FreeBSD optimizations used by Netflix to serve video at 800Gb/s [pdf]

https://people.freebsd.org/~gallatin/talks/euro2022.pdf

by ltadeut • 4 years ago

390 177 4 years ago

people.freebsd.org

_trackno52• 4 years ago

Recording of the presentation can be found here: https://www.youtube.com/watch?v=36qZYL5RlgY

Pretty cool stuff

eatonphil2• 4 years ago

From what I can see in a quick search (and from this presentation), Netflix only uses FreeBSD for serving video and they run these servers themselves in their own datacenters I guess. In contrast their apps on EC2 use Linux [0]. Sounds like the time has not yet come when AWS is paying anyone full time to support FreeBSD on EC2.

[0] https://twitter.com/brendangregg/status/1412201241472471048

vbezhenar• 4 years ago

What does it mean to support FreeBSD on EC2? Surely it's just a KVM so you can run whatever you want?

sanxiyn2• 4 years ago

It means, for example, writing a FreeBSD kernel driver for Elastic Network Adapter (ENA). Both Linux kernel driver and FreeBSD kernel driver is available at https://github.com/amzn/amzn-drivers

cotillion• 4 years ago

Netflix works because they move content close to the users. This is done by either having the ISP establish a peering connection directly to Netflix hosted servers or by having the ISPs host "Open Connect Appliances" which cache the most requested content. These appliances are based on FreeBSD.

The AWS egress savings from this setup must be immense.

https://openconnect.netflix.com/

ilyt3• 4 years ago

Yup, cloud bandwidth is insanely expensive considering to what you actually pay to get link to your datacenter.

And you pay either by 95th percentile (basically "peak usage") or by whole link, not per megabyte sent

erk__2• 4 years ago

cperciva whom you link have worked quite a bit on EC2 support for FreeBSD, a lot of it documented on their blog [0] and supported by Patreons at [1].

But yeah it would be nice if there was someone who could work on it full time

[0] https://www.daemonology.net/blog/2022-03-29-FreeBSD-EC2-repo...

[1] https://www.patreon.com/cperciva

eatonphil2• 4 years ago

Yep! In the thread he describes how he is not enough.

throw0101a2• 4 years ago

And to think not that long ago I remember being excited when the V.92 standard was released and I could get 56 kb/s on my dial-up connection:

* https://en.wikipedia.org/wiki/V.92

rwl4• 4 years ago

How about the marvel that was Walnut Creek's cdrom.com that served 10,000 simultaneous FTP connections back in 1999? [1]

I was always blown away by how much more efficient FreeBSD's network stack was compared to Linux at the time. It convinced me to go FreeBSD-only for a few years.

[1] http://www.kegel.com/c10k.html

alberth7• 4 years ago

compared to Linux at the time

Do you consider that not to still be the case?

drewg12319• 4 years ago

Author here, happy to answer questions

erk__2• 4 years ago

Do you know if there is any documentation regarding interfacing with the KTLS, eg to implement support for a new library?

sanxiyn2• 4 years ago

For Linux, there is a documentation at kernel.org: https://docs.kernel.org/networking/tls.html

drewg12319• 4 years ago

The ktls(4) man page is a start. The reference implementation is OpenSSL right now. I did support for an internal Netflix library a while ago, I probably should have documented it at the time. For now feel free to contact me via email with questions (the username in the URL, but @netflix.com)

monotux• 4 years ago

How long will you be able to keep up with this near yearly doubling of bandwidth used for serving video? :)

drewg12319• 4 years ago

It depends on when we get PCIe Gen5 NICs and servers with DDR5 :)

crest• 4 years ago

Which NIC and driver combinations support kTLS offloading to the NIC?

How did you deal with the hardware/firmware limitations on the number of offloadable TLS sessions?

drewg12319• 4 years ago

We use Mellanox ConnectX6-DX NICs, with the Mellanox drivers built into FreeBSD 14-current (which are also present in FreeBSD 13).

alberth7• 4 years ago

A. Just curious, are these servers performing any work besides purely serving content? Eg user auth, album art, show description, etc?

B. What’s the current biggest bottleneck preventing higher throughout?

C. Has everything been up streamed? Meaning, if I were to theoretically purchase the exact same hardware - would I be able to achieve similar throughout?

(Amazing work by the way in these continued accomplishments. These posts over thr years are always my favorite HN stories.)

drewg12319• 4 years ago

a) These are CDN servers, so they serve CDN stuff. Some do serve cover art, etc sorts of things.

b) Memory bandwidth and PCIe bandwidth. I'm eagerly awaiting Gen5 PCIe NICs and Gen5 PCIe / DDR5 based servers :)

c) Yes, everything in the kernel has been upstreamed. I think there may be some patches to nginx that we have not upstreamed (SO_REUSEPORT_LB patches, TCP_REUSPORT_LB_NUMA patches).

amelius• 4 years ago

At what point will it make more sense to use specialized hardware, e.g. network card that can do encryption?

drewg12319• 4 years ago

We already do. The Mellanox ConnectX6-Dx with crypto support.. It does inline crypto on TLS records as they are transmitted. This saves memory bandwidth, as compared to a traditional lookaside card.

waynesonfire• 4 years ago

How did you generate those flamegraphs and what other tools did you use to measure performance?

My motivation for asking comes from these findings in the pdf,

Did the graph show the bottleneck contention on aio queue? Did the graph show that "a lot of time was spent accessing memory"?

What made freebsd a better platform compared to Linux to begin tackling this problem?

Thanks! Super interesting. Both a freebsd fan and I have workloads that I'd love to explore benchmarking to squeeze more performance.

drewg12319• 4 years ago

How did you generate those flamegraphs and what other tools did you use to measure performance?

We have an internal shell script that takes hwpmc output and generates flamegraphs from the stacks. It also works with dtrace. I'm a huge fan of dtrace. I also make heavy use of lockstat, AMD uProf, and Intel Vtune.

Did the graph show the bottleneck contention on aio queue? Did the graph show that "a lot of time was spent accessing memory"?

See the graph on page 32 or so of the presentation. It shows huge plateaus in lock_delay called out of the aio code. Its also obvious from lockstat stacks (run as lockstat -x aggsize=4m -s 10 sleep 10 > results.txt)

See the graph on page 38 or so. The plateaus are mostly memory copy functions (memcpy, copyin, copyout).

We already use FreeBSD on our CDN, so it just made sense to do the work in FreeBSD.

The talk is on Youtube https://youtu.be/36qZYL5RlgY

smokel• 4 years ago

The flame graphs might be generated using Brendan Gregg's utility, see https://www.brendangregg.com/flamegraphs.html

hzhou321• 4 years ago

What prevents linux to achieve the same bandwidth?

_trackno52• 4 years ago

Not sure about all other optimisations, but Linux doesn't have support for async sendfile.

dist1ll• 4 years ago

How involved was Netflix in the design of the Mellanox NIC? How many stakeholders does this type of networking hardware have, relatively speaking?

Also, what percentage of CDN traffic that reaches the user is served directly from your co-located appliances?

_-david-_• 4 years ago

There are a lot of slides and I am on my phone, so sorry if it was addressed in the slides.

How does Linux compare currently? I know in the past FreeBSD was faster, but are there any current comparisons?

alberth7• 4 years ago

How much "U's" of space do ISP typically give you (e.g. 4U, 8U, etc)?

nixgeek• 4 years ago

This is going to be a “How long is a piece of string?”. Each ASN will be unique, and even within any large ISP, there may be many OCA deployment sites (there won’t just be one for Virgin Media in UK) and each site will likely have subtly different traffic patterns and content consumption patterns, meaning the OCA deployment may be customized to suit, and the content pushed out (particularly to these NVME-based nodes) will be tailored accordingly.

Since the alternative for an ISP is to be carrying the bits for Netflix further, the likelihood is they’ll devote whatever space is required because that’s much cheaper than backhauling the traffic and ingressing over either a settlement-free PNI or IXP link to a Netflix-operated cache site, or worse, ingressing the traffic over a paid transit link.

Meanwhile, on the flipside, since Netflix funds the OCA deployments they have a strong interest in not “oversizing” the sites. That said I’m sure there is an element of growth forecasting involved once a site has been operational for a period of time.

kloch• 4 years ago

What filesystem(s) are you using for root and content?

And If ZFS, what options are you using?

drewg12319• 4 years ago

We use ZFS for root, but not content. For content we use UFS. This is because ZFS is not compatible with "zero-copy" sendfile, since it uses its own ARC cache rather than the kernel page cache, meaning sending data stored on ZFS requires an extra data copy out of the ARC. Its also not compatible with async sendfile, as it does not have the methods required to call the sendfile completion handler after data is read from disk into memory.

vkaku2• 4 years ago

Read the presentation. Had super noobie level questions.

Is the RAM mostly used by page content read by the NICs due to kTLS?

If there was better DMA/Offload could this be done with a fraction of the RAM? (NVME->NIC)

If there was no need to TLS, would the RAM usage drop dramatically?

drewg12319• 4 years ago

These are actually fantastic questions.

Yes, the RAM is mostly used by content sitting in the VM page cache.

Yes, you could go NVME->NIC with P2P DMA. The problem is that NICs want to read data at once TCP mss (~1448b) and NVME really wants to speak in 4K sized chunks. So there needs to be some buffers somewhere. It might eventually be CXL based memory, but for now it is host memory.

EDIT: missed the last question. No, with NIC kTLS, the host RAM usage is about the same as it would be without TLS at all. Eg, connection data sitting in the socket buffers refers to pages in the host vm page cache which can be shared among multiple connections. With software kTLS, data in the socket buffers must refer to private, per-connection encrypted data which increases RAM requirements.

eddyg• 4 years ago

Wondering if there's a video presentation to go along with the slides?

notaplumber1• 4 years ago

This talk was given at this years EuroBSDcon in Vienna, recording is up on YouTube.

https://2022.eurobsdcon.org/

https://www.youtube.com/watch?v=36qZYL5RlgY

Some really great talks this year from all the *BSDs, highly recommend checking a look: https://www.youtube.com/playlist?list=PLskKNopggjc6_N7kpccFZ...

coredog64• 4 years ago

And is the video presentation on Netflix?

fsckin• 4 years ago

What tools do you use for load testing / benchmarking?

drewg12319• 4 years ago

At a very basic microbenchmark level, I use stream, netpef, a few private VM stress tests, etc. But the majority of my testing is done using real production traffic.

PYTHONDJANGO• 4 years ago

* How is the DRM applied? * Is the software, that does DRM open source, too?

tame3902• 4 years ago

1. I got excited when I saw arm64 mentioned. How competitive is it? Do you think it will be a viable alternative for Netflix in the future?

2. On amd, did you play around with BIOS settings? Like turbo, sub-numa clustering or cTDP?

drewg12319• 4 years ago

Arm64 is very competitive. As you can see from the slides, the Ampere Q80-30 is pretty much on-par with our production AMD systems.

Yes, I've spent lots of time in the AMD BIOS over the years, and lots of time with our AMD FAE (who is fantastic, BTW) poking at things.

alberth7• 4 years ago

If a "typical" NIC was used, what do you think the throughput would be?

I have to imagine considerably less (e.g. 100 Gb/s instead of 800).

toast06• 4 years ago

Not the OP, but that's basically in the slides. When it's kTLS, but not NIC kTLS. Maybe you could optimize that a bit more around the edges if NIC kTLS wasn't an option.

drewg12319• 4 years ago

Back of the envelop guess is ~400Gb/s. Each node has enough memory BW for about 240Gb/s, then factor in some efficiency loss for NUMA..

shaklee32• 4 years ago

what do you mean by typical nic? these are COTS NICs anyone can buy

toast06• 4 years ago

Could you do a mix of NIC oriented siloing and Disk oriented siloing?

It seems like the bottlenecks are different, so if you serve X% of traffic on the NIC node and Y% on the disk node, you might be able to squeeze a bit more traffic?

Also, how amenable to real time analysis is this? Could you look at the request where the request comes in on node A and the disk is on node B, and tell which NIC is less loaded and send the output through that NIC. (Some selection algorithm based on relative loading anyway)

drewg12319• 4 years ago

Those are good ideas. The solution I was thinking of would be to do replication of the most popular content to both nodes.

krylon4• 4 years ago

And here I sit like a chump with my home server connected to a 100MBit switch. (I paid for that switch, and I'm not replacing it until it gives up the ghost.) (And before you ask, the server also runs FreeBSD, and I'm very happy with the result.)

nix236• 4 years ago

Bring it to the max with multipath ;) since you have already Freebsd, no need to throw those beautiful reliable thing's away, maybe just buy a second...third? Dirt-cheap 100MBit card:

https://en.wikipedia.org/wiki/Multipath_TCP#Implementation

krylon4• 4 years ago

The server has a second NIC, but the switch has no more free ports. I briefly thought of bonding, but stopped when I read that the switch would need to support it (which it almost certainly does not).

But my point was that for my requirements, 100MBit are actually sufficient and FreeBSD still is a good choice for me, I was just being snarky about it. (I do find it aesthetically displeasing, though, that my wifi is now faster than my wired network, but I can live with that.)

toast06• 4 years ago

I understand the motivation, but $20 gets you an 8-port gigE switch, so it seems like the wrong hill to die on. :)

krylon4• 4 years ago

I know, but so far 100MBit is sufficient, actually, I rarely move Gigabytes of data around. When it becomes annoying, I'll get a new switch, but so far the pressure is really low.

ilyt3• 4 years ago

It's nice to see someone actually still does proper engineering instead of farting something about cloud and webscale and just throwing money at a problem.

kleiba• 4 years ago

Yet, in order to watch Netflix on FreeBSD, you have to jump through such hoops as "downloading either google chrome, vivaldi, or brave, and [using] a small shell script which basically creates a small jail for some ubuntu binaries that actually install widevine which is essential for viewing some DRM content such as Netflix" [1]

[1] https://www.youtube.com/watch?v=mBYor4wL62Q

IntelMiner• 4 years ago

Devils advocate: The people who work on the server engineering at Netflix don't exactly have much control over copyright holders being lawyer brained man children

alberth7• 4 years ago

which is essential for viewing some DRM content such as Netflix

Are you complaining that Netflix doesn't want people to pirate content, content they might have licensed from 3rd parties which contractually bind them to not let being pirated?

This + is the development/resources/cost of serving such few people on FreeBSD even worth it.

Note: I'm a huge FreeBSD fan. But consider this totally understandable on Netflix part.

somehnguy• 4 years ago

But it doesn’t prevent it from being pirated at all. You can get any Netflix release you want within minutes of release on any torrent site. Sometimes before the official release even.

It just makes normal people jump through hoops to watch the things they are trying to pay for. That’s a DRM issue in general though, I acknowledge this isn’t just a Netflix thing.

Thaxll3• 4 years ago

DRM makes 0 sense since you can get any content using torrents in 2min. It's not protecting anything, as a matter of fact it's just making people download more since it's a painful experience.

For example on Windows with Chrome you only get 720p playback for Netflix, complete nonsense.

__MatrixMan__4• 4 years ago

BitTorrent is probably easier. I just wish there was a good way to send money to the artists without also funding DRM enhancements.

andsoitis• 4 years ago

So you want to send money to all the people who worked on the TV Show or the Movie you just downloaded?

I don’t think you realize how impractical that is. Take a look at the credits at the end of a movie some time. Or look up the list of people who worked on a particular episode of a show (yes, it can vary throughout a season).

SpaceInvader• 4 years ago

FreeBSD is not a "desktop first" system and has strengths elsewhere. I use it for 20+ years constantly. Sadly my experiments with FreeBSD desktop ended years ago as there always was something "not working".

asveikau2• 4 years ago

Typing this on a FreeBSD laptop.

Haven't tried using netflix on it though.

pjmlp2• 4 years ago

UNIX's strength was never in the desktop experience, rather server room.

akreal• 4 years ago

Not true for macOS.

nix236• 4 years ago

That's just not true, it's not just apple but also SGI, maybe not directly desktop but workstation for sure.

jbirer• 4 years ago

That is the problem with the BSD license, it says "use my work and don't give anything back". Of course, GPL gets violated too, but would be very difficult by an American company like Netflix.

nix236• 4 years ago

Since Netflix is NOT deploying it's code/Devices to users it would be exactly the same with a GPL.

it says "use my work and don't give anything back".

That's not written.

But compared to GPL you just could read the whole BSD-2 license in 1 minute.

nix236• 4 years ago

I really think Netflix could make some good money being a multimedia-cdn (even for "competitors")

coldpie• 4 years ago

Wow, that is actually a really interesting idea in the context of developing a YouTube competitor. Delivery & bandwidth are a really high barrier to entry, and piggy-backing off of Netflix's existing network could really lower those costs. I agree the "providing services to your direct competition" is probably a stumbling block, and likely Netflix has other irons in the fire. But anyway it's a cool idea to think about.

virtuallynathan2• 4 years ago

It works with such high efficiency because we know how to place content in advance, and the catalog is relatively small. Trying to serve 800Gbps of YouTube content would be a nightmare.

ilyt3• 4 years ago

I wonder how much of YT traffic is the "big" (say >200k viewers in a month) vs the small guys.

But yeah, once your hot data size exceeds cache byebye efficiency

nix236• 4 years ago

I don't talk directly about Youtube, but also serving Disney, hula, but ESPECIALLY national/continental portals like Arte.tv, Play-SRF, ARD-Mediathek etc.

drewg12319• 4 years ago

Indeed. YT has a much different problem, which is to determine which video is going to go viral, and then transcode it into popular formats when it does.

In comparison, we pre-transcode everything to exacting standards, so all our CDN has to do is serve is static files.

jedberg5• 4 years ago

I thought the same thing 10 years ago when I worked there. At the time management was not interested in losing focus on doing anything other than streaming movies to customers.

But it should be noted that the FreeBSD Openconnect boxes are highly optimized to Netflix's use case. Which is serving a predefined set of content that has been pre-rendered. Youtube and its ilk are a completely different use case.

The Netflix cache is so optimized for serving Netflix movies that for many years we still used Akamai for all of our other CDN needs, but it looks like they may have finally moved that to Netflix's own CDN now.

ksec• 4 years ago

Just some napkin maths. ( Correct me if I am wrong )

Looking at the 800Gbps Config, Dell R7525 with Dual 64C / 128T and 4x Connect-DX 800Gbps in 2U.

With Zen 4C, 128C and PCI-E 5.0, Connect-7, two node could fit into 2U. i.e doubling to 1.6Tbps per 2U.

That is going from 16Tbps to 32Tbps per Rack. ( Using 40U only )

To things in perspective, if every user were to use 20Mbps Stream at the same time, ( not going to happen due to time zone difference ), the 250M Netflix subscribers worldwide would need 5000M Mbps or 5000 Tbps. That is less than 200 Racks to serve every single of their customer on planet earth. ( Ignoring Storage. ) You could ship a Rack to every Region, State, Nation, Jurisdiction or Local ISP and Exchange and be done with it.

I hope Lisa Su sent drewg123 and his team at Netflix with Zen 4C ASAP to play, cough, I mean help them test it.

Note: We have PCI-E 6.0 ( and 7.0 ), DDR6 on Roadmap. The 200 Racks could be down to 50 Racks by the end of this decade. Assuming Netflix is still streaming at the same bitrate.

alberth7• 4 years ago

That is going from 16Tbps to 32Tbps per Rack ... only need 200 racks

I doubt ISP's give an entire rack to Netflix. I wouldn't be surprised if they only get like 4U total (hence why throughput per server is so important to Netflix).

BonoboIO3• 4 years ago

The minimum requirements

https://openconnect.zendesk.com/hc/en-us/articles/3600345383...

I think it depends on the size of isp, probably a rack would be too much even for the biggest isps, but one 4u too less.

meltedcapacitor• 4 years ago

Why not? It's top bandwidth consumer for a retail ISP and surely any reasonable amount of rack space is worth the savings in interconnect bandwidth.

jedberg5• 4 years ago

Some ISPs give a full rack, some don't. It depends on how much traffic they have and how willing they are.

But a lot of the racks sit at internet exchange points, where Netflix rents one or more racks at a time.

reaperducer2• 4 years ago

That is less than 200 Racks to serve every single of their customer on planet earth. ( Ignoring Storage. )

If you're going to ignore storage, Netflix could just ship a low-end video server to every one of its customers and be done with it.

Every problem is an easy problem if your pretend the hard parts don't exist.

OliverGuy• 4 years ago

How much storage does Netflix actually need for its whole library?

It's got about 17,000 titles globally [1]. If they have copies in SD, 720p, HD and 4k that would be 68,000 versions (plus some extra audio tracks for stuff dubbed in multiple languages, but I suspect this is fairly minimal in terms of storage though)

Let's assume that the resolutions have the bitrates at 5, 10, 15 and 20 mbps.

The average length of a Netflix original movie is ~90mins [2]

So that would require about 575TB in storage if I have done my maths correctly.

You would need about 20x30TB Kioxia CD6 SSDs for all that. Very expensive but definitely technically possible.

I could totally see it being possible to fit those drives in a single node to push the 800gbps required, not increasing the over rack requirement at all. Not sure if the bandwidth from that many drives is enough, might have to cache some of the most watched stuff to ram)

Not gonna see any in home boxes with all the titles pre loaded any time soon though. As a hard drive array that's still 30x20TB drives.

[1] https://www.comparitech.com/blog/vpn-privacy/netflix-statist...

[2] https://stephenfollows.com/netflix-original-movies-shows/#:~...)

loeg5• 4 years ago

Netflix is more likely to use a single box of this kind of throughout at any given POP than a rack of them. For bigger installations they can use cheaper, less throughput-dense hardware (although I don’t know if they do).

lillecarl3• 4 years ago

Take a look at the hardware, it isn't particularly expensive stuff.

aeyes• 4 years ago

You are probably overestimating Netflix traffic by a lot.

IX.br peak traffic is 20Tb/s, DE-CIX peak traffic is 14Tb/s, AMS-IX is around 11Tb/s.

The 800Gpbs machine is probably enough for a country.

Netflix traffic stats at PIT Chile, this is their only peering connection in Chile: https://www.pitchile.cl/wp/graficos-con-indicadores/streamin...

lostlogin2• 4 years ago

I notice people streaming in very low resolution without realising it, and sometimes intervene when the pain gets too great.

I’d be vey surprised if the average bitrate was anywhere near the appropriation.

However that wasn’t the point of the calculation, it was looking for a maximum.

srmn• 4 years ago

This assumption misses out on all the private interconnect links and deployed OpenConnect appliances within ISP networks - a majority of Netflix's traffic today. IXes are only a small portion of overall internet traffic.

recuter• 4 years ago

There is the rather intriguing prospect of NVM Express over Fabrics (NVMe-oF): https://en.wikipedia.org/wiki/NVM_Express#NVMe-oF

Marvel Octeon 10 DPU (with an integrated 1 Terabit switch): https://www.marvell.com/content/dam/marvell/en/company/media...

Probably pretty soon you'll be able to chuck in a few hot swappable 100 TB Nimbus exadrives (https://nimbusdata.com/products/exadrive/) in there and call it a day. 1T in 1U. :)

Melatonic2• 4 years ago

Interesting to see that Infiniband is still kicking

virtuallynathan2• 4 years ago

Back of the Napkin Zen4 / Genoa gets you to ~500GB/s PCIe and ~500GB/s of DRAM bandwidth -- nearly 4Tbps! Zen3/Rome is ~300GB/s PCIe and ~300GB/s DRAM -- about 2.4Tbps. A single 2U box with Genoa might scale to 1.25Tbps+ of useful Netflix traffic. We'll have to see what magic Drew can pull :)

Melatonic2• 4 years ago

The people doing this might also be doing infra as code for the virtualization layer on the hardware itself - which this might not be able to satisfy. At minimum they surely have a ton of this stuff deployed already so changing hardware specs big time might not be worth the cost.

Also are you taking into account encryption for those specs?

Matthias247• 4 years ago

You could ship a Rack to every Region, State, Nation, Jurisdiction or Local ISP and Exchange and be done with it.

And when a single rack is down the whole region might be impacted negatively. The blast radius of such a setup would be huge!

As a CDN you really want to avoid this. Even for regular operations you need to be able to afford some servers going out of service for software update and other maintenance - without impacting availability, latency for users (the latter would happen if you route them to a complete different region), or infrastructure costs (which increase if you can't serve data from as close to the user as possible anymore, and have to pay for additional networking fees).

All major CDN providers have hundreds of regions (all with multiple hosts), and you can't really avoid the former. You could run less hosts per region, but whether 1 is the feasible will depend a lot on other parts of your system.

BonoboIO3• 4 years ago

It amazes me, that Netflix is capable of such top of the line engineering things (really mindblowing stuff, one machine that streams nearly 1 Terabit pers second), but is for the love of god unable to stream HD Content to my iPhone (newest firmware, all up2date). Tried everything gigabit wifi, cellular, multiple ISPs ...

It is better for me to pirate their content, play it with Plex and be happy. I pay for Netflix, but still have to download it, to see it an acceptable quality. Absurd. The support couldn‘t help. It doesn’t affect, because I have my Torrent/Plex Setup, but for 99.9% of people it is a subpar experience.

I think the best years are over for Netflix. The hard awakening is here to make content that the users want and they are a movie/tv content company, not primarily a „tech company“.

leetharris• 4 years ago

You live in a bubble. The vast majority of the world likely cannot even tell the difference between HD and 4K. Netflix continues to grow its content and retain subscribers.

BonoboIO3• 4 years ago

Netflix is a media company as I said.

Well 4K vs HD, you are right, but 480p on a Retina display right in front of me. Really obvious.

selfhoster69• 4 years ago

unable to stream HD Content to my iPhone

Yeah this has been the case since forever. It prioritizes instant playback vs forcing 1080p or similar.

Can't speak for iPhone, but on iPad, I've moved to using the website which goes goes to 1080p immediately.

still have to download it, to see it an acceptable quality

Downloaded content do contain a whole lot more compression than streaming at max phone supported quality, so just a tiny FYI.

nicholasjarnold• 4 years ago

I come from the time when the first internet connection my house had was a 56k modem...just before cable modems/DOCSIS started rolling out in the midwest. These speeds are somewhat mind boggling to me. (Yeah, yeah, datacenter vs home, but it's still somewhat hard to imagine saturating pipes like those.)

While standing in a state of mild awe at 800Gb/s I read reviews and consider upgrading my house to 2.5Gb/s equipment... Should I just wait for 10Gbit to get a bit cheaper? Should I ditch copper and go fiber like that guy who was on the front page here recently (probably not, but that was cool)? Maybe raw single core CPU performance is starting to level off a bit, but it seems that networking technologies are still advancing a rapid clip!

seized• 4 years ago

Fiber 10Gb is very cheap. NICs and SFPs from Ebay, fibre from FS.com in whatever length you want. I got a plenum rated 100 ft 4 pair cable from FS.com for $100 or so, and it was only that expensive for the plenum rating as it runs through my cold air returns.

pyuser583• 4 years ago

I think it’s weird and cool how Netflix used FreeBSD/Dlang.

Linux is just the automatic go to. It’s great the big tech companies are rethinking these basics.

loeg5• 4 years ago

Where are you seeing any mention of Dlang?

emilper• 4 years ago

https://github.com/Netflix/vectorflow

ocbyc• 4 years ago

"In networking units"

this15testing2• 4 years ago

FreeBSD optimizations used by Netflix to serve video at 800Gb/s [pdf]

Related Stories