{"id":42,"date":"2013-10-03T01:41:10","date_gmt":"2013-10-02T17:41:10","guid":{"rendered":"http:\/\/www.lixiaodong.com\/?p=42"},"modified":"2014-02-03T07:07:54","modified_gmt":"2014-02-02T23:07:54","slug":"2pmms2-phase-multiway-merge-sort%e7%ae%97%e6%b3%95java%e7%89%88","status":"publish","type":"post","link":"https:\/\/www.lixiaodong.com\/?p=42","title":{"rendered":"2PMMS(2-phase multiway merge sort)\u7b97\u6cd5java\u5b9e\u73b0"},"content":{"rendered":"<p>2PMMS(2-phase multiway merge sort)\uff0c\u987e\u540d\u601d\u4e49\uff0c\u75312\u4e2a\u9636\u6bb5\u7ec4\u6210\u3002\u7b2c\u4e00\u4e2a\u9636\u6bb5\u5c06\u6e90\u6587\u4ef6\u5206\u6279\u6b21\u8bfb\u5230\u5185\u5b58\u4e2d\uff0c\u91c7\u7528\u67d0\u79cd\u5185\u90e8\u6392\u5e8f\u7b97\u6cd5\u8fdb\u884c\u6392\u5e8f\uff0c\u7136\u540e\u5c06\u6bcf\u6b21\u6392\u5e8f\u540e\u7684\u7ed3\u679c\u5199\u5230\u786c\u76d8\u4e0a\uff0c1\u4e2a\u6279\u6b21\u5bf9\u5e941\u4e2a\u6587\u4ef6\uff1b\u7b2c\u4e8c\u4e2a\u9636\u6bb5\u5c06n\u4e2a\u6587\u4ef6\u6279\u91cf\u8bfb\u5230\u7f13\u5b58\u91cc\uff0c\u4eceN\u4e2a\u7f13\u5b58\u5e8f\u5217\u91cc\u627e\u5230\u6700\u5c0f\u6570\uff0c\u653e\u5230\u8f93\u51fa\u7f13\u5b58\u91cc\u3002\u6bcf\u6b21\u653e\u5b8c\u540e\uff0c\u8981\u68c0\u67e5\u6bcf\u4e2a\u8f93\u5165\u7f13\u5b58\u662f\u5426\u8bfb\u5b8c\uff0c\u5982\u679c\u8bfb\u5b8c\u5219\u4ece\u5bf9\u5e94\u7684\u6587\u4ef6\u91cc\u7ee7\u7eed\u8bfb\u6570\u636e\u5230\u7f13\u5b58\uff0c\u76f4\u5230\u6587\u4ef6\u5168\u90e8\u8bfb\u5b8c\u4e3a\u6b62\uff1b\u5e76\u68c0\u67e5\u8f93\u51fa\u7f13\u5b58\u662f\u5426\u5df2\u6ee1\uff0c\u5982\u6ee1\u5219\u8f93\u51fa\u5230\u6587\u4ef6\u5e76\u6e05\u7a7a\u8f93\u51fa\u7f13\u5b58\u3002<\/p>\n<p>4\u4e2a\u7c7b\uff1a<\/p>\n<p>(1)<\/p>\n<pre class=\"brush: java; title: ; notranslate\" title=\"\">\r\npublic class Main {\r\n\r\n\/*\r\n* 2 arguments\r\n* first argument: inputfile ex: d:\\\\data\\\\input.dat\r\n* second argument: output folder ex: d:\\\\data\r\n*\/\r\npublic static void main(String&#x5B;] args) throws Exception {\r\n    if(args.length!=2){\r\n        System.out.println(&quot;arguments are not valid&quot;);\r\n        return;\r\n    }\r\n    long startTime = System.currentTimeMillis();\r\n    \/\/read data, sort, output sorted data to files\r\n    PhaseOne phaseOne = new PhaseOne();\r\n    List&lt;String&gt; fileNameList = phaseOne.generateSortedFile(args&#x5B;0], args&#x5B;1]);\r\n    \/\/merge\r\n    PhaseTwo phaseTwo = new PhaseTwo();\r\n    phaseTwo.merge(fileNameList, args&#x5B;1]);\r\n    \/\/delete temp files\r\n    for(String fileName: fileNameList){\r\n        new File(fileName).delete();\r\n    }\r\n    long finishTime = System.currentTimeMillis();\r\n    double totalTime = (double)(finishTime - startTime) \/ 1000;\r\n    System.out.println(&quot;total time: &quot; + totalTime + &quot; seconds&quot;);\r\n}\r\n\r\n}\r\n<\/pre>\n<p>(2)<\/p>\n<pre class=\"brush: java; title: ; notranslate\" title=\"\">\r\nimport java.io.BufferedReader;\r\nimport java.io.BufferedWriter;\r\nimport java.io.FileReader;\r\nimport java.io.FileWriter;\r\nimport java.io.IOException;\r\nimport java.util.ArrayList;\r\nimport java.util.List;\r\n\r\npublic class PhaseOne {\r\n\r\nprivate final int amountOnce = 550000;\r\n\r\npublic List&lt;String&gt; generateSortedFile(String inputFile, String outputFolder) {\r\n    List&lt;String&gt; fileNameList = new ArrayList&lt;String&gt;();\r\n    BufferedReader reader = null;\r\n    try {\r\n        reader = new BufferedReader(new FileReader(inputFile));\r\n        String tempString = null;\r\n        int line = 0;\r\n        int&#x5B;] array = new int&#x5B;amountOnce];\r\n        while ((tempString = reader.readLine()) != null) {\r\n        \/\/System.out.println(&quot;line &quot; + line + &quot;: &quot; + tempString);\r\n        array&#x5B;line % amountOnce] = Integer.parseInt(tempString);\r\n        if(line % amountOnce == amountOnce -1){\r\n            String fileNO = (line\/amountOnce)&lt;10 ? &quot;0&quot;+ line\/amountOnce : String.valueOf(line\/amountOnce);\r\n            String fileName = outputFolder+ &quot;\/tmp&quot;+ fileNO+&quot;.dat&quot;;\r\n            writeData(array, line%amountOnce+1, fileName,fileNameList);\r\n        }\r\n        line++;\r\n    }\r\n    String fileNO = (line\/amountOnce)&lt;10 ? &quot;0&quot;+ line\/amountOnce : String.valueOf(line\/amountOnce);\r\n    String fileName = outputFolder+ &quot;\/tmp&quot;+ fileNO+&quot;.dat&quot;;\r\n    writeData(array, line%amountOnce, fileName, fileNameList);\r\n    reader.close();\r\n} catch (IOException e) {\r\n    e.printStackTrace();\r\n} finally {\r\n    if (reader != null) {\r\n    try {\r\n        reader.close();\r\n    }\r\n    catch (IOException e1) {\r\n    }\r\n}\r\nreturn fileNameList;\r\n}\r\n}\r\n\r\nprivate void writeData(int&#x5B;] array, int length, String fileName, List&lt;String&gt; fileNames) throws IOException{\r\n    Quicksort.quickSort(array, length);\r\n    BufferedWriter output = new BufferedWriter(new FileWriter(fileName));\r\n    for(int i=0;i&lt;length;i++){\r\n        output.write(String.valueOf(array&#x5B;i]));\r\n        output.write(&quot;\\r\\n&quot;);\r\n    }\r\n    output.close();\r\n    fileNames.add(fileName);\r\n    array = new int&#x5B;amountOnce];\r\n}\r\n}\r\n<\/pre>\n<p>(3)<\/p>\n<pre class=\"brush: java; title: ; notranslate\" title=\"\">\r\nimport java.io.BufferedReader;\r\nimport java.io.BufferedWriter;\r\nimport java.io.FileOutputStream;\r\nimport java.io.FileReader;\r\nimport java.io.IOException;\r\nimport java.io.OutputStreamWriter;\r\nimport java.util.ArrayList;\r\nimport java.util.List;\r\n\r\npublic class PhaseTwo {\r\n\r\npublic static int inputBufferSize = 30000;\r\n\r\npublic static int outputBufferSize = 100000;\r\n\r\npublic static boolean finishFlag = false;\r\n\r\npublic void merge(List&lt;String&gt; fileNameList, String outputFolder) {\r\n    int&#x5B;] outputBuffer = new int&#x5B;outputBufferSize];\r\n    int&#x5B;]&#x5B;] inputBuffers = new int&#x5B;fileNameList.size()]&#x5B;inputBufferSize];\r\n    List&lt;BufferedReader&gt; readers = new ArrayList&lt;BufferedReader&gt;();\r\n    try {\r\n        for(int i=0;i&lt;fileNameList.size();i++){\r\n            BufferedReader reader = new BufferedReader(new FileReader(fileNameList.get(i)));\r\n            readers.add(reader);\r\n            String tempString = null;\r\n            int line = 0;\r\n            while ((tempString = readers.get(i).readLine()) != null) {\r\n            inputBuffers&#x5B;i]&#x5B;line] = Integer.parseInt(tempString);\r\n            line++;\r\n            if(line &gt;= inputBufferSize){\r\n                break;\r\n            }\r\n        }\r\n    }\r\n\r\n\r\n    int&#x5B;] pointers = new int&#x5B;fileNameList.size()];\r\n\r\n    int outputBufferIndex = 0;\r\n    while(true){\r\n        int min = findMin(inputBuffers,pointers, readers);\r\n        outputBuffer&#x5B;outputBufferIndex % outputBufferSize] = min;\r\n        if(outputBufferIndex % outputBufferSize == outputBufferSize-1){\r\n            outputData(outputBuffer, outputBufferSize, outputFolder+&quot;\/result.dat&quot;);\r\n        }\r\n        outputBufferIndex++;\r\n        if(finishFlag){\r\n            break;\r\n        }\r\n    }\r\n    outputData(outputBuffer, outputBufferIndex%outputBufferSize, outputFolder+&quot;\/result.dat&quot;);\r\n\r\n    for(int i=0;i&lt;fileNameList.size();i++){\r\n        readers.get(i).close();\r\n    }\r\n    System.out.println(&quot;------------------&quot;);\r\n\r\n} catch (IOException e) {\r\n    e.printStackTrace();\r\n} finally {\r\n    for(int i=0;i&lt;fileNameList.size();i++){\r\n    if (readers.get(i) != null) {\r\n    try {\r\n        readers.get(i).close();\r\n    }\r\n    catch (IOException e1) {\r\n    }\r\n}\r\n}\r\n}\r\n}\r\n\r\nprivate int findMin(int&#x5B;]&#x5B;] arrays, int&#x5B;] pointers, List&lt;BufferedReader&gt; readers) throws IOException{\r\n    int min = 0;\r\n    int index = 0;\r\n    for(int i=0;i&lt;arrays.length;i++){\r\n        if(arrays&#x5B;i]&#x5B;pointers&#x5B;i]]&gt;0){\r\n            min = arrays&#x5B;i]&#x5B;pointers&#x5B;i]];\r\n            index = i;\r\n            break;\r\n        }\r\n    }\r\n\r\n    for(int i=index+1;i&lt;arrays.length;i++){\r\n        if(arrays&#x5B;i]&#x5B;pointers&#x5B;i]]&lt;=0){\r\n            continue;\r\n        }\r\n        if(arrays&#x5B;i]&#x5B;pointers&#x5B;i]] &lt; min){\r\n            min = arrays&#x5B;i]&#x5B;pointers&#x5B;i]];\r\n            index = i;\r\n        }\r\n    }\r\n    pointers&#x5B;index] = pointers&#x5B;index]+1;\r\n\r\n    for(int i=0;i&lt;pointers.length;i++){\r\n        if(pointers&#x5B;i]&gt;=inputBufferSize){\r\n            arrays&#x5B;i] = new int&#x5B;inputBufferSize];\r\n            String tempString = null;\r\n            int line = 0;\r\n            while ((tempString = readers.get(i).readLine()) != null) {\r\n                arrays&#x5B;i]&#x5B;line] = Integer.parseInt(tempString);\r\n                line++;\r\n                if(line &gt;= inputBufferSize){\r\n                    break;\r\n                }\r\n            }\r\n            pointers&#x5B;i]= 0;\r\n        }\r\n    } \r\n\r\n    \/\/update finishFlag\r\n    if(!finishFlag){\r\n        boolean flag = true;\r\n        for(int i=0;i&lt;arrays.length;i++){\r\n        if(arrays&#x5B;i]&#x5B;pointers&#x5B;i]]&gt;0){\r\n            flag = false;\r\n            break;\r\n        }\r\n    }\r\n    finishFlag = flag;\r\n}\r\nreturn min;\r\n}\r\n\r\nprivate void outputData(int&#x5B;] outputBuffer, int length, String outputFile) throws IOException{\r\nBufferedWriter output = new BufferedWriter(new OutputStreamWriter(new FileOutputStream(outputFile, true)));\r\nfor(int i=0;i&lt;length;i++){\r\n    StringBuffer sb = new StringBuffer();\r\n    sb.append(outputBuffer&#x5B;i]);\r\n    sb.append(&quot;\\r\\n&quot;);\r\n    output.write(sb.toString());\r\n}\r\noutput.close();\r\noutputBuffer = new int&#x5B;outputBufferSize];\r\n}\r\n\r\n}\r\n<\/pre>\n<p>(4)<\/p>\n<pre class=\"brush: java; title: ; notranslate\" title=\"\">\r\npublic class Quicksort {\r\n    public static void quickSort(int&#x5B;] data, int length) {\r\n        recQuickSort(data, 0, length-1);\r\n    }\r\n\r\nprivate static void recQuickSort(int&#x5B;] data,int left, int right) {\r\n    if(right-left &lt;= 0){\r\n    return;\r\n}\r\nelse {\r\n    long pivot = data&#x5B;right];\r\n    int partition = partitionIt(data, left, right, pivot);\r\n    recQuickSort(data, left, partition-1);\r\n    recQuickSort(data, partition+1, right);\r\n}\r\n}\r\n\r\nprivate static int partitionIt(int&#x5B;] data, int left, int right, long pivot) {\r\n    int leftPtr = left-1;\r\n    int rightPtr = right;\r\n    while(true) {\r\n        while(data&#x5B;++leftPtr] &lt; pivot )\r\n        ;\r\n        while(rightPtr &gt; 0 &amp;&amp; data&#x5B;--rightPtr] &gt; pivot)\r\n        ;\r\n        if(leftPtr &gt;= rightPtr){\r\n        break;\r\n    }\r\n    else{\r\n        swap(data, leftPtr, rightPtr);\r\n    }\r\n}\r\nswap(data, leftPtr, right);\r\n    return leftPtr;\r\n}\r\n\r\nprivate static void swap(int&#x5B;] data, int dex1, int dex2) {\r\n    int temp = data&#x5B;dex1];\r\n    data&#x5B;dex1] = data&#x5B;dex2];\r\n    data&#x5B;dex2] = temp;\r\n}\r\n}\r\n<\/pre>\n","protected":false},"excerpt":{"rendered":"<p>2PMMS(2-phase multiway merge sort)\uff0c\u987e\u540d\u601d\u4e49\uff0c &hellip; <a href=\"https:\/\/www.lixiaodong.com\/?p=42\">\u7ee7\u7eed\u9605\u8bfb <span class=\"meta-nav\">&rarr;<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[8],"tags":[7,42],"class_list":["post-42","post","type-post","status-publish","format-standard","hentry","category-8","tag-2pmms","tag-java"],"_links":{"self":[{"href":"https:\/\/www.lixiaodong.com\/index.php?rest_route=\/wp\/v2\/posts\/42","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.lixiaodong.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.lixiaodong.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.lixiaodong.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.lixiaodong.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=42"}],"version-history":[{"count":7,"href":"https:\/\/www.lixiaodong.com\/index.php?rest_route=\/wp\/v2\/posts\/42\/revisions"}],"predecessor-version":[{"id":129,"href":"https:\/\/www.lixiaodong.com\/index.php?rest_route=\/wp\/v2\/posts\/42\/revisions\/129"}],"wp:attachment":[{"href":"https:\/\/www.lixiaodong.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=42"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.lixiaodong.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=42"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.lixiaodong.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=42"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}