For many years we have had a local instance of the Ookla speedtest.net on our network, and while it is pretty good some other tests seem include more detailed results.
I am aware of the following speedtest systems that an operator can likely have a local instance of:
A lot of people have crappy performance to those. For example, from a 10G
server to fast.com I was pulling around 9Mbps up/down. 1 hop away from a
Netflix open connect appliance.
Ah, this is the first I've heard of slow fast.com performance with someone actually connected to them. Usually it's an ISP that's a few AS hops away from Netflix.
First, you only get down from fast.com not up - so the up/down is a bit
suspect there.
Second, this is a more 'real world' test than iperf - if you want to ensure
that your NIC is operating at the rated speed I'd imagine you'd have the
ability to setup an iperf target and check Layer2/Layer3 transfer
speeds/etc.
Third, you should really look into that if you are 1 hop away and getting
that type of speed. Clearly you deserve better.
80Mbps result (with comparison link if you don't like that one): http://i.imgur.com/Cnr92Ag.png - of course I'm on a 240Mbps WAN connection:
*Last Result:*
Download Speed: *236960* kbps (29620 KB/sec transfer rate)
Upload Speed: *22991* kbps (2873.9 KB/sec transfer rate)
Latency: *12* ms
Jitter: *2* ms
12/5/2016, 10:57:56 AM
(Those results are from my provider in the Tampa Bay area at: speedtest.bhn.net).
Right, it's mostly ISPs that don't understand the BGP world or how speedtests work. I think, you, Paul and myself were the only ones participating that really knew.
A lot of people can't differentiate between what the test is testing, a bad test and connectivity issues producing bad results on an otherwise good test.
I'd say that most of the time, it's the last category.
I'm impressed that openspeedtest.com supports IPv6! I haven't noticed this
in too many speedtests yet, and its something I've been asked about on
occasion.
I like nperf.com as I usually always get consistent results and you
can keep track of your results if you sign up. They only have one
server in Canada (hosted by OVH in Beauharnois) but you can host your
own like Ookla's Speedtest.net.
Generally I don't bother with speed testers unless I'm wanting a quick
guesstimate -- I wouldn't recommend using them as a measure of how "fast"
an internet connection is because there's always other factors in
contention, and it only tests the path to the speedtester.
Having said that, despite the obnoxious adverts on the site, speedof.me
provides a really nice non-flash interface that is an excellent teaching
tool for showing people how TCP congestion windows work, and lets you
demonstrate to people the effect of buffers in a network etc.
I'll be taking notes if anyone provides anything else similar that does a
similar (if not better) job, maybe one that can be self-hosted too!
The problem with fast.com is that they use HTTPS for the test. The user needs a fast computer to decode the SSL at full speed. Even if you have a very fast computer the test will max out at 100-200 Mbps because the Netflix servers are apparently not able to encode SSL any faster. Maybe we would get better speed if multiple SSL connections were used.
I just did a test on fast.com and got 150 Mbps. Click the compare on speedtest.net button and I got 940 Mbps at beta.speedtest.net. The computer is Intel i7 5820K, the OS is Ubuntu 16.04 and the internet is 1 Gbps delivered on GPON. The test runs on IPv6.
We are directly peered with Netflix with 2x10G and there is plenty of capacity. It appears fast.com is on Akamai but the test itself is downloading data via our peering.