hbase scan客户端服务端流程

hbase scan客户端服务端流程

一:基础知识了解:

scanner可分为两种InternalScanner和KeyValueScanner,区别如下

1.InternalScanner,可以理解为包含其他scanner的scanner,它的主要接口为next(),作用是从其包含的scanner中获取下一个KeyValue,它的角色可以理解为雇佣KeyValueScanner

2.KeyValueScanner,从内存或文件中获取KeyValue的scanner,

大致:

scanner, RegionScanner、StoreScanner属于InternalScanner,

而MemstoreScanner、StoreFileScanner、StoreScanner属于KeyValueScanner

二:hbase中client端rpc服务定义:

service ClientService {
    rpc Get (GetRequest) returns (GetResponse);
    rpc Mutate (MutateRequest) returns (MutateResponse);
    rpc Scan (ScanRequest) returns (ScanResponse);
    rpc BulkLoadHFile (BulkLoadHFileRequest) returns (BulkLoadHFileResponse);
    rpc ExecService (CoprocessorServiceRequest) returns (CoprocessorServiceResponse);
    rpc ExecRegionServerService (CoprocessorServiceRequest) returns (CoprocessorServiceResponse);
    rpc Multi (MultiRequest) returns (MultiResponse);
}

关于PB的rpc具体实现可以看源代码。

三:直接代码跟踪了解流程。scan-client客户端源码跟踪

从客户端scan-api代码开始:

table.getScanner(scan)

然后进入HTable的getScanner(final Scan scan)方法

    if (scan.getBatch() > 0 && scan.isSmall()) {
        throw new IllegalArgumentException("Small scan should not be used with batching");
    }

    if (scan.getCaching() <= 0) {
        scan.setCaching(getScannerCaching());
    }
    if (scan.getMaxResultSize() <= 0) {
        scan.setMaxResultSize(scannerMaxResultSize);
    }

    if (scan.isReversed()) {
        if (scan.isSmall()) {
            return new ClientSmallReversedScanner(getConfiguration(), scan, getName(), this.connection, this.rpcCallerFactory, this.rpcControllerFactory, pool, connConfiguration.getReplicaCallTimeoutMicroSecondScan());
        } else {
            return new ReversedClientScanner(getConfiguration(), scan, getName(), this.connection, this.rpcCallerFactory, this.rpcControllerFactory, pool, connConfiguration.getReplicaCallTimeoutMicroSecondScan());
        }
    }

    if (scan.isSmall()) {
        return new ClientSmallScanner(getConfiguration(), scan, getName(), this.connection, this.rpcCallerFactory, this.rpcControllerFactory, pool, connConfiguration.getReplicaCallTimeoutMicroSecondScan());
    } else {
        return new ClientScanner(getConfiguration(), scan, getName(), this.connection, this.rpcCallerFactory, this.rpcControllerFactory, pool, connConfiguration.getReplicaCallTimeoutMicroSecondScan());
    }

根据不同条件执行不同操作,接着进入ClientScanner的构造函数中,

先是属性的取值;再是initializeScannerInConstruction();

进入initializeScannerInConstruction(),发现是一个nextScanner(this.caching, false),初始化scanner;

进接着进入nextscanner;

关闭之前的scanner;下个scanner开始key;判断是否在表的末尾;接着:

        callable = getScannerCallable(localStartKey, nbRows);
        // Open a scanner on the region server starting at the
        // beginning of the region
        call(callable, caller, scannerTimeout);
        this.currentRegion = callable.getHRegionInfo();
        if (this.scanMetrics != null) {
            this.scanMetrics.countOfRegions.incrementAndGet();
        }

生成一个ScannerCallableWithReplicas;

然后利用生成的ScannerCallableWithReplicas通过call函数用来在RS上的region的开始打开一个scanner,扫描数据

进入call方法:

try {
        callable.prepare(false);
        return callable.call(callTimeout);
    }

请求由ScannerCallable.call()发起,

        this.scannerId = openScanner();
        ScanResponse response = getStub().scan(controller, request);
        long id = response.getScannerId();  scannerID存在的话:
        获取服务观返回的数据结果集:
                    response = getStub().scan(controller, request);
                    nextCallSeq++;
                    long timestamp = System.currentTimeMillis();
                    setHeartbeatMessage(response.hasHeartbeatMessage() && response.getHeartbeatMessage());
                    // Results are returned via controller
                    CellScanner cellScanner = controller.cellScanner();
                    rrs = ResponseConverter.getResults(cellScanner, response);

四:scan-server服务端源码跟踪

1.进入HRegion的getScanner(Scan scan, List<KeyValueScanner> additionalScanners, long nonceGroup, long nonce)开始

列族的判断添加及检查

然后返回初始化的RegionScanner: instantiateRegionScanner(scan, additionalScanners, nonceGroup, nonce);

实现在RegionScannerImpl中,初始化scanner

initializeScanners(Scan scan, List<KeyValueScanner> additionalScanners)

根据列族获取对应store:

                Store store = stores.get(entry.getKey());
                KeyValueScanner scanner;
                try {
                    scanner = store.getScanner(scan, entry.getValue(), this.readPt);
                } catch (FileNotFoundException e) {
                    throw handleFileNotFound(e);
                }
                instantiatedScanners.add(scanner);
                if (this.filter == null || !scan.doLoadColumnFamiliesOnDemand() || this.filter.isFamilyEssential(entry.getKey())) {
                    scanners.add(scanner);
                } else {
                    joinedScanners.add(scanner);
                }
            }
            initializeKVHeap(scanners, joinedScanners, region);

通过memstore,snapshot,所有的storefiles打开一scanner,并且不处于compact情形;

然后为一个scan构造一个ScanQueryMatcher

然后根据ScanQueryMatcher执行

seekScanners(scanners, matcher.getStartKey(), explicitColumnQuery && lazySeekEnabledGlobally, parallelSeekEnabled);

包括定位起始row的所有scanner,布隆过滤器的检查删除标记等;

最后会将所有的scanner放入堆中;

2.接着对于客户端的scanner的next()函数服务端实现;

在storesanner中的next:

从store中获取下个row数据;

        do {
        // Update and check the time limit based on the configured value of cellsPerTimeoutCheck
        if ((kvsScanned % cellsPerHeartbeatCheck == 0)) {
            scannerContext.updateTimeProgress();
            if (scannerContext.checkTimeLimit(LimitScope.BETWEEN_CELLS)) {
                return scannerContext.setScannerState(NextState.TIME_LIMIT_REACHED).hasMoreValues();
            }
        }

        if (prevCell != cell)
            ++kvsScanned; // Do object compare - we set prevKV from the same heap.
        checkScanOrder(prevCell, cell, comparator);
        prevCell = cell;

        ScanQueryMatcher.MatchCode qcode = matcher.match(cell);
    其中包含:MatchCode的状态 {
    INCLUDE,
    SKIP,
    NEXT,
    DONE,
    SEEK_NEXT_ROW, 
    SEEK_NEXT_COL,
    DONE_SCAN,
    SEEK_NEXT_USING_HINT,
    INCLUDE_AND_SEEK_NEXT_COL,INCLUDE_AND_SEEK_NEXT_ROW;
    
    } while ((cell = this.heap.peek()) != null);

    if (count > 0) {
        return scannerContext.setScannerState(NextState.MORE_VALUES).hasMoreValues();
    }

    // No more keys
    close();
    return scannerContext.setScannerState(NextState.NO_MORE_VALUES).hasMoreValues();

最终返回匹配的resultsanner,中间含有数据结果集;

五:只是简单地流程跟踪,并没有写出更多的细节实现。可以自己阅读源码解读。

    原文作者:奈文摩尔ST
    原文地址: https://www.jianshu.com/p/3d4058fabf16
    本文转自网络文章,转载此文章仅为分享知识,如有侵权,请联系博主进行删除。
点赞