jsr166y/forkjoin/package-info.java

/*
 * Written by Doug Lea with assistance from members of JCP JSR-166
 * Expert Group and released to the public domain, as explained at
 * http://creativecommons.org/licenses/publicdomain
 */


/**
 * A fine-grained parallel computation framework. ForkJoinTasks and
 * their related support classes provide a very efficient basis for
 * obtaining platform-independent parallel speed-ups of
 * computation-intensive operations.  They are not a full substitute
 * for the kinds of arbitrary processing supported by Executors or
 * Threads. However, when applicable, they typically provide
 * significantly greater performance on multiprocessor platforms.
 *
 * <p> Candidates for fork/join processing mainly include those that
 * can be expressed using parallel divide-and-conquer techniques: To
 * solve a problem, break it in two (or more) parts, and then solve
 * those parts in parallel, continuing on in this way until the
 * problem is too small to be broken up, so is solved directly.  The
 * underlying <em>work-stealing</em> framework makes subtasks
 * available to other threads (normally one per CPU), that help
 * complete the tasks.  In general, the most efficient ForkJoinTasks
 * are those that directly implement this algorithmic design pattern.
 *
 * <p>While direct implementation of parallel divide-and-conquer
 * algorithms is often straightforward, it can also be tedious and
 * code-intensive. For this reason, a number of solution "templates"
 * are available for common kinds of operations on lists and arrays:
 * applying some operation to all elements, combining elements
 * according to some function, and so on.  In this preliminary
 * release, these are presented via some interfaces describing the
 * associated code bodies in TaskTypes, along with an evolving set of
 * implementations for lists and arrays of objects and scalars.
 */
package jsr166y.forkjoin;
Revision:	1.3
Committed:	Tue Jan 6 14:34:59 2009 UTC (15 years, 4 months ago) by dl
Branch:	MAIN
CVS Tags:	HEAD
Changes since 1.2:	+0 -0 lines
State:	*FILE REMOVED*
Log Message:	Repackaging
#	Content
1	/*
2	* Written by Doug Lea with assistance from members of JCP JSR-166
3	* Expert Group and released to the public domain, as explained at
4	* http://creativecommons.org/licenses/publicdomain
5	*/
6
7
8	/**
9	* A fine-grained parallel computation framework. ForkJoinTasks and
10	* their related support classes provide a very efficient basis for
11	* obtaining platform-independent parallel speed-ups of
12	* computation-intensive operations. They are not a full substitute
13	* for the kinds of arbitrary processing supported by Executors or
14	* Threads. However, when applicable, they typically provide
15	* significantly greater performance on multiprocessor platforms.
16	*
17	* <p> Candidates for fork/join processing mainly include those that
18	* can be expressed using parallel divide-and-conquer techniques: To
19	* solve a problem, break it in two (or more) parts, and then solve
20	* those parts in parallel, continuing on in this way until the
21	* problem is too small to be broken up, so is solved directly. The
22	* underlying <em>work-stealing</em> framework makes subtasks
23	* available to other threads (normally one per CPU), that help
24	* complete the tasks. In general, the most efficient ForkJoinTasks
25	* are those that directly implement this algorithmic design pattern.
26	*
27	* <p>While direct implementation of parallel divide-and-conquer
28	* algorithms is often straightforward, it can also be tedious and
29	* code-intensive. For this reason, a number of solution "templates"
30	* are available for common kinds of operations on lists and arrays:
31	* applying some operation to all elements, combining elements
32	* according to some function, and so on. In this preliminary
33	* release, these are presented via some interfaces describing the
34	* associated code bodies in TaskTypes, along with an evolving set of
35	* implementations for lists and arrays of objects and scalars.
36	*/
37	package jsr166y.forkjoin;