Distributed systems and calculations being carried out in parallel

• Hamid Reza Khaleghzadeh

  Finding a Memory Tracer for Multi-Threaded applications

  I need an application memory tracer. The tracer must support Multi-threaded applications such as OpenMP and PThread applications and log address of all acceded data by the application. Could you helpI need an application memory tracer. The tracer must support Multi-threaded applications such as OpenMP and PThread applications and log address of all acceded data by the application. Could you help me find an appropriate memory tracer?

  Recent replies ⋅ Show All (5)

  • 
    Hamid Reza Khaleghzadeh replied

    Thanks Theodore for your helpful answer.

    Apr 19, 2012
• Marcelo Alaniz

  anyone has used a Bulk Synchronous Parallel (BSP) Programming model? I need some information about using a BSP Library in a Multilevel Hierarchy

  That's because i'm looking for an affinity of BSP Process on hybrid architectures (clusters Multicore), trying to minimize the latency of barriers. A picture about the idea:That's because i'm looking for an affinity of BSP Process on hybrid architectures (clusters Multicore), trying to minimize the latency of barriers. A picture about the idea: https://www.evernote.com/shard/s125/sh/5bb14c67-a62a-4364-b889-94a6d0a00a87/e95cab4b14546927a8195ca7ec7daaa8

  Recent replies ⋅ Show All (3)

  • 
    Marcelo Alaniz replied

    Thanks you very much. I understand your point of view. But from my side i think like Valiant... today all programming models in parallel computing don't provide a good cost function for know how longThanks you very much. I understand your point of view. But from my side i think like Valiant... today all programming models in parallel computing don't provide a good cost function for know how long will run you program... but BSP has one and the model is great for design algoritms a lot better than PRAM from my point of view. That's why i'm trying to implement a library for support MultiBSP (paper 2010 of Valiant but for hybrid computing - Multicore Clusters). Again thanks for that papers, i will read... Please take a look in this: people.seas.harvard.edu/.../bridging-2010.pdf www.umiacs.umd.edu/~vishkin/T.../Valiant.ppt CheerS!

    Apr 17, 2012
• Ivaylo Penev

  What is the difference between coarse-grained and embarrassing tasks?

  I am doing a research for the possibilities of distributed computing for parallel realization of Monte Carlo simulations and portfolio estimations with large amounts of data. I need to classify theseI am doing a research for the possibilities of distributed computing for parallel realization of Monte Carlo simulations and portfolio estimations with large amounts of data. I need to classify these parallel tasks according to granularity. In the literature coarse-grained and embarrassing parallel tasks are described. But how could I formally differentiate it the tasks are coarse-grained or embarrassing?

  Recent replies ⋅ Show All (4)

  • 
    Ivaylo Penev replied

    Thank you so much, colleagues! Be safe and sound!

    Apr 17, 2012