scnresearch.com
SCN Research: Associates
http://www.scnresearch.com/company/associates.html
Shrubbery Networks, Inc.
puliu.blogspot.com
Main BLOGGER: 02/01/2005 - 03/01/2005
http://puliu.blogspot.com/2005_02_01_archive.html
This is Pop's BLOG, for his bad memory. Sunday, February 27, 2005. A constructor taking a single argument is by default an implicit conversion operator:. C(int i);/ constructor and implicit conversion operator. C = 5; / implicit conversion of 5 to a C object and. The compiler re-edits the above sample as if the programmer had written:. C=5;" transformed by the compiler into something like this:. C temp(5);/ temporary object instantiated,. C = temp; / assigned using operator =. Explicit String (int sz);.
elepar.blogspot.com
Elemental Parallelism: January 2014
http://elepar.blogspot.com/2014_01_01_archive.html
Methods for engineering correct, portable, performant, scalable software for concurrent environments (HPC, multicore, cloud, grid, distributed). Wednesday, January 29, 2014. ScalPL 1.1: A few enhancements to the book. This post was slightly edited on 1/30/14 to correct errors in the examples in part 4.]. 1 New Rule: Resource arrays allow any type for indices. In section 13.1.1 of the book (page 230), when introducing arrays, it says (in the second sentence):. In section 13.1.2 (page 232), I say t...Becau...
elepar.blogspot.com
Elemental Parallelism: Big Data, Map-Reduce, and ScalPL
http://elepar.blogspot.com/2014/03/big-data-map-reduce-and-scalpl.html
Methods for engineering correct, portable, performant, scalable software for concurrent environments (HPC, multicore, cloud, grid, distributed). Monday, March 03, 2014. Big Data, Map-Reduce, and ScalPL. There is Big Focus on Big Data these days, and specifically tools (like Hadoop. Possible High-Level Context for MapReduce. Before we get started with the MapReduce. The blocks (of course) may not be stored or indexed contiguously at some low level, but by the time the MapReduce. Indexed by key, and changi...
elepar.blogspot.com
Elemental Parallelism: Actors vs ScalPL tactics: Hidden state vs WYSIWYG
http://elepar.blogspot.com/2012/10/actors-vs-scalpl-tactics-hidden-state.html
Methods for engineering correct, portable, performant, scalable software for concurrent environments (HPC, multicore, cloud, grid, distributed). Friday, October 19, 2012. Actors vs ScalPL tactics: Hidden state vs WYSIWYG. I stumbled into a video. I did not consult much of the more extensive written documentation, like Hewitt's and/or Agha's books. So I do not pretend to be fully up on all aspects of this work.). Those are nice attributes. But they pale in comparison to what ScalPL. And the F-Nets model.
elepar.blogspot.com
Elemental Parallelism: Comparing ScalPL and Concurrent Collections (CnC)
http://elepar.blogspot.com/2014/04/comparing-scalpl-and-concurrent.html
Methods for engineering correct, portable, performant, scalable software for concurrent environments (HPC, multicore, cloud, grid, distributed). Tuesday, April 15, 2014. Comparing ScalPL and Concurrent Collections (CnC). The previous post here, showing a Cholesky decomposition in ScalPL. Together with previous posts here and in the book. CnC Cholesky diagram (See Knobe video). Brief description of CnC. ScalPL Cholesky strategy (see previous post). Some Comparisons with ScalPL. CnC does not allow update-i...
elepar.blogspot.com
Elemental Parallelism: March 2014
http://elepar.blogspot.com/2014_03_01_archive.html
Methods for engineering correct, portable, performant, scalable software for concurrent environments (HPC, multicore, cloud, grid, distributed). Saturday, March 22, 2014. Cholesky Factorization in ScalPL. And a NUMA implementation (Jeannot). It should be noted that the latter effectively computes the result in place, while the former apparently doesn't, but since data can be assumed to move between processors in both cases, the notion of being "in place" is not as clear as one might expect. Other use...
elepar.blogspot.com
Elemental Parallelism: April 2014
http://elepar.blogspot.com/2014_04_01_archive.html
Methods for engineering correct, portable, performant, scalable software for concurrent environments (HPC, multicore, cloud, grid, distributed). Tuesday, April 15, 2014. Comparing ScalPL and Concurrent Collections (CnC). The previous post here, showing a Cholesky decomposition in ScalPL. Together with previous posts here and in the book. CnC Cholesky diagram (See Knobe video). Brief description of CnC. ScalPL Cholesky strategy (see previous post). Some Comparisons with ScalPL. CnC does not allow update-i...
elepar.blogspot.com
Elemental Parallelism: June 2014
http://elepar.blogspot.com/2014_06_01_archive.html
Methods for engineering correct, portable, performant, scalable software for concurrent environments (HPC, multicore, cloud, grid, distributed). Sunday, June 29, 2014. Exciting new directions for ScalPL: Splatr. I haven't posted much over the last few months, in part because I have been working hard in an exciting new direction, and figured it would make more sense to organize and firm up some of those ideas before posting them here. In fact, I have been using a certain workshop. Many of these existing p...
elepar.blogspot.com
Elemental Parallelism: Exciting new directions for ScalPL: Splatr
http://elepar.blogspot.com/2014/06/exciting-new-directions-for-scalpl.html
Methods for engineering correct, portable, performant, scalable software for concurrent environments (HPC, multicore, cloud, grid, distributed). Sunday, June 29, 2014. Exciting new directions for ScalPL: Splatr. I haven't posted much over the last few months, in part because I have been working hard in an exciting new direction, and figured it would make more sense to organize and firm up some of those ideas before posting them here. In fact, I have been using a certain workshop. Many of these existing p...
