Serving Netflix Video Traffic at 800Gb/s and Beyond [pdf]

Re:Serving Netflix Video Traffic at 800Gb/s and Beyond [pdf] by geodel on Friday August 19, 2022 @10:59AM (#32521842)

Seems to be mixing too many things here. Many scaling/ hardware challenges need a lot of people but it can still be true that Netflix has choke full of engineers making half-assed turd Java frameworks day in and day out. I know this because we are forced to use these crappy tools as they are made by Netflix so supposed to be best.

It's just that they succeeded in streaming market with low competition and great success bring in lot of post facto justifications on how outrageously great Netflix tech infra is.

I mean it may be excellent for their purpose but to think their solution can be industry wide replicated seems not true to me.

[ Parent | on HN ]

Re:Serving Netflix Video Traffic at 800Gb/s and Beyond [pdf] by tankenmate on Friday August 19, 2022 @11:11AM (#32521979)

Que? You don't seem to have much justification for your points; it seems more like a rant as you have had a bad experience using software provided by Netflix. It would be great if you could provide more details about what was wrong with it rather than just "we are forced to use these crappy tools". I'm genuinely interested.

In my personal experience lots of companies (admittedly all large companies, but many of which sell their services / software / hardware to smaller companies) have a use for serving hundreds of Gbps of static file traffic as cheaply as possible. And the slides for this talk seem exactly on the money (again from my experience slinging lots of static data to lots of users).

[ Parent | on HN ]

Re:Serving Netflix Video Traffic at 800Gb/s and Beyond [pdf] by paxys on Friday August 19, 2022 @11:18AM (#32522069)

So Netflix published a framework which seemingly isn't suitable for your use case, your managers forced you to use it, and your response is to blame...Netflix?

[ Parent | on HN ]