Solved

Connecting 6 computers to a site through one connection - whats more important, UP or DOWN?

Posted on 2004-10-02
7
200 Views
Last Modified: 2010-03-17
Hi

I am working on a project that will require 6 computers to connect to one web site on a certain day, then all at the same time hit a certain button - a 'bid' basically. The first computers to connect wins type of thing. I know that the website goes by timestamp and login name, I have 6 different login names, and I need the timestamps to be sent as quickly as possible. The person with the fastest connection gets in first basically.

So, right now Im on a RoadRunner cable connection with a 3mb download and 328kb upload. I want to increase the upload speed because I will not have just 1 or 2 computers, but 6 - I have the option to upgrade our line here to a 6mb download and a 512kb up, or a 2mb both ways, or a 3mb down and a 1.5mb up. (theres a few more options, but they dont seem plausible)

Im worried about many things:
1. Should upload be more than download, or vice-versa?
2. I am connecting them through a router/switch, will they have to take 'turns' sending the information, no matter what I do?
3. Will my operating systems have anything to do with bottleneck and/or computer hardware?


The people will log-in, wait for the dynamic clock to show a certain time, they will hit a button, a graphic question will be sent to them (to stop bots from doing it) and once they answer the question they qill submit the info.

It's very very vital I get this correct the first time around, my job depends on it. lol

Any help is greatly appreciated!
0
Comment
Question by:StaindSou11
[X]
Welcome to Experts Exchange

Add your voice to the tech community where 5M+ people just like you are talking about what matters.

  • Help others & share knowledge
  • Earn cash & points
  • Learn & ask questions
7 Comments
 
LVL 11

Accepted Solution

by:
infotrader earned 250 total points
ID: 12207964
1.  Theoretically, upload is more important than download in this particular case.  HOWEVER, I doubt that the information you are transferring is so big that it'd make too much difference.

2.  Depends, but I think most router/switches sends out packet in a first-come-first-server basis (I could be wrong, but that's my impression).

3.  Yes...  It's not necessarily your O.S. per se, but what's loaded.  For example, if you use any kind of spyware sweeping utility or anything that analyze the packets, then you'd have to go through the software, get the packet analyzed, etc., before sending out the info.  Even something like Antivirus would effect the performance.

However, I think your biggest challenge is not the 3 items mentioned above, but latency.  It doesn't matter how fast your bandwidth is, but how fast can it response...  Basically, what matters the most is how many hops between your computer and the server.  It doesn't matter if you have a 100Mbps connection going from your computer to the Internet, if you have to go through 20 hops just to get to the destination, you are still going to get beat by a slower-connection computer that only has 5 routers between the server and the computer...

- Info
0
 

Author Comment

by:StaindSou11
ID: 12208107
I appreciate the comment.

I also agree, I don't think the size of the information I will be sending will matter too much, so 'I' (in my complete lack of experience) think the 6mb down and 512k up will be fine BUT now I need to worry about 'waiting in line' - can you (or anyone, dont want to pressure you) offer some advice on how I can check 'waiting in line' or if there are solutions to get around that (spare no expense really).

I tried setting up a test by taking two of the computers on the current connection, opening cmd up on them and then running ping (insert site) and pressing enter at the same time. Results:
One always waits for the other... which isnt good if that's how it's going to be.

Any more suggestions on testing this situation? and any ideas on how to get around it? Or anymore input on anything I might be missing?

The information is GREATLY appreciated and really means a lot to me (it means my job actually)

thanks!
-Staind

0
 

Author Comment

by:StaindSou11
ID: 12208155
btw on the Latency part of it -

We had tested tracert to a site that we will be doing this action from (not the main one though, because it wont let us) and they had 16 hops (it took over 2k ms to get to each one though) and ours here had 17 (and we took 57ms to get to each one, average) Will this be a problem also?

I know Im asking of quite a bit, but I have until Monday to make this decision, otherwise my job will be at stake.

All Im asking is for a little information to even just point me in the right direction,

again thanks in advance

~Staind
0
Industry Leaders: We Want Your Opinion!

We value your feedback.

Take our survey and automatically be enter to win anyone of the following:
Yeti Cooler, Amazon eGift Card, and Movie eGift Card!

 
LVL 79

Assisted Solution

by:lrmoore
lrmoore earned 250 total points
ID: 12212692
>I tried setting up a test by taking two of the computers on the current connection, opening cmd up on them and then running ping (insert site) and pressing enter at the same time. Results:
One always waits for the other... which isnt good if that's how it's going to be.

The fallacy of that test is that when you're talking in milliseconds for one packet to win the race to the output queue, humans are simply not capable of doing anything at exactly the same time. There are subtle differences between the two systems that you are using as test clients that can also change the dynamics of which one wins the race to the queue.

Packet serialization delay and network latency has much more to do with how long the "wait in line" is than does bandwidth, either up or down. No matter how you change your bandwidth parameters, you will still have the same physical media that determines the serialization. Change from T1 to cable, change from cable to DS3, Change from DS3 to Gigabit Ethernet, etc, and you change the serialization delay of your local network egress point.

http://www.dtool.com/delay.html
http://www.more.net/services/bandwidth/documentation/qos-bp/plan.html
0
 
LVL 40

Expert Comment

by:Fatal_Exception
ID: 12213885
I wonder more at the person who really believes that such a test could be definitive...   There are just too many factors involved here that could affect the outcome.  Even collisions could skew your data.    I would go to your super or director and lay out the reasons that this is not a reliable test, and the possibilities of failure are high from the start...  I would certainly not wager my job on the success of this...

FE
0
 

Author Comment

by:StaindSou11
ID: 12217715
OK, thanks everyone for your help - we're going with 6mb down/ 512k up. I will also be getting a new router/switch to remap our local network to decrease serialization and local latencies.

FE - My job depends on the success of the project, not of the little test, lol - I would have to agree with you totally if it did

Thanks

~Staind
0
 
LVL 40

Expert Comment

by:Fatal_Exception
ID: 12217776
Then I wish you the best with the overall project..!!  In this respect, have a great week..

FE
0

Featured Post

Industry Leaders: We Want Your Opinion!

We value your feedback.

Take our survey and automatically be enter to win anyone of the following:
Yeti Cooler, Amazon eGift Card, and Movie eGift Card!

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

Cable Modem Provisioning from DPoE compliant server  This Article is to support CMTS administrators to provide an overview of DOCSIS compliance configuration file, and to provision a cable modem located at customer place from a Back office serve…
Sometimes you have to pull out old tricks to get a new firewall to work… While we were installing a new Sonicwall at a customers site we found that sites they were able to visit before were not working.  It seemed random and we could not understa…
This video Micro Tutorial shows how to password-protect PDF files with free software. Many software products can do this, such as Adobe Acrobat (but not Adobe Reader), Nuance PaperPort, and Nuance Power PDF, but they are not free products. This vide…
If you’ve ever visited a web page and noticed a cool font that you really liked the look of, but couldn’t figure out which font it was so that you could use it for your own work, then this video is for you! In this Micro Tutorial, you'll learn yo…

734 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question