News for hackers. Stuff that matters.
Re:Fastly S-1 by treis on Friday April 19, 2019 @10:46AM (#19700039)
>The whole "edge compute" space seems poised to do well to me.

IMHO, I think it's mostly a solution in search of a problem. The internet backbone is fast enough to not be noticeable for end users. A centrally located server in the U.S. will have a maximum ping of 40 ms to anywhere in the U.S. That's faster than is perceptible to the end user. The only mainstream usage I can see is cloud gaming where ping is that critical.

Re:Fastly S-1 by nyc_pizzadev on Friday April 19, 2019 @11:58AM (#19700582)
That's only good for small media like websites and images. Videos and downloads need to be cached in the provider networks, you simply cannot serve terrabits of data from a single origin. Well, you can, but networks don't like to operate like that. So ya, at scale, you need edges. The internet is dominated by companies at scale.
Re:Fastly S-1 by treis on Friday April 19, 2019 @12:36PM (#19700890)
That's more of a CDN than edge compute. Edge compute would be something like having your Rails app run in a bunch of edge locations and then doing some eventually consistent magic to sync them up.
Re:Fastly S-1 by merreborn on Friday April 19, 2019 @06:36PM (#19703583)
https://hpbn.co/transport-layer-security-tls/#leverage-early...

Establishing a connection can require several round trips, and the latency adds up. "early termination" with a point of presence near your end user makes a pretty massive difference, especially if your user isn't on the same continent. Every additional 100ms costs amazon 1% of their revenue https://news.ycombinator.com/item?id=273900

Re:Fastly S-1 by bogomipz on Friday April 19, 2019 @12:22PM (#19700771)
>IMHO, I think it's mostly a solution in search of a problem.

I agree, I've been looking at discussion one here recently on how someone was working on the problem of "data storage at the edge" to compliment their "compute at the edge" offering. I think you can argue that that when you move compute and data storage to the edge you've effectively moved the data center to the edge. At some point if you put enough data at the edge you operating your own backbone makes economic sense and CDN don't maintain backbones they mostly buy paid peering and transit and try to leverage public peering wherever they can. I would even say edge compute is a solution in search of revenue stream and hedge in a heavily commoditized market.

Re:Fastly S-1 by stickfigure on Friday April 19, 2019 @11:07AM (#19700191)
See also: Page takes 2500ms to render javascript.
Re:Fastly S-1 by paxys on Friday April 19, 2019 @09:18PM (#19704272)
Gaming is a bad example because it isn't the ping to the edge server that matters but ping to another user.
Re:Fastly S-1 by gmanley on Friday April 19, 2019 @11:05AM (#19700178)
It may not make sense if all your customers are in the US but locations like Australia and others very much have an issue with latency. Especially if the server they are hitting is around the globe.
Re:Fastly S-1 by treis on Friday April 19, 2019 @11:14AM (#19700254)
NYC to Sydney is still only ~200 ms. That's probably in the realm of being noticeable but only barely. Even then, the real need is more like one server per continent rather than every cell tower.
Re:Fastly S-1 by bogomipz on Friday April 19, 2019 @12:49PM (#19700997)
That statement is categorically false. One simple example is online gaming/MMOGs where a latency of 200 ms would be be a miserable experience. The upper bound often targeted for "acceptable interactivity" is somewhere between 100-130ms.
Re:Fastly S-1 by treis on Friday April 19, 2019 @02:12PM (#19701661)
(1) I excluded gaming as something that is legitimately ping dependent

(2) You can achieve 100-130 ms latency with a "like one data center per continent" sort of network. You don't need edge computing on hundreds or thousands of edge sites to achieve that.

Re:Fastly S-1 by cagenut on Friday April 19, 2019 @02:30PM (#19701811)
I'm sorry but what are you talking about? Fastly has about 60 PoPs (physical locations). No one is talking about hundreds of thousands of sites. Even akamai, who made it one of their primary marketing metrics, only ever got into the thousands.
Re:Fastly S-1 by treis on Friday April 19, 2019 @02:43PM (#19701921)
The GP is talking about putting them in cell phone towers.
Re:Fastly S-1 by cagenut on Friday April 19, 2019 @02:39PM (#19701884)
Your numbers are wrong but your line of reasoning is actually the basis of how Fastly's network was designed to be different from Akamai's. There's a graph about it at the 2:40 mark in this video: https://vimeo.com/132842124
Re:Fastly S-1 by lapnitnelav on Friday April 19, 2019 @05:10PM (#19703030)
That presentation was refreshingly honest and interesting to boot with.
Re:Fastly S-1 by huac on Friday April 19, 2019 @12:29PM (#19700833)
200ms is human reaction time, you can definitely notice it; the rule of thumb is 100ms https://www.pubnub.com/blog/how-fast-is-realtime-human-perce...
Re:Fastly S-1 by treis on Friday April 19, 2019 @02:15PM (#19701688)
Your cite says that < 1s is "for users to feel they are interacting freely with the information". Even a 400ms round trip is well within that limit.
Re:Fastly S-1 by sajal83 on Friday April 19, 2019 @04:00PM (#19702525)
Most, if not all websites require more than one round trip.
Re:Fastly S-1 by wbl on Friday April 19, 2019 @09:21PM (#19704281)
HTTP2 can reduce this significantly.