lucene-对多个索引的搜索和多线程搜索

1、如果应用程序架构由多个LUCENE索引组成，则可以通过MutltiSearcher把所有索引搜索。也可以通过ParallelMultiSearcher进行多线程搜索。在单核的情况下，MultiSearcher比ParallelMultiSearcher性能更高。

2、MultiSearcher

搜索2个搜索，把动物按首字母在字母表中的位置分成2部分，一部分一个索引

public class MultisearcherTest extends TestCase{

   private Indexsearcher[]searchers;

   public void setUp() throwsException{ 

       String[] animals={"aardvark","beaver","coati","dog","lemur",

                         "python","vicuna","zebra"};

       Ananlyzer analyzer=ne WhitespaceAnalyzer();

       Directory aTomDirectory =new RAMDirectory();

       Directory nTOzDirectory=new RAMDirectory();

//建立2个索引

        IndexWriteraTomwriter=new IndexWriter(atomDirectory,analyzer,true);

        IndexWriternTozwriter=new IndexWriter(aTozDirectory,analyzer,true);

        

        for(int i=0;i<anaimals.length;i+){

           Document doc=new Document();

            Stringanimal=animals[i];

            doc.add(Filed.Keyword("animal",animal));

           if (animal.compareToIgnoreCase("n")<0){

              aTomWriter.addDocument(doc);//前半部分索引a-m

           }else{

              nTozWriter.addDocument(doc);//后半部分索引 n-z

            }

       }

       

        aTomwriter.close();

       nTozwriter.close();

       searchers=newIndexsearcher[2];

       searcher[0]=new IndexSearcher(aTOmDirectory);

       searcher[1]=new IndexSearhcer(nTOzDirectory);

 

      

       }

       public void testMulti() throws Exception{

           MultiSearcher searcher=new MultiSearcher(searchers);

            //对2个索引进行搜索

           Query query=new RangeQuery(new Term("animal","h"),newTerm("animal","t"),true);

           Hits hits= searcher.search(query);

        }

 }

3、ParallelMultiSearcher多线程搜索

   搜索操作为每个Searchable分配一个线程，直到所有线程都完成其搜索。基本搜索和进行过滤的搜索是并行执行的。

  lucene通过RMI为用户提供搜索远程索引功能。

  RMI服务器绑定了一个RemoteSearchable的实例，它和IndexSearcher、MultiSearch一样，实现Searchable接口

 

1)把文档按26个字母切分为26个索引。服务器端向客户端提供2个RMI调用

public class SearchServer{

    privatestatic final String ALPHABET="abcdefghijklmnopqrstuvwxyz";

    publicstatic void main(String[] args) throws Exceptino{

       if (args.length!=1){

           System.err.printLn("Usage:Searchserver<basedir>");

            System.exit(-1);

        }

       String basedir=args[0];

//为每个索引建立一个IndexSearcher对象

       Searchable[] searchables=new Searchable[ALPHABET.length()];

       for (int i=0;i<ALPHABET.length;i++){

            searchables[]=new IndexSearcher(newFile(basedir,""+ALPHABET.charAt(i)).getAbsolutePath());              

        }    

//注册可供客户端调用服务的端口

       LocateRegistry.createRegistry(1099);

//使用     multiSearcher完成所有索引的搜索  

        SearchermultiSearcher=new MultiSearcher(searchables);

        RemoteSearchablemultiImpl=new RemoteSearchables(multiSearcher);

       Naming.rebind("//localhost/LIA_Multi",multiImpl);//注册RMI方法

//使用   parallelSearcher   完成搜索 

       Searcher parallelSearcher=newParallelMultiSearcher(searchables);

        RemoteSearchableparallelImpl=new RemoteSearchables(parallelSearcher);

       Naming.rebind("//localhost/LIA_Parallel",parallelImpl);//注册RMI方法       

       System.out.println("server started");

        }

    }

2)客户端

public class SearchClient{

   private static HashMapsearchercache=new HashMap();

  

   public static voidmain(String[] args) throws Exception{

     if (args.length!=1){

        System.err.println("Usage:SearchClient<query>");

        System.exit(-1);

     }

      Stringword=args[0];

     for (int i=0;i<5;i++){

         search("LIA_Multi",word);//调用服务器的multi方法搜索

         search("LIA_Multi",word);//调用服务器的multi方法搜索       

     }      
  }

  private static void search(String name,Stringword) throws Exception{

     TermQuery query=new TermQuery(new Term("word",word));

     MultiSearcher searcher=(MultiSearcher)searcherCache.get(name);//检查缓存中是否有该搜索器，该搜索器是带缓存功能的

     if (searcher==null){//没有该搜索，则生成新的搜索

          searcher=new MultiSearcher(newSearchable[]{lookupRemote(name)});

           searcherCache.put(name,searcher);

      }

//统计时间

     long begin=new Date().getTime();

     Hits hits=searcher.search(query);

     long end=new Date().getTime()

     

     ...........

     ...........

      //不要关闭searcher对象

    }

    private static  Searchable lookupRemote(Stringname) throws Exception{

         return (Searchable) Naming.lookup("//localhost/"+name);

    }

}