Is Netflix too much traffic for your ISP?

Forums › Online streaming services › Is Netflix too much traffic for your ISP?

1 | ... | 4 | 5 | 6 | 7 | 8 | 9

DonGould

3892 posts

Uber Geek

#1300651 8-May-2015 13:04

BigPipeNZ:

well the Truenet report shows that Bigpipe (us) and Orcon both maintained good peak time speeds at 98%.

In fact Bigpipe have never dropped below 95% in the time Truenet have been reporting our results.

so you could start there.

(we also now have UFB available in Wellington - so could be an option for some existing VF cable customers)

Interestingly Netflix actually makes up quite a small proportion of our traffic. Youtube is waaaaaay bigger.

What marketing is your company doing to UFB areas that are covering saturated cable areas?

Do you currently get a heads up from CFH every time a new area comes online so you can start direct marketing into that area?

Do you do direct marketing into these areas as they come on line?

Promote New Zealand - Get yourself a .kiwi.nz domain name!!!

Check out mine - i.am.a.can.do.kiwi.nz - don@i.am.a.can.do.kiwi.nz

raytaylor

4017 posts

Uber Geek

Trusted

#1300732 8-May-2015 14:39

BigPipeNZ:

Interestingly Netflix actually makes up quite a small proportion of our traffic. Youtube is waaaaaay bigger.

I must say that in looking at traffic trends on our network, I must ask "What is this netflix you speak of?"
I dont think our customers know it exists.

Ray Taylor

There is no place like localhost

Spreadsheet for Comparing Electricity Plans Here

BigPipeNZ

1170 posts

Uber Geek

Trusted

BigPipe

#1300800 8-May-2015 15:21

raytaylor:
BigPipeNZ:

Interestingly Netflix actually makes up quite a small proportion of our traffic. Youtube is waaaaaay bigger.

I must say that in looking at traffic trends on our network, I must ask "What is this netflix you speak of?"
I dont think our customers know it exists.

It makes me wonder about the iiNet claims that Netflix has gone from 3% of their traffic to 25% in just 3 weeks.

I know it's a different country, with different usage profiles etc etc but I just can't see how that happened in such a short time.

(Netflix is about 35% of US traffic by comparison, but that's after several years of solid growth)

bigpipe.co.nz
https://www.facebook.com/BigPipeNZ
https://twitter.com/BigPipeNZ

raytaylor

4017 posts

Uber Geek

Trusted

#1300832 8-May-2015 16:01

BigPipeNZ:

I know it's a different country, with different usage profiles etc etc but I just can't see how that happened in such a short time.

Netflix had a massive week of growth after a black friday thanksgiving sale where their app was included in all the smart tv's a few years ago in the USA

Also this is going to get interesting
Take a harvey norman salesman, in-store promo material, and suddenly every smart tv sold will have netflix capability - and the customers *will know it*
http://blog.netflix.com/2015/04/announcing-first-netflix-recommended-tvs.html

Ray Taylor

There is no place like localhost

Spreadsheet for Comparing Electricity Plans Here

DonGould

3892 posts

Uber Geek

#1300865 8-May-2015 16:45

BigPipeNZ:

It makes me wonder about the iiNet claims that Netflix has gone from 3% of their traffic to 25% in just 3 weeks.

I know it's a different country, with different usage profiles etc etc but I just can't see how that happened in such a short time.

Did they do some marketing to their customers?

Did they have a bunch of customers who were using VPNs, meaning that the traffic wasn't measured that way?

Promote New Zealand - Get yourself a .kiwi.nz domain name!!!

Check out mine - i.am.a.can.do.kiwi.nz - don@i.am.a.can.do.kiwi.nz

nigelj

856 posts

Ultimate Geek

#1300891 8-May-2015 17:39

DonGould:
BigPipeNZ:

It makes me wonder about the iiNet claims that Netflix has gone from 3% of their traffic to 25% in just 3 weeks.

I know it's a different country, with different usage profiles etc etc but I just can't see how that happened in such a short time.

Did they do some marketing to their customers?

Did they have a bunch of customers who were using VPNs, meaning that the traffic wasn't measured that way?

iiNet and Internode (I think they were the 2nd) did a big "Netflix Traffic is Unmetered" on our network blitz pre-launch with countdown clocks and all. I had the feeling that if you went to their websites you would've needed glasses if you didn't know Netflix was coming :)

Vodafone is the only company I'm aware of that has done any 3rd-party Netflix marketing (edit: in New Zealand (obviously)) and that is because of their 'OnAccount get X months free on a new contract' promo.

ripdog

548 posts

Ultimate Geek
Inactive user

#1302219 11-May-2015 19:40

I'm curious as to why Vodafone let their HFC network get so overloaded in the first place. For as long as I've been reading them, I've been astounded at the way VF cable download rate drops like a stone at prime time, incomparible to any ADSL, VDSL or fibre products. No apparent change for over a year, and now Netflix is the straw which finally pushed the 9pm performance into unusable territory?

Over the same time period all other products showed significant improvement - but cable stayed static then got worse.

Combined with their long-standing issues with helpdesk wait times (how long has that been going on for now?), it's beginning to seem like VF is just a grossly mismanaged company. Perhaps their market share allows them to rest on their laurels for some time?

Quic Broadband

Move to New Zealand's best fibre broadband service (affiliate link). Free setup code: R587125ERQ6VE. Note that to use Quic Broadband you must be comfortable with configuring your own router.

Darren0

70 posts

Master Geek

#1302380 11-May-2015 23:55

Netflix have started publishing their numbers on our ISPs performance

http://ispspeedindex.netflix.com/new-zealand

Graph view is fairly boring for now ;)

Talkiet

4793 posts

Uber Geek

Trusted

#1302381 12-May-2015 00:05

Darren0: Netflix have started publishing their numbers on our ISPs performance

http://ispspeedindex.netflix.com/new-zealand

Graph view is fairly boring for now ;)

Ouch.

I'd say the Spark numbers on that graph should be taken with a grain of salt at this stage as they are reporting over a month. The change made on Friday has improved the performance of most sites for a not insignificant number of Spark customers.

I'll be very interested to see the May numbers, although even then there is nearly a third of the month without the upgrade in the data...

Cheers - N

Please note all comments are from my own brain and don't necessarily represent the position or opinions of my employer, previous employers, colleagues, friends or pets.

Sideface

9365 posts

Uber Geek

Trusted

Lifetime subscriber

#1302401 12-May-2015 04:48

Darren0: Netflix have started publishing their numbers on our ISPs performance

http://ispspeedindex.netflix.com/new-zealand

Read the small print under the graph:

"The Netflix ISP Speed Index is a measure of prime time Netflix performance on a particular ISP and not a measure of overall performance for other services/data that may travel across the specific ISP network."

Sideface

JohnButt

374 posts

Ultimate Geek

Trusted

#1302579 12-May-2015 10:31

Sideface:
Darren0: Netflix have started publishing their numbers on our ISPs performance

http://ispspeedindex.netflix.com/new-zealand

Read the small print under the graph:

"The Netflix ISP Speed Index is a measure of prime time Netflix performance on a particular ISP and not a measure of overall performance for other services/data that may travel across the specific ISP network."

Yes, this could simply be a measure of Netflix's connectedness, and very little to do with ISP performance. Our tests downloading youtube video have a VERY different ISP order and very different results.

best speed = 3.77Mbps? even at 9pm that is just a nonsense number, less than 8% of our youtube downloads are less than 3.78Mb/s at 2100hrs. Even the worst ISP has just 35% of downloads less than 3.78Mb/s in April. One ISP never gets above 4%.

Note, this is not a weighted average of technologies and will be biased towards Fibre and Cable due to being limited to our MikroTik probes, however doing a similar analysis on DSL (35% of MikroTik probes) shows slightly better numbers. Sorry not able to name ISPs due to lack of cross checking so far.

Darren0

70 posts

Master Geek

#1302587 12-May-2015 10:35

JohnButt:
Sideface:
Darren0: Netflix have started publishing their numbers on our ISPs performance

http://ispspeedindex.netflix.com/new-zealand

Read the small print under the graph:

"The Netflix ISP Speed Index is a measure of prime time Netflix performance on a particular ISP and not a measure of overall performance for other services/data that may travel across the specific ISP network."

Yes, this could simply be a measure of Netflix's connectedness, and very little to do with ISP performance. Our tests downloading youtube video have a VERY different ISP order and very different results.

best speed = 3.77Mbps? even at 9pm that is just a nonsense number, less than 8% of our youtube downloads are less than 3.78Mb/s at 2100hrs. Even the worst ISP has just 35% of downloads less than 3.78Mb/s in April. One ISP never gets above 4%.

Note, this is not a weighted average of technologies and will be biased towards Fibre and Cable due to being limited to our MikroTik probes, however doing a similar analysis on DSL (35% of MikroTik probes) shows slightly better numbers. Sorry not able to name ISPs due to lack of cross checking so far.

Yup, absolutely it's not a measure of overall performance. YouTube publishes better data in my opinion, and Spark does quite well since you can distinguish services - https://www.google.com/get/videoqualityreport/

1 | ... | 4 | 5 | 6 | 7 | 8 | 9