Parallel

Syndicate content 2 

Attached code contains a very basic $system.WorkMgr example.

It uses several jobs (workers) to update different chunks of rows of a table.

Steps:

  • Creates a table with 100 records.
  • Split table  in chunks 
  • Initialize WorkMgr and queue chunks to workers.
    • Every worker simply sets its process number in the Job field of the processed row. 

In this case, I have tested the example in a 8-core laptop

Last comment 27 October 2018
0   0 3
378

views

0

rating

Hi

I'm starting work with the Atelier. 

Starting from the point that Eclipse uses local files on the workstation, COS development can be versioned with market tools (For example GIT or Subversion).

I would like to know how the community has worked with code versioning, to create a consistent versioning model.

I initially have the following doubts:

  1.  Creation of WorkSpaces and Projects
  2.  Merge files
  3. Change Branch Functional Development
  4. Promotion of Environments: Development -> Homologation -> Production
  5. Generating Historical Tags

Thank you all.

This post has been translated by Google Translate, sorry for writing errors

Last answer 13 September 2018
+ 5   0 2
0

comments

195

views

+ 5

rating

Hello community,

I need to perform some processing-heavy operations on a set of input objects and I would like to utilize more than a single processor core to do the heavy lifting.

Is there any way to call methods in parallel and wait for the results in a blocking way?

Basically, I am looking for an equivalent of the pythonese

with Pool(n) as p:
    results = p.map(function, data)

Thanks

Jiri

EDIT: Correct English :)

Last answer 9 March 2018 Last comment 9 March 2018
+ 1   0 2
231

views

+ 1

rating

Several years ago everyone got mad about BigData – nobody knew when smallish data will become BIGDATA, but all knows that it’s trendy and the way to go. Time passed, BigData is not a buzz anymore (most of us missed the moment when Gartner has removed BigData term from their 2016 buzzword 2016 curve http://www.kdnuggets.com/2015/08/gartner-2015-hype-cycle-big-data-is-out-machine-learning-is-in.html), so it’s probably a good time to look back and realize what it is (what it was)…

When it becomes “BigData”?

Let’s start from the beginning: what is the moment when “not so big data” becomes BigData? Here was the answer in 2015 from David Kanter[1], one of most respected, well known x86 architecture specialist

Last comment 28 March 2017
+ 6   0 3
1183

views

+ 6

rating

In part I of this series we have introduced MapReduce as a generic concept, and in part II we started to approach Caché ObjectScript implementation via introducing abstract interfaces. Now we will try to provide more concrete examples of applications using MapReduce.

Last comment 28 March 2017
+ 4   0 4
581

views

+ 4

rating