Massively parallel on single core?

 
Thread Tools Search this Thread
Special Forums UNIX and Linux Applications High Performance Computing Massively parallel on single core?
# 1  
Old 02-11-2010
Question Massively parallel on single core?

Hia all,

I am not sure how many people actually follow the HPC forum on unix.com, but you may be interested in discussing the following (academic) problem:

Assume you want to run a *very* large number (say 100.000) of very lightweight synchronous operations. As an example, assume that you want to run 100.000 instances of

Code:
sleep (3600); // thats one hour sleep

The trivial (aka braindead) approach would be

Code:
for ( int i = 0; i < 100000; i++ )
{
  ::sleep (3600);
}

Takes about 15 years to finish ;-)

One could start 1000 threads, and run a sleep in each of them. That reduces the runtime to 100 hours - still 4 days, and the system is totally idle all the time.

So, using more threads? Won't work, as the max-threads-per-process limit will be hit at some point.

So, spawn 100 processes which spawn 1000 threads each?
The max-threads-per-process limit is, on Linux, close to the max-threads-per-system limit, so that won't work. On other Unixes that is different, but I don't think you get 100.000 threads on a normal single CPU system. Do you?

So, what would your approach be?

I am not looking for a sleep replacement: so saying that I should set alarm or something similar is of not much use. Sleep is obviously only an example here - replace it with an extremely lightweight job, like running a very time consuming synchronous remote operation.

I am looking forward to the ideas you guys can come up with! :-)

Cheers, Andre.

Last edited by Andre_Merzky; 02-11-2010 at 08:48 AM.. Reason: layout...
# 2  
Old 02-15-2010
Seems overly academic.....

If practice, most people who have a requirement to run 100,000 parallel applications, they would turn do some distributed processing package, for example cluster management software.

Hardware and existing distributed processing software is cheaper (and more practical) than attempting to design a single-core solution (the title of this thread).

In general, you should design your HPC application as a distributed architecture and make the centralized approach a special case of a distributed architecture.
# 3  
Old 02-15-2010
Hi Neo,

thanks for your reply!

I agree abut your remark as distributed architectures. This is my day-job, and I like it a lot :-)

I did not make the problem clear enough I think: the workload I am talking about are mostly idle jobs, so the CPU and memory load for each job is *very* low. Yes, I can beat the problem with more cores or nodes, but that seems very much like a waste, as those would be all idling most of the time.

Assume you plan for 1000 threads per core, and use quad code nodes - that would require 25 nodes which all idle all day long :-(

Some more detail, if that helps: the idle processes/threads are basically watchers, which represent a CPU/Memory heavy remote job they spawned, and whose state they are watching. Only when that state changes they become active, and kick of data movements or spawn new jobs.

We can't control the design of the remote job startup API very well (third party, synchronous API only), thus our technical options for obtaining state information about those jobs are limited, and boil down to
Code:
void * run_job (void * data)
{
   // this call runs a remote job, and blocks for hours
   remote_api_call (data);
   store_output_data (data);
}

#define NJOBS 100000

int main ()
{
  pthread_t threads[NJOBS]
  for ( int i = 0; i < NJOBS; i++ )
  {
     pthread_create (threads[i],  run_job, ...)
  }

  for ( int i = 0; i < NJOBS; i++ )
  {
     pthread_join (threads[i]);
  }
}

So, I can throw 25 nodes on that large for loop, and that is what we do basically - but what a waste...

The *real* workload are 100.000 CPU/Memory heavy remote jobs, which have sufficient resources to run concurrently. I am talking about the management side (our workflow engine).

Thanks, Andre.
Login or Register to Ask a Question

Previous Thread | Next Thread

9 More Discussions You Might Find Interesting

1. Shell Programming and Scripting

Python GNU parallel single command on multiple cores

Hello, I have a 4 core machine. Here is my initial script cd /work/ python script.py input.txt output.txt 1 2 3 This script runs for 1.5hrs. So I read across the web and figured out that you can use GNU parallel to submit multiple jobs using parallel. But I am not sure if I can run... (4 Replies)
Discussion started by: jacobs.smith
4 Replies

2. Shell Programming and Scripting

Paste 2 single column files to a single file

Hi, I have 2 csv/txt files with single columns. I am trying to merge them using paste, but its not working.. output3.csv: flowerbomb everlon-jewelry sofft steve-madden dolce-gabbana-watchoutput2.csv: http://www1.abc.com/cms/slp/2/Flowerbomb http://www1.abc.com/cms/slp/2/Everlon-Jewelry... (5 Replies)
Discussion started by: ajayakunuri
5 Replies

3. Shell Programming and Scripting

How to replace file massively?

Hi Gurus, I need to change a large amout of file name's. for example: current file name: file_ABCDE_sufix.txt I need to change them as file_FGHIGHKE_sufix.txt. Is there any way I can change them with script. Thanks in advance (1 Reply)
Discussion started by: ken6503
1 Replies

4. Shell Programming and Scripting

For loop in parallel

Hello, My script shell is: for i in $(seq $nb_lignes) do //command java done Please, how can i execute all iteration in parallel ? Thank you so much. (9 Replies)
Discussion started by: chercheur857
9 Replies

5. Shell Programming and Scripting

Multiple lines in a single column to be merged as a single line for a record

Hi, I have a requirement with, No~Dt~Notes 1~2011/08/1~"aaa bbb ccc ddd eee fff ggg hhh" Single column alone got splitted into multiple lines. I require the output as No~Dt~Notes 1~2011/08/1~"aaa<>bbb<>ccc<>ddd<>eee<>fff<>ggg<>hhh" mean to say those new lines to be... (1 Reply)
Discussion started by: Bhuvaneswari
1 Replies

6. Shell Programming and Scripting

Replace single quote with two single quotes in perl

Hi I want to replace single quote with two single quotes in a perl string. If the string is <It's Simpson's book> It should become <It''s Simpson''s book> (3 Replies)
Discussion started by: DushyantG
3 Replies

7. Programming

how to know the application run on which core, and run how many times on this core

I have a dual core pc, I write a application with two child process. I know I can add sched_get_cpu to know the process run on which core, but, it just when the sched_get_cpu is called, it will tell me the result, my quesion is how to know the child proceess spend how many times on one core. (2 Replies)
Discussion started by: yanglei_fage
2 Replies

8. UNIX for Dummies Questions & Answers

difference between Dual-core & Core-to-duo

Can anybody tell What is the exact difference between a Dual-core processor and a Core-to-duo processor ?Advance thanks to all my friends. (1 Reply)
Discussion started by: Ajith kumar.G
1 Replies

9. Shell Programming and Scripting

parallel processing

Hi I want to run two shell script files parallely. These two scripts are interacting with the database. can any body help on this Pls Regards Audippa naidu.M (3 Replies)
Discussion started by: audippa
3 Replies
Login or Register to Ask a Question