Talk on Parallel Computing

Parallel computing used to be the province of scientists and engineers. The tools and algorithms to do this has always been available on the net. However, since the advent of multicore machines, there is now a compelling reason for all programmers to learn and master it. Parallel programming is already the future of computing and there future is now.

I remember way back in 1998, I was introduced to parallel computing in MPI. I also wrote some papers on parallel algorithms and still active in parallel algorithms research. Now I am advocating and teaching a lot of people in parallel computing.


Published by

Bobby Corpus

Loves to Compute!

13 thoughts on “Talk on Parallel Computing”

  1. sige, if my talks with my other research cool-leagues and cool-labor-rationists will push thru, I’ll push for parallelization of the many heuristic hybrids we can use for our computational researches, especially if we’ll have population-based heuristics in the hybridization.

    Parallelization and hybridization of heuristics have many powerful applications in extreme computing we can work together in the many researches we can pursue later,if those whom I try to inspire will catch the fire!!!

    Pero, please huwag isama ang mga puta na kilala mo na sino sila na dati ko ng kinasusuklaman na magaling lang magamit ng tao, as in friendly-user na mga tae!!!

  2. UP Diliman EEE Dept. will be offering a UC Berkeley course on Parallel Programming this coming semester. 🙂

  3. Hi uProcessor, that’s great news! We are already in the age of multicore yet our programmers don’t have the skills yet. BTW, I used to teach in EEE also way back 1998 on Probability theory. During that time Ms. Guevarra just came back from states and Dayco was outgoing chairman.

  4. Boobby yeow,

    try to put in population-based heuristics, e.g., GA, EA, etc…, for faster global search, then hybridize with a local fine tuning via like SA, TS, or whatever is good for local search, and then parallelize the hybrid. so far, in adora’s thesis, that is the best kind of heuristic to have even like perfect optimization, and if you can remember, ZERO conflicts in the timetabling settings!!!!



  5. baka marinig itong ating tete-a-tete via the internet ng mga friendly users sa math!!!! gagamitin lang tayo for their stupid idiot quixotic searches, like one problem they have now that has a very elegant model, but so poor in computational approaches!!! hirap if they will pick your brains for their use!!!

  6. we can still further fine tune and tweak what we did in adora’s thesis, kasi madaming mga heuristics na puede natin i-hybrid and parallelize!!!

    If you can go to UP, ask Joy to let you xerox Geoffrey Chua’s thesis, kasi puede natin i-parallelize ang mga hybrid heuristic doon. Maganda din ang mga statistics namin doon. Sana, if meron pa kaming time (graduating na kasi siya noon), madami pa kaming magawa, like more experimental designs and regression.

    Sonia’s thesis puede rin i-parallelize … actually, lahat na pina-thesis ko na merong mga population-based heuristics (e.g., GA, EA, etc…) can be parallelized, together with their hybrids with local searches (e.g., SA, TS, etc..) that can further fine tune and tweak the heuristic search.

    Can you like put some kind of computer mechanism here in your site where I’ll be alerted in my gmail if you have posted some more comments, so that I can read and reply and react kaagad? Maybe, ask the webmaster of multiply how they do such alerts in the email, so that mabilis ang ating tete-a-tete sa internet!!!

    Actually, even when I was still in UP, I had long already tried to encourage my friends and advisees, like one in EEE, to try parallel programming. Pero, wala naman akong clout, kasi hindi naman ako politico like alam no na!!! Kayo lang ang puede mag-trailblaze nito this time.

  7. Hi Auggie,

    I think whenever you post it will immediately email to your gmail account whenever i reply. Did you take a look at the blog I created for you? Maybe you can tell your students to post their thesis there.

  8. wala akong natatanggap na mga notice about your replies, unlike multiply.

    mahirap yata if i-post natin ang mga theses ng mga advisees ko, kasi nakawin ang ating mga ideas by alam mo na kung sinong mga pirata sa UP. For our consumption na lang, kasi mahirap na maunahan at manakawan tayo ng ideas at tauhan!!!

    By the way, how about setting up some repository of all the literature, papers, links, even theses that we can surf into when we discuss and do our computational adventures? Kaya lang, dapat exclusive lang sa mga members, kasi as I just have discussed above, merong mga magnanakaw at manunulot ng ating mga ideas and tauhan!!! Hindi sila nice to even acknowledge what they plagiarize!!! very despicable kasi!!!!

  9. Yes, having a repository would be a good idea. I’ll see what kind of software i’ll install to do that.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s