The Dataflow Model: Balancing Correctness, Latency, and Cost in Data Processing [pdf]

http://static.googleusercontent.com/media/research.google.com/en//pubs/archive/43864.pdf

by timclark • 11 years ago

56 12 11 years ago

static.googleusercontent.com

dang2• 11 years ago

Url changed from http://blog.acolyer.org/2015/08/18/the-dataflow-model-a-prac..., which points to this.

zweiterlinde• 11 years ago

This is unfortunate---Colyer's summaries are well worth reading.

dang2• 11 years ago

They're good summaries, but after thinking about this a fair bit, I don't think they're enough to trump HN's preference for original sources. Anyone who wants to can read both, since when we change a URL we include the previous one in the comments, as above.

chrisseaton• 11 years ago

Why is there no related work section in this paper? I'm not sure simply calling it 'The' dataflow model is very friendly either to all the other previous dataflow models for parallelism that have been developed over the last four decades or so. Why can this implementation be the definitive one so much that it doesn't even need a qualified name and why aren't any of the others even worth a mention?

demian• 11 years ago

IMHO the same thing happened with the "object" concept in the '90s.

scott_s• 11 years ago

The introduction is essentially the related works section.

obulpathi• 11 years ago

Wow ... this is awesome! Quoting from the paper, "live and breathe under the assumption that we will never know if or when we have seen all of our data, only that new data will arrive, old data may be retracted, and the only way to make this problem tractable". That's another amazing mindshift!

eternalban• 11 years ago

Fundamentally still operating in an 'anticipatory' [1] model of computing. A trending shade of pink lipstick for the old pig.

[1]: the margin of this post is too small to contain an elaboration on this ;)

wslh• 11 years ago

Eric Schmidt is in the list of authors, I wonder if he is doing CS research again as an executive.

trequartista• 11 years ago

And his work email is cloude@google.com? That is quirky

chubot• 11 years ago

There are multiple Eric Schmidt's at Google. That one is not the Chairman/former CEO :)

nfoz• 11 years ago

Wow. Just imagine all the opportunity for social engineering, with a name like that :)

The Dataflow Model: Balancing Correctness, Latency, and Cost in Data Processing [pdf]

Related Stories