Keeping Netflix Reliable Using Prioritized Load Shedding
netflixtechblog.com
Keeping Netflix Reliable Using Prioritized Load Shedding
1–10 of 77 posts
Re: Keeping Netflix Reliable Using Prioritized Load Shedding
#2Totally anecdotal evidence, but I was in a rural NY house served by DSL for the past 6 months. The DSL has consistent packet loss between 4 and 6%. The only video service that could handle this level of packet loss well was Amazon Prime. Netflix couldn't even load its browse screen until the past two weeks, where something changed, and suddenly Netflix could handle the high packet loss as well as Amazon Prime.
Thank you to the engineers and developers!
Re: Keeping Netflix Reliable Using Prioritized Load Shedding
#3I love the phrase "Prioritized Load Shedding" as corporate-speak for "dropping the less-important traffic."
Re: Keeping Netflix Reliable Using Prioritized Load Shedding
#4I love the phrase "Prioritized Load Shedding" as corporate-speak for "dropping the less-important traffic."
Dropping some traffic to avoid complete melt down.
Re: Keeping Netflix Reliable Using Prioritized Load Shedding
#5Totally anecdotal evidence, but I was in a rural NY house served by DSL for the past 6 months. The DSL has consistent packet loss between 4 and 6%. The only video service that could handle this level of packet loss well was Amazon Prime. Netflix couldn't even load its browse screen until the past two weeks, where something changed, and suddenly Netflix could handle the high packet loss as well as Amazon Prime. Thank…
I'd believe it. When you know that there is going to be packet loss (whether from the user's spotty internet or from internal load-shedding), building your applications to be as resilient as possible to it makes sense. The infrastructure experimentation platform mentioned in the article is probably helpful for sniffing out potential trouble-spots in applications.
Re: Keeping Netflix Reliable Using Prioritized Load Shedding
#6So you can drop all that traffic and the users are unimpacted... So why not just always drop that traffic and don't even bother writing the code to implement those features that clearly nobody cares about?
Re: Keeping Netflix Reliable Using Prioritized Load Shedding
#7I love the phrase "Prioritized Load Shedding" as corporate-speak for "dropping the less-important traffic."
engineer-speak for dropping requests that are less noticeable to users when dropped in order to give services room to recover
Re: Keeping Netflix Reliable Using Prioritized Load Shedding
#8So you can drop all that traffic and the users are unimpacted... So why not just always drop that traffic and don't even bother writing the code to implement those features that clearly nobody cares about?
It's not that nobody cares; they're just less important. Obvious example is analytics. It helps Netflix deliver a better experience to their users, (video watchers and product placement advertisers -- whoever) but it isn't mission critical to making sure users can watch their content. If you miss some log data, that's not a huge deal. That doesn't mean you want to stop logging.
Some of the things they mentioned were also user impacting, like not being able to select a video's language, but less critical. You obviously still want that feature, but it's less important than being able to watch at all.
Re: Keeping Netflix Reliable Using Prioritized Load Shedding
#9I love the phrase "Prioritized Load Shedding" as corporate-speak for "dropping the less-important traffic."
A lot of which is non-essential as well as NON_CRITICAL, like tracking and telemetry
Re: Keeping Netflix Reliable Using Prioritized Load Shedding
#10According to the diagram, Netflix is injecting chaos into the chaos control panel. Is that right?
Looks like the arrow goes the wrong direction.