Added: 4 years ago
From: GoogleDevelopers
Views: 121,826
Sort by time | Sort by thread (beta)

Link to this comment:

Share to:
see all

All Comments (61)

Sign In or Sign Up now to post a comment!
  • hy i want to create a distributed clustr plz provide some information...

  • pwede ko mangayo sa imong slide? please!

  • Ah, very interesting. I was wondering how synchronization primitives avoided being subject to race conditions themselves. :)

  • I have very little knowledge of parallel/distributed computing or even some of the lower level network behavior that was covered.

    However you maintained my interest and the material was explained clearly and concisely. Awesome stuff.

  • I love how concise this is

  • thanks

  • good stuff but the network part was a bit out of whack ...leave that to network guys ;-)

  • can't post the link for the slides here it seems, but if you remove _content_ from the provided link, it should work (or a google search :) )

  • All the slides and notes can be found by googling "mapreduce-minilecture listing" or you can just go here: bit.ly forwardslash 44YUv

  • Strange lecture mixing advanced and trivial concepts.

  • +1 for a slide on Parallel Vs Distributed Computing !!

    I worked on a parallel computing distributed real time environment : MPI based cluster of nodes with multicore processing capabilities..

  • I recomend this lecture to students offering Distributed & Parallel Systems Paper Of the BRITISH COMPUTER SOCIETY PGD LEVEL EXAM good intro Thanks R.Y.Maheswaran

  • Good video! Thanks google.

    P.S. Most of the audience in this lecture seems to be clueless hah.

  • Of course plenty of people remember it. He also mentioned it somewhere around the 18 minute mark.

  • turnin go boilers

  • Hey Can anybodytell me where to start if iam new to hadoop. I have an idea of scaling biologcial dataset using Hadoop. Can I know If anyone is interested in life science ?

  • @yogiprasanna If you like Java and getting your hands dirty then have a look at HadoopStudio by Karmasphere. It may help you visualise the map reduce process. Also look at the stuff written by Cloudera (who Aaron now works for I believe)

  • Great Information for newbie like me, Thanks Aaron. Please Doing Some more.

  • The requested URL /edu/content/submissions/mapre­duce-minilecture/listing.html was not found on this server.

  • no one remembers SETI anymore?

  • lol the solemn near-silence, punctuated only by the whiteboard marker's scritching, as he laboriously hand-writes an example

  • Great vid, Aaron's cool, though I get slightly annoyed when the young intellectuals think that over enunciation makes them sound more intelligent.

  • i agree especially with words that end in t

  • Surely you're kidding... right? Proper enunciation is merely the individuals ability to apply grade school grammar when speaking. I find it frustrating that speaking properly (especially in a lecture) somehow suggests a 'simulated' intellect as apposed to taking grammatical shortcuts. I mean seriously.... find another reason to Judge the youth of this Country.... Grammatical accuracy is a completely ridiculous reason.

  • I further cant understand  how you would even comment on something so trivial when you have a large population of people who cant seem to speak a word of Proper English. Ebonics clearly demonstrates this point.... If only we were more concerned with what really matters. With all of the youth and young adults that have found the hopelessness of drugs and substance abuse, you find it most prudent to comment on over-enunciation..? Seriously..?

  • sockets are now the standard whey to do inter program communication even on local machines....

    think DBus...

  • excellent lecture

  • Nice material, except...

    There is a deadlock in the example slide around 28:45

    Good refresher, however.

  • great obs, if bar() locks the semaphore before foo() begins, then they deadlock.

  • MapReduce is really cool...but I think the future is combining structured and unstructured data by integrating with scalable databases...that's how you can really popularize analytics - check out Aster Data for example, they just announced a MapReduce database.

  • great vid! keep up with the good work!

  • Mr/Ms GoogleDevelopers, the lecture slides are not available on the link. Can you provide a new link

  • @rational8

    remove content from the given link

    

  • The correct link for Lecture Slides code.google.com/edu/submission­s/mapreduce-minilecture/listin­g.html

  • Good presentation. I liked the historical slides the most. This issue has been around for many many years. Good effort.

  • Can the dual core processing be considered as an application of distributed computing? I don't have time to watch the whole lecture and I admit it's somewhat fool question :)

  • As I understand it, dual core processing is parallel computing as both processors share the same memory resources.

  • Does this mean that cluster or gucha is a form of Distributed computing and not parallel computing?

  • Wauw no questions. Brilliant audience.

  • This obviously isn't a new concept but its good to see google pursuing it. Look at Standford Folding@home project...just imagine what you could do with all the excess CPU power on home computers these days. I envision a time when all home computers are running some sort of beneficial cpu utilization program.

  • Very interesting stuff! I have learned a lot already in the first 10 minutes.

  • Nice course! :)  I liked it!

  • Muy buenas estas charlas, ya si las traducis al castellano, o simplemente consubtitulos, seriais unos cracks!!

    Very good!

  • i just love the good work that google is doing and pls continue to add such informative videos!!!

  • amen

  • başka dilelrdede olsa şu şeyler altyazı felan geçse ne güzel olur :(

  • its great, i have this idea near half year back / здорово , ум еня была такая же идея полгода назад . =)

  • классно у меня такая же идея была гдето полгода назад =)

  • In the "Final Example" slide where it shows Thread 1 calling foo() and Thread 2 calling bar() there is a deadlock condition if bar() is called first. bar() locks sem and waits for the fooFinishedCV to be notified (sem is still locked). foo() gets to run, but only for a short while because it is blocked when trying to lock sem. foo() never is able to notify the fooFinishedCV and both threads are in a deadlock condition :(

  • I guess I should have kept listening, the CV unlocks sem :(

  • good.

    what i can do using it?

  • Datamine those thousands of fMRI scans you have lying around for your next neuroscience meta-analysis.

    ;)

  • You should better identify problems and solve your own rather than blaming others.

  • You don't view streaming video on ADSL, you use cable for that ;)

  • I use ADSL and I "view streaming video"....

Loading...
0 / 00Unsaved Playlist Return to active list
    1. Your queue is empty. Add videos to your queue using this button:
      or sign in to load a different list.
    Loading...Loading...Saving...
    • Clear all videos from this list
    • Learn more