What is MapReduce?

Building efficient data centers that can hold thousands of machines is hard enough. Programming thousands of machines is even harder. One approach pioneered by Google is known as MapReduce. MapReduce provides a programming model that simplifies programming thousands of machines by breaking down distributed programs into two steps: map, and reduce.

Credits: Talking: Geoffrey Challen (Assistant Professor, Computer Science and Engineering, University at Buffalo). Producing: Greg Bunyea (Undergraduate, Computer Science and Engineering, University at Buffalo).

[test-mmbv]

thanks for explaining this so clearly.
this is one of the rare explanations in youtube that isn't inane or polluted with pointless jargon.

[kevinohare]

A simple topic made unnecessarily difficult to understand by others poor explanation. Thanks so much, concept is now crystal clear to me.

[Rahul_.]

This is the single best explanation on the topic on YouTube. It helped me 6 years ago and it still holds up now that I'm revising. Thanks!

[standcontractdelta]

FINALLY. This made sense. I watched a few others try to explain it in the weirdest way, immediately jumping into details instead of giving a bigger picture first.

[johnny]

straight to the point and specific, great explanation

[newbie]

This the most explicit and best explanation on MapReduce concept ever!!!

[arocco]

How random. Came here looking to learn more about Map Reduce before a Facebook interview next week, and the guy doing the video ends up being one of my old CS professors.

[ciloc]

This video was super helpful for me!! Not only talked about the counting example but also explained that there are some tasks that do not fit in this framework!! Now I think I got the concept!! Thanks!!

[skipa]

I still refer to this video and I get to understand MR better with every play, if only I could like this video every time I watched it. Thank you for the content.

[igcc]

Finally a good professor on this matter.

[jotarokuzo]

Thank you so much for the simple yet effective visuals. Perfect for people who just need the basic knowledge before stepping into all of the complicated parts.

[fostercityjoe]

I just shed a tear. That was beautiful!

[tshaylatte]

Once you understand the why, it is so much clear. Thank you

[aemethes]

Brilliantly put, I must say, when you explain it like that, it makes our understanding of distributed computing much easier to understand. Saving your video is a play list for future references and sharing. thank you~

[xianshuzeng]

Thanks a lot! I struggled a bit to learn what is map reduce from documents. This video makes everything so clear!

[Alexandra-heol]

Best explanation in the YouTube! Thanks🙏🏻

[high_fly_bird]

so cool, so consistent! Five minutes and I already understand the essence! Thank u so much!

[Shermila]

Got a clear idea about it. Thanks from India.

[ziad]

High quality explanation in a simple way.

[kenii]

amazing. cramming for a parallel and distributed systems in an hour. thanks so much!