Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Elon Musk said SpaceX's multi 1000 satellite Internet service will be IP-less

twitter.com

Post reply on HN
Print view View original

91–100 of 105 posts

Re: Elon Musk said SpaceX's multi 1000 satellite Internet service will be IP-less

#91

Post by fealios »

walrus01 wrote: post #78
mozumder wrote: post #76

Earlier quoted context omitted.

It's most likely a version of the ATM protocol[1]. Point-to-point, small packets, etc.. We used a version of that when I was working on the Iridium project. Likely the exact same thing here, especially if they hired the same engineers. [1]: https://en.wikipedia.org/wiki/Asynchronous_Transfer_Mode

You worked on the original Iridium satellite architecture? I am sure a lot of people here may not know how fully groundbreaking it was. Considering it is now 20-year-old tech the satellite-to-satellite Ka-band trunk links are an amazing technical achievement. That and the voice codec they came up with to make humans intelligible at 2400bps.

Damn, those are some sexy credentials.

Re: Elon Musk said SpaceX's multi 1000 satellite Internet service will be IP-less

#92

Post by treebeard901 »

kakarot wrote: post #85
petee wrote: post #57

Earlier quoted context omitted.

Beyond that, there is entirely no substance here - unless he wants to release details, this is non-news, just more business fluff - "Ours will be even better. It'll be huge. The best. Others, so sad" "And if its hacked, we'll just patch the firmware. So easy."

I'd be bragging about my satellite fleet, too. The guy's only human.

Do you have a source for this?

Re: Elon Musk said SpaceX's multi 1000 satellite Internet service will be IP-less

#93

Post by gnode »

KMag wrote: post #87
gnode wrote: post #29

I wonder if the reason here is that the topology of the network will change rapidly. E.g. as equatorial and polar satellites cross each other, they will be in direct communication with each other for only brief periods of time. There are ways to cope with routing changes in IP, but I don't know if it'd be seamless enough.

It's easy to read too much into a tweet, but it sounds like the header size overhead and number of options are the primary reasons for not using IPv6. Note also that IPv4 and IPv6 are nearly independent of the routing protocol. Also, my first job was writing discrete event models of wireless communications equipment, including a couple satellite models. You'd be surprised how simple satellites themselves are. A lot o…

I expect that internet traffic will still travel over the network using IP, but encapsulated in the SpaceX protocol. I expect translating the headers to and from IP would not make the protocol "simpler than IPv6", and would not be justifiable complexity to save a few bytes.

I think this "simple" protocol is just to simplify routing, and allow non-IP traffic over the network (in-band control messages; phone calls?).

Re: Elon Musk said SpaceX's multi 1000 satellite Internet service will be IP-less

#94

Post by bchallenor »

sho wrote: post #89
jsjohnst wrote: post #56

Earlier quoted context omitted.

1) Starlink will be in LEO around 800mi up unlike most communications satellites (like HughesNet) which are in geostationary orbit at ~22,200mi. Expect 25-35ms latency as compared to 600+. 2) due to having a lot more birds in the air at lower altitude with newer tech, they should be competitive with 5G networks speed wise as compared to existing satellite internet. Edit: fixed distance typo for HughesNet

Huh. That's really close. Your numbers are off though. Firstly it's ~40ms up to the bird if it's right above you, then 40 down to the base station if it is next door to your house. If it's on the other side of the world and has to travel around the whole mesh, add a hundred ms at best. Then from the base station to the destination - which could be anywhere in the world, and back, then back up to the bird, then back d…

I make it 4ms [1], not 40ms.

[1] http://www.wolframalpha.com/input/?i=800mi+%2F+speed+of+ligh...

Re: Elon Musk said SpaceX's multi 1000 satellite Internet service will be IP-less

#96

Post by madez »

jsjohnst wrote: post #45
madez wrote: post #24

I don't think the market for satellite Internet is only where other services don't provide fast enough internet. Of course, the technology of glass fiber in the ground can't be beaten qualitywise, but glass fiber can more easily be controlled by your ISP and your government, unlike satellite internet. If SpaceX manages to provide unfiltered, net neutral, cheap internet, I'd consider getting it to complement my curren…

> but glass fiber can more easily be controlled by your ISP and your government, unlike satellite internet. What would lead you to believe that? The reality is that the reverse is true, satellite internet is more heavily scrutinized as it’s far easier to intercept and very few companies are doing it.

On my earthbound connection my government can influence the ISP, but how is the government of [insert-autocratic-country] here supposed to control what its citizen do over satellite internet?

I assume that all communication to, in, and from the satellite network is secured by cryptography to prevent eavesdropping.

Re: Elon Musk said SpaceX's multi 1000 satellite Internet service will be IP-less

#97

Post by sho »

bchallenor wrote: post #94
sho wrote: post #89

Earlier quoted context omitted.

Huh. That's really close. Your numbers are off though. Firstly it's ~40ms up to the bird if it's right above you, then 40 down to the base station if it is next door to your house. If it's on the other side of the world and has to travel around the whole mesh, add a hundred ms at best. Then from the base station to the destination - which could be anywhere in the world, and back, then back up to the bird, then back d…

I make it 4ms [1], not 40ms. [1] http://www.wolframalpha.com/input/?i=800mi+%2F+speed+of+ligh...

* facepalm

Too late to edit. Misplaced the decimal point. Should have noticed that if 22k is 120, 800 can't possibly be 40. Duh.

Still, the actual latency is going to be very dependent on the distance to the base station. If it's nearby then shit, yeah, that will be within 20ms of 5G. Point conceded @jsjohnst.

Re: Elon Musk said SpaceX's multi 1000 satellite Internet service will be IP-less

#98

Post by mustardo »

petee wrote: post #57
imtringued wrote: post #28

The headline is completely misleading. The transport layer won't use IPv6. It's like saying ethernet doesn't use IPv6. That's unsurprising.

Beyond that, there is entirely no substance here - unless he wants to release details, this is non-news, just more business fluff - "Ours will be even better. It'll be huge. The best. Others, so sad" "And if its hacked, we'll just patch the firmware. So easy."

Do you have small hands?

Re: Elon Musk said SpaceX's multi 1000 satellite Internet service will be IP-less

#99

Post by KMag »

gnode wrote: post #93
KMag wrote: post #87

Earlier quoted context omitted.

It's easy to read too much into a tweet, but it sounds like the header size overhead and number of options are the primary reasons for not using IPv6. Note also that IPv4 and IPv6 are nearly independent of the routing protocol. Also, my first job was writing discrete event models of wireless communications equipment, including a couple satellite models. You'd be surprised how simple satellites themselves are. A lot o…

I expect that internet traffic will still travel over the network using IP, but encapsulated in the SpaceX protocol. I expect translating the headers to and from IP would not make the protocol "simpler than IPv6", and would not be justifiable complexity to save a few bytes. I think this "simple" protocol is just to simplify routing, and allow non-IP traffic over the network (in-band control messages; phone calls?).

Looking back at the context, I think you're right. It looks like Elon was just trying to explain to someone without much networking background that, like most link layers, they'd be using a link layer more simple than IPv6. He probably wasn't making a statement about lossy header compression that discards uncommonly used features.

Re: Elon Musk said SpaceX's multi 1000 satellite Internet service will be IP-less

#100

Post by jsjohnst »

sho wrote: post #97
bchallenor wrote: post #94

Earlier quoted context omitted.

I make it 4ms [1], not 40ms. [1] http://www.wolframalpha.com/input/?i=800mi+%2F+speed+of+ligh...

* facepalm Too late to edit. Misplaced the decimal point. Should have noticed that if 22k is 120, 800 can't possibly be 40. Duh. Still, the actual latency is going to be very dependent on the distance to the base station. If it's nearby then shit, yeah, that will be within 20ms of 5G. Point conceded @jsjohnst.

It doesn’t make sense to bounce a signal over North America / Europe / etc all around on the mesh. The best course is to send it to the ground almost immediately (aka So yes, my point still stands for the majority use case.

Also, to be clear, I was making following claims:

1) bandwidth on par with initial 5G deployments (aka up to 1gb/sec offering)

2) Latency on par or better than existing LTE networks.

Post reply on HN