How to properly test throughput / capacity of FTTC with BT

[cpcnw]

Hi All,

I need to make some kind of assessment of FTTC with BT and compare it to a clients current services.

As I understand it speed tests alone aren't too useful when it comes to this kind of task because real world use of small networks consists of say 10-20 people all downloading / uploading data, from many different locations, at different times of the day and sometimes at the same time.

I see words like 'bandwidth', 'throughput' and 'capacity' used in these type of discussions although I think I am right in saying that mostly these refer to how much data you can simultaneously move up/down a link on a continuing basis and how that might vary with times of day, or other factors like how many other users are on a particular circuit / exchange etc.

My guess is I would have to recreate two way traffic continuously over say 24-48 hours, during peak and off-peak and somehow collate and make sense out of that data?

Any advice / pointers welcomed

 

[Mark.J]

I'm not 100% sure what you're trying to do here. Are you trying to test a specific active connection or are you consulting to find the best connection for a particular business / size of business? If it's the latter then I'd start by trying to identify what kind of data the individual businesses applications actually NEED and how much they use them (as well as what they might need in the future).

We don't know anything about the setup of this business, but there are various pieces of software around for monitoring network traffic and those would also come in handy to help establish a baseline. Consulting work can get complicated very quickly, so it's not something that would be easy to thin down into a forum post.

 

[Captain_Cretin]

Set up and download multiple torrents in the 200-500GB range and watch the charts.

Just remember not to do anything else on the connection, as the torrent programs will restrict themselves to let other traffic use the connection.

Last time I did the above, I was getting a constant 55-65Mbps over many hours.

As a result of the test, I now have many hours of "The Gilmore Girls", "Blakes Seven" and many other classic 70's and 80's TV series to watch

 

[cpcnw]

I'm not 100% sure what you're trying to do here.

I have stats from an existing service. I want to try and figure out if BT Infinity is 'at least as good' in it's performance. The client is moving about 160GB data / month and has 25 staff mostly sending large amounts of email data and some surfing but very little streaming.

The existing service is facilitated via roof top WiFi but is relatively expensive. Now Infinity is available I have to be able to show clearly that it can do the job, or not as the case may be. The new VOIP phone system uses Infinity and I can test this overnight or over a weekend to make an assessment on the real world performance.

 

[ManOfMeans]

Whats the bandwidth of the current WIFI service?

 

[cpcnw]

Whats the bandwidth of the current WIFI service?

This any help?

https://jpegbay.com/gallery/006573764-.html

 

[ManOfMeans]

Not really, that looks like more of a internal wireless network speed than anything else.

 

[Mark.J]

You should just be able to ask the current provider in order to determine the link speed, assuming you don't know what it is. However if the capacity really is around 340Mbps then obviously an 'up to' 76Mbps FTTC line would be quite a drop (more so if the actual estimated speed ends up being much lower), but then again it might not matter if you aren’t using the current service to capacity.

150GB (GigaBytes) per month for 25 users sounds incredibly low for such a business. Just to put that into some perspective, it's close to last year's average for an individual consumer broadband connection . If that's the case then it seems like you don't need anything terribly fast and a 'up to' 76Mbps or 'up to' 38Mbps FTTC line would be fine, assuming the ISP actually says you can get that sort of speed (actual estimates tend to be lower on longer copper lines).

You also have to factor upload performance too as it's possible that the business may be more demanding on that side (e.g. a 76Mbps FTTC service usually comes with 19.5Mbps uploads). But we're all flying in the blind with our comments a bit as there's not a lot of detail to go off.

 

[cpcnw]

there's not a lot of detail to go off.

Thanks for the replies. Taking this step by step then, what detail exactly should I be seeking?

I have been told that speed tests alone are meaningless and that 'paper specs' are rarely seen in the wild.

Without some sort of yard stick how does one ever make any decisions?

Perhaps I am still struggling with my lack of understanding of 'capacity'? Anyway I am in the office tomorrow and intend to do some more testing and will get back to you on what I find.

Its interesting that you pick up on one factor that I think has a real bearing on all this and that's the upload speed. A lot of the data is outgoing emails with lots of drawing attached. This is via smtp. I have noticed that having busy outgoing connections can effectively reduce your incoming speeds as the ack packets aren't able to get out, hence reduced requests for incoming.

So, the determining factor could well be how easily swamped the uplink is on FTTC?

Am I right in saying that, for example if the uplink throughput is rated at say 20Mbps then you could only have 4 computers transmitting 5Mb /s on a continual basis?

 

[ManOfMeans]

Speed tests are useful who ever said they are meaningless isn't right, they are the only thing you can go off because it is a "live" statistic.

Its best to use a couple of different speed tests and on like speedtest.net use a few different servers which would give you a good idea of what it is able to do, you do how ever have to take in to account what the current traffic levels are at.

You could use things like iperf but still that's only really a speed test as well.

 

[Mark.J]

Am I right in saying that, for example if the uplink throughput is rated at say 20Mbps then you could only have 4 computers transmitting 5Mb /s on a continual basis?

Certainly if you have 4 computers uploading at 5Mbps each then that would gobble your bandwidth, although the network would usually auto-balance this so that other users could still upload (i.e. it would reduce the upload speed across the board as more users gobble up the bandwidth). The exception may be if you're doing something different with Quality of Service / filtering on the network to force some control over upstream speeds for prioritising specific tasks over others.

However it's very important to remember that the real-world environment of a network will be quite dynamic, so users may only be making heavier use of that upstream link for short periods and outside of that their traffic gobbling would be insignificant. So if you send an email with a 5MB file attached then that transfer would be quite short lived, thus the user won't constantly need to suck from the upstream pipe.

On the other hand if the business often has to upload hundreds of MegaBytes or a few GigaBytes in files then having a 20Mbps upstream max might become a much bigger problem. A consultant will often breakdown the use expectations of each staff member and application in order to better understand the businesses requirements, as well as monitoring current link activity.

 

[cpcnw]

Thanks for replies. Could you comment on whether the below is valid / meaningful?

As a preliminary test I used this tool ;

https://download.peplink.com/files/Broadband Speed Test v3.1.exe

on a Windows 7 Pro desktop PC with a 1GB nic. I opened up multiple instances of the tool and selected the 85Mb test file via Google in each case apart from Test4 where I used the following 100Mb test file locations;

http://speedtest.tokyo.linode.com/100MB-tokyo.bin
http://speedtest.london.linode.com/100MB-london.bin
http://speedtest.newark.linode.com/100MB-newark.bin
http://speedtest.atlanta.linode.com/100MB-atlanta.bin
http://speedtest.dallas.linode.com/100MB-dallas.bin
http://speedtest.fremont.linode.com/100MB-fremont.bin

I added the individual results together to get the total throughput as well as the finish times of the first and last process.

Infinity Test 1: 85Mb x 3
1st Finished: 25s
Last Finished: 25s
Throughput: 66.89Mbps

Infinity Test 2: 85Mb x 6
1st Finished: 49s
Last Finished: 51s
Throughput: 67.35Mbps

Infinity Test 3: 85Mb x 9
1st Finished: 69s
Last Finished: 71s
Throughput: 71.74Mbps

Infinity Test 4: 100Mb x 6
1st Finished: 37s
Last Finished: 77s
Throughput: 77.77Mbps

WiFi Test 1: 85Mb x 9
1st Finished: 76s
Last Finished: 78s
Throughput: 60.72Mbps

WiFi Test 2: 85Mb x 6
1st Finished: 56s
Last Finished: 52s
Throughput: 59.58Mbps

WiFi Test 3: 85Mb x 3
1st Finished: 27s
Last Finished: 28s
Throughput: 59.7Mbps

WiFi Test 4: 100Mb x 6
1st Finished: 39s
Last Finished: 107s
Throughput: 64.83Mbps

Screen Grabs
https://jpegbay.com/gallery/006581264-1.html#1

To clarify the Infinity account has only VOIP phones on and there was low traffic during testing whilst on the WiFi account the entire office activity was in full swing. That said the results aren't nearly as different as I expected.

Next I will be testing multiple uploads when I can come up with a decent test method.

One last thing; would running multiple tests on a single PC be affected by the limitations of the single 1GB nic?

 

[ManOfMeans]

Theres not much between them, the FTTC connection is slightly faster

 

[cpcnw]

Theres not much between them, the FTTC connection is slightly faster

Must admit was quite surprised myself although it is slightly unfair in that the infinity account was pretty much idling att of the tests.

I also did consider that as we have a couple of lines, if one additional FTTC was not good enough it would still actually be cheaper to run 2 and use some kind of bonding router, maybe sending email over one and http etc over the other.