c# - Is Interlocked.Decrement(i) right for Parallel.ForEach()? -

-

अतीत में मैंने यह किया है: 

  सूची & gt; आइटम & gt; _Items = GetItems (); Int _CountDown = _Items.Count; (पृष्ठभूमिवर्कर _Worker = नया बैकग्राउंडवर्क ()) का उपयोग करना {_Worker.DoWork + = (s, arg) = & gt; {डोसोमिंग (_Items [_CountDown]); }; _Worker.RunWorkerCompleted + = (s, arg) = & gt; {If (System.Threading.Interlocked.Decrement (ref _CountDown) == 0) RaiseAllDoneEvent (); अन्य _Worker.RunWorkerAsync (); }; _Worker.RunWorkerAsync (); }   
 समानांतर के साथ मैं ऐसा कुछ करना चाहता हूं:  <पूर्व>  सूची & gt; आइटम & gt; _Items = GetItems (); Int _CountDown = _Items.Count; System.Threading.Tasks.Parallel.ForEach (_Items, (i) = & gt; {DoSomething (i); यदि (System.Threading.Interlocked.Decrement (ref _CountDown) == 0) RaiseAllDoneEvent ();});   
 मेरा असली प्रश्न यह है कि  इंटरलॉक किया गया है। अधिग्रहण () यहां सही है?    
 
 
 समानांतर। फोरेईक तब तक अवरुद्ध करे जब तक कि सभी वस्तुओं को संसाधित नहीं किया जाता है- यहां पर करने के लिए सही बात कॉलर से रैसेअले डोनइवेंट () को समानांतर कॉल करने के तुरंत बाद कॉल करने के लिए प्रतीत होती है। FOREach: < / P> 
  सूची & lt; आइटम & gt; _Items = GetItems (); System.Threading.Tasks.Parallel.ForEach (_Items, (i) = & gt; {DoSomething (i);}); RaiseAllDoneEvent ();   
 दूसरे शब्दों में, नीचे गिनने की ज़रूरत नहीं है (या इस बात पर नज़र रखो कि प्रक्रिया में कितनी वस्तुएं हैं) यदि आप समानांतर संचालन को अवरुद्ध नहीं करना चाहते हैं, तो आप इसे कार्य के एक गुच्छा में बदल सकते हैं:  
  सूची & lt; आइटम & gt; _Items = GetItems (); कार्य। फ़ैक्टर। स्टार्टन्यू (() = & gt; {टास्क। प्रतीक्षा (सभी आइटम्स)। चयन (i = & gt; कार्य। फ़ैक्टर। स्टार्टन्यू (() = & gt; डूसममिंग (i)))। ToArray ());})। जारी रखें (t = & gt; RaiseAllDoneEvent ()); उस स्थिति में, आप स्पिन अप करने के लिए एक बाहरी कार्य शुरू करेंगे और फिर कार्य के एक गुच्छा (प्रत्येक आइटम के लिए एक) पर प्रतीक्षा करें, और फिर अंततः सभी पूर्ण घटना बढ़ाएं। इनमें से कोई भी मूल कॉलर को अवरोधित नहीं करेगा।

Comments

Post a Comment

Popular posts from this blog

php - Creating canonical URLs with custom route-classes -

-
I am trying to implement the canonical url and to combine with custom route classes. URL-SACAME is something like this: / category-x / article / 123 / category-y / article / 123 I am creating a custom route - the expansion of the class Zend_Controller_Router_Route_Regex and checks that the article is 123 and the correct category-name is included in the URL. If the article is in 123 square-X and the user is reaching square-y, then I want to redirect to the correct URL. But there is no clear possibility to do this directly to the roads. What is the best practice approach here? I often do this in my action controller like something ... // Get / Category-Y / Article / 123 $ $ article- & gt; URL is generated, and it contains / category-x / article / 123 if (this-> --request-> getRequestUri ()! = $ Article-> URL) {return $ this- & gt; ; _helper-> Redirector-> GoToUrl ($ paragraph-> url); } In this example, $ article-> The url will need ...
Read more

mysql - Formatting problem with PHP DateTime Difference when used as part of an equation -

-
I am trying to write a simple invoicing system. I want to use the Datetime Diff to calculate the cost of each invoice. I have so far: / * In this example $ dayLength and $ ratePerSec is not really the case, but now we use the hypotethical values ​​of 8.0 and 0.0054 Will * / $ date1 = new date time ("Time 1 Form Database", [DateTimeJonof Chase]); $ Date2 = new date time ("Time 2 Form Database", [Datetime Joneof Choice]); $ Diff = $ date2- & gt; Diff ($ date1); $ LenInSec = $ diff- & gt; Format ("% y") * 365 * $ dayLength * 3600 + $ diff- & gt; Format ("% d") * $ dayLength * 3600 + $ diff- & gt; Format ("% H") * 3600 + $ diff- & gt; Format ("% i") * 60 + $ diff- & gt; Format ("% s"); $ Cost = $ lenInSec * $ rate per second; When I try to output cost using $ lenInSec generated from the Deltaime Diff, I actually get weird results. Occasionally, in 1 and 3 out of 1 to 3 times, I get the...
Read more

Mysql Local server hardware -

-
In anticipation of a live scenario, I am creating my entire site, which has a 500 MHz processor with 256 MB A very large 100 MHz RAM I have loaded my database with 200k + rows of garbage data, trying to test my queries. I have seen - obviously because of this, a large display hit all my FAQ using the index, although some 15+ to join (using all indexes) using MySQL To enable 16MB cache, which brings down the query times to nothing but what results? With a dynamic website, I can not see how the cache can help completely when I allow unique queries such as a search feature. I am also using the eaccelerator for PHP. When I use the COUNT (*) total with multiple code JOIN and WHERE , then 10 + There are other queries. Other queries that only get the same line or two do at the same level when I had only a few lines I also noticed that almost 99% of the CPU in CPU CPU stress during a query Comes. Basically I do not know whether the software is on the fault or too much of the h...
Read more