http://erlang-china.org erlang 中文社区 Wed, 28 Jul 2010 06:52:46 +0000 hourly 1 http://wordpress.org/?v=3.0 http://erlang-china.org/help/comment-page-1#comment-9867 棘手问题高手请帮忙解答 Fri, 02 Jan 2009 12:55:27 +0000 http://erlang-china.org/help/#comment-9867 The goal of this assignment is to make a distributed fault-tolerant application for a simple computation of prime numbers. (Large prime numbers are user in security algorithms, if you like to know a practical application of this algorithm). A prime number is a number that is only dividable by 1 and itself, i.e. for N to be prime it must hold that for all M between 2 and N-1 the erlang expression N rem M evaluates to something unequal zero. The first few prime numbers are 1,2,3,5,7,11,13,… Write a demonstrator for a fault-tolerant distributed application. A number of computers should be connected in a network, say 1 to N. It is known how they are positioned, i.e., 1 nest to 2, next to 3, … next to N. On each computer you run the same program. This program computes the prime numbers, but it skips numbers that other computers have already analyzed. Thus, if computer 1 starts to compute whether 1 is a prime, then it informs the others, which decide to go on to the next number (2, 3, 4, etc). In the end, all computers print a list of all prime numbers, but never compute a prime number if one of the other computers already took that task. As soon as a computer is disconnected from the network, or the Erlang node that it runs is killed (simulating a crash), then the other computers continue working. When a crashed computer is restarted (or the Erlang node restarted), the program joins the pool and starts doing computations again. The goal of this assignment is to make a distributed fault-tolerant application for a simple
computation of prime numbers. (Large prime numbers are user in security algorithms, if you
like to know a practical application of this algorithm).
A prime number is a number that is only dividable by 1 and itself, i.e. for N to be prime it
must hold that for all M between 2 and N-1 the erlang expression N rem M evaluates to
something unequal zero. The first few prime numbers are 1,2,3,5,7,11,13,…
Write a demonstrator for a fault-tolerant distributed application. A number of computers
should be connected in a network, say 1 to N.
It is known how they are positioned, i.e., 1 nest to 2, next to 3, … next to N.
On each computer you run the same program. This program computes the prime numbers, but
it skips numbers that other computers have already analyzed. Thus, if computer 1 starts to
compute whether 1 is a prime, then it informs the others, which decide to go on to the next
number (2, 3, 4, etc). In the end, all computers print a list of all prime numbers, but never
compute a prime number if one of the other computers already took that task.
As soon as a computer is disconnected from the network, or the Erlang node that it runs is
killed (simulating a crash), then the other computers continue working.
When a crashed computer is restarted (or the Erlang node restarted), the program joins the
pool and starts doing computations again.

]]> http://erlang-china.org/help/comment-page-1#comment-49 Erlang-China » Blog Archive » 加了Erlang代码的语法加亮，show一下 Mon, 16 Apr 2007 13:15:36 +0000 http://erlang-china.org/help/#comment-49 [...] 完整的示例可以参考这里。 [...] [...] 完整的示例可以参考这里。 [...]

]]>