100's NDT works for me easily even on 1 gigabit LAN and will give me results of 940 meg up/down. Speedtest.net has some servers and great connections but definitely not the best software setup to do speed tests. I have never gotten over 150 meg up to a speedtest.net server and my best (down) being like 520 meg and the upload was only like 20 meg to the Los Angeles server. I think the NDT ones simply are limited by what you can get to that server.
I would really suggest testing web100.rit.edu again but at like 2 or 3 in the morning if possible (off hours). I am sure they don't have a lot of excess download capacity during peak hours due to the fact its a school.
--
Chugging along on 2x 6016/768k DSL Extreme DSL lines and one 6016/768 ATT DSL DIrect line as well as one 10mb/1mb Charter cable line for a combined total of just over 26 meg download and 3 meg up (after overhead). yay! |
|
CoxCable4
banned from most servers for cheating
join:2002-10-02
PwnZone
·RoadRunner Cable
| according to rit's site they have already graduated so I would think around now before they really get into summer school would be the best time for a speedtest.
1gbps is pretty cool to have for work connect. What do they give you all that bandwidth for? |
|
powuh
join:2008-05-12
New York, NY | what if i said porn...
nah, not really. it's a server/colo place, so it's obvious there's bw laying around |
|
powuhWORK
@nuclearfallout.net | »www.speedtest.net/result/279339994.png
 |
|
a333
A hot cup of integrals please
join:2007-06-12
Corona, NY | I hate you. =\ |
|
