Skip to Content.
Sympa Menu

charm - Re: [charm] Urgent : Queries Regarding Charm++

charm AT lists.cs.illinois.edu

Subject: Charm++ parallel programming system

List archive

Re: [charm] Urgent : Queries Regarding Charm++


Chronological Thread 
  • From: Raghuvar Prajapati <201351003 AT iiitvadodara.ac.in>
  • To: "Kale, Laxmikant V" <kale AT illinois.edu>
  • Cc: "charm AT lists.cs.illinois.edu" <charm AT lists.cs.illinois.edu>
  • Subject: Re: [charm] Urgent : Queries Regarding Charm++
  • Date: Tue, 11 Apr 2017 16:48:53 +0530

Hello,

The duration of project completion is for only this semester (Actually now only one month left). And We want to run our application (Travelling Salesman Problem) on Charm++ framework in parallel distributed environment. We are facing problems with file handling (File read/write operations) coding syntax and basic structure to create .ci(interface file) and .h(header file) in Charm++ environment. We have gone through your tutorials available on the http://charmplusplus.org/tutorial/  (currently not working properly) and http://charm.cs.illinois.edu/research/charm. But we were unable to find out the relevant applications that could help us to code our application (TSP - Travelling Salesman Problem).

So, can you help us with the issues of prescribed above?

Raghuvar



On Fri, Apr 7, 2017 at 9:48 PM, Kale, Laxmikant V <kale AT illinois.edu> wrote:

People mean different things by “heterogeneous environments”.  Charm allows you to use machines with different configurations (that you mention in Q3). Using machines with different data representations (e.g. big-endian vs little endian… this is rare these days) will require a data conversion.  Number of cores being different is something that requires additional effort currently.  For speed variation, a simple suggestion is to use a refinement-based load balancing strategy. More advanced will be rate-aware load balancers.  Cache and RAM size variations are no problem (and the impact on speed can be handled by load balancing).

(and if you mean use of accelerators such as GPGPUs, that is possible via a specialized interface).

 

What is your time frame for the project? This semester? Or later in the year.

 

          -Sanjay

 

 

Hello,

I am working on the charm++ parallel programming framework for my Undergrad project and I am facing some problems and want to know more about it. So, can you people help me out with these queries?


1. Does charm++ works in the heterogeneous environment?
 2. Can we run charm++ programs in different machines having different architecture simultaneously?
3. What are the parallel programming components that we can deviate in charm++ programs like:
   
(a)    Cache Size (L1/L2 cache), Memory Hierarchy.
(b)    Processing power.
(c)    RAM size.
 (d)   Number of core.

4. If we can accomplish the third step then how (basic idea)?

5. Can we deploy or run Charm++ programs in Smart-phones?

Basically, we want to run an application on distributed mobile environment (distributed mobile).
So is it possible to do the same using charm++ programs?  

Regards.

Raghuvar Prajapati
Indian Institute of Information Technology, Vadodara

Final Year, B.Tech (CSE)

 




--
Raghuvar Prajapati
Indian Institute of Information Technology, Vadodara
Final Year, B.Tech (CSE)





Archive powered by MHonArc 2.6.19.

Top of Page