Search in sources :

Example 6 with AggregateResponse

use of org.apache.hadoop.hbase.protobuf.generated.AggregateProtos.AggregateResponse in project hbase by apache.

the class AggregateImplementation method getMedian.

/**
   * Gives a List containing sum of values and sum of weights.
   * It is computed for the combination of column
   * family and column qualifier(s) in the given row range as defined in the
   * Scan object. In its current implementation, it takes one column family and
   * two column qualifiers. The first qualifier is for values column and 
   * the second qualifier (optional) is for weight column.
   */
@Override
public void getMedian(RpcController controller, AggregateRequest request, RpcCallback<AggregateResponse> done) {
    AggregateResponse response = null;
    InternalScanner scanner = null;
    try {
        ColumnInterpreter<T, S, P, Q, R> ci = constructColumnInterpreterFromRequest(request);
        S sumVal = null, sumWeights = null, tempVal = null, tempWeight = null;
        Scan scan = ProtobufUtil.toScan(request.getScan());
        scanner = env.getRegion().getScanner(scan);
        byte[] colFamily = scan.getFamilies()[0];
        NavigableSet<byte[]> qualifiers = scan.getFamilyMap().get(colFamily);
        byte[] valQualifier = null, weightQualifier = null;
        if (qualifiers != null && !qualifiers.isEmpty()) {
            valQualifier = qualifiers.pollFirst();
            // if weighted median is requested, get qualifier for the weight column
            weightQualifier = qualifiers.pollLast();
        }
        List<Cell> results = new ArrayList<>();
        boolean hasMoreRows = false;
        do {
            tempVal = null;
            tempWeight = null;
            hasMoreRows = scanner.next(results);
            int listSize = results.size();
            for (int i = 0; i < listSize; i++) {
                Cell kv = results.get(i);
                tempVal = ci.add(tempVal, ci.castToReturnType(ci.getValue(colFamily, valQualifier, kv)));
                if (weightQualifier != null) {
                    tempWeight = ci.add(tempWeight, ci.castToReturnType(ci.getValue(colFamily, weightQualifier, kv)));
                }
            }
            results.clear();
            sumVal = ci.add(sumVal, tempVal);
            sumWeights = ci.add(sumWeights, tempWeight);
        } while (hasMoreRows);
        ByteString first_sumVal = ci.getProtoForPromotedType(sumVal).toByteString();
        S s = sumWeights == null ? ci.castToReturnType(ci.getMinValue()) : sumWeights;
        ByteString first_sumWeights = ci.getProtoForPromotedType(s).toByteString();
        AggregateResponse.Builder pair = AggregateResponse.newBuilder();
        pair.addFirstPart(first_sumVal);
        pair.addFirstPart(first_sumWeights);
        response = pair.build();
    } catch (IOException e) {
        CoprocessorRpcUtils.setControllerException(controller, e);
    } finally {
        if (scanner != null) {
            try {
                scanner.close();
            } catch (IOException ignored) {
            }
        }
    }
    done.run(response);
}
Also used : InternalScanner(org.apache.hadoop.hbase.regionserver.InternalScanner) ByteString(com.google.protobuf.ByteString) AggregateResponse(org.apache.hadoop.hbase.protobuf.generated.AggregateProtos.AggregateResponse) ArrayList(java.util.ArrayList) IOException(java.io.IOException) Scan(org.apache.hadoop.hbase.client.Scan) Cell(org.apache.hadoop.hbase.Cell)

Example 7 with AggregateResponse

use of org.apache.hadoop.hbase.protobuf.generated.AggregateProtos.AggregateResponse in project hbase by apache.

the class AggregationClient method min.

/**
   * It gives the minimum value of a column for a given column family for the
   * given range. In case qualifier is null, a min of all values for the given
   * family is returned.
   * @param table
   * @param ci
   * @param scan
   * @return min val &lt;R&gt;
   * @throws Throwable
   */
public <R, S, P extends Message, Q extends Message, T extends Message> R min(final Table table, final ColumnInterpreter<R, S, P, Q, T> ci, final Scan scan) throws Throwable {
    final AggregateRequest requestArg = validateArgAndGetPB(scan, ci, false);
    class MinCallBack implements Batch.Callback<R> {

        private R min = null;

        public R getMinimum() {
            return min;
        }

        @Override
        public synchronized void update(byte[] region, byte[] row, R result) {
            min = (min == null || (result != null && ci.compare(result, min) < 0)) ? result : min;
        }
    }
    MinCallBack minCallBack = new MinCallBack();
    table.coprocessorService(AggregateService.class, scan.getStartRow(), scan.getStopRow(), new Batch.Call<AggregateService, R>() {

        @Override
        public R call(AggregateService instance) throws IOException {
            RpcController controller = new AggregationClientRpcController();
            CoprocessorRpcUtils.BlockingRpcCallback<AggregateResponse> rpcCallback = new CoprocessorRpcUtils.BlockingRpcCallback<>();
            instance.getMin(controller, requestArg, rpcCallback);
            AggregateResponse response = rpcCallback.get();
            if (controller.failed()) {
                throw new IOException(controller.errorText());
            }
            if (response.getFirstPartCount() > 0) {
                ByteString b = response.getFirstPart(0);
                Q q = getParsedGenericInstance(ci.getClass(), 3, b);
                return ci.getCellValueFromProto(q);
            }
            return null;
        }
    }, minCallBack);
    log.debug("Min fom all regions is: " + minCallBack.getMinimum());
    return minCallBack.getMinimum();
}
Also used : AggregateRequest(org.apache.hadoop.hbase.protobuf.generated.AggregateProtos.AggregateRequest) ByteString(com.google.protobuf.ByteString) AggregateResponse(org.apache.hadoop.hbase.protobuf.generated.AggregateProtos.AggregateResponse) IOException(java.io.IOException) RpcController(com.google.protobuf.RpcController) RpcCallback(com.google.protobuf.RpcCallback) CoprocessorRpcUtils(org.apache.hadoop.hbase.ipc.CoprocessorRpcUtils) AggregateService(org.apache.hadoop.hbase.protobuf.generated.AggregateProtos.AggregateService)

Example 8 with AggregateResponse

use of org.apache.hadoop.hbase.protobuf.generated.AggregateProtos.AggregateResponse in project hbase by apache.

the class AggregationClient method getStdArgs.

/**
   * It computes a global standard deviation for a given column and its value.
   * Standard deviation is square root of (average of squares -
   * average*average). From individual regions, it obtains sum, square sum and
   * number of rows. With these, the above values are computed to get the global
   * std.
   * @param table
   * @param scan
   * @return standard deviations
   * @throws Throwable
   */
private <R, S, P extends Message, Q extends Message, T extends Message> Pair<List<S>, Long> getStdArgs(final Table table, final ColumnInterpreter<R, S, P, Q, T> ci, final Scan scan) throws Throwable {
    final AggregateRequest requestArg = validateArgAndGetPB(scan, ci, false);
    class StdCallback implements Batch.Callback<Pair<List<S>, Long>> {

        long rowCountVal = 0l;

        S sumVal = null, sumSqVal = null;

        public synchronized Pair<List<S>, Long> getStdParams() {
            List<S> l = new ArrayList<>(2);
            l.add(sumVal);
            l.add(sumSqVal);
            Pair<List<S>, Long> p = new Pair<>(l, rowCountVal);
            return p;
        }

        @Override
        public synchronized void update(byte[] region, byte[] row, Pair<List<S>, Long> result) {
            if (result.getFirst().size() > 0) {
                sumVal = ci.add(sumVal, result.getFirst().get(0));
                sumSqVal = ci.add(sumSqVal, result.getFirst().get(1));
                rowCountVal += result.getSecond();
            }
        }
    }
    StdCallback stdCallback = new StdCallback();
    table.coprocessorService(AggregateService.class, scan.getStartRow(), scan.getStopRow(), new Batch.Call<AggregateService, Pair<List<S>, Long>>() {

        @Override
        public Pair<List<S>, Long> call(AggregateService instance) throws IOException {
            RpcController controller = new AggregationClientRpcController();
            CoprocessorRpcUtils.BlockingRpcCallback<AggregateResponse> rpcCallback = new CoprocessorRpcUtils.BlockingRpcCallback<>();
            instance.getStd(controller, requestArg, rpcCallback);
            AggregateResponse response = rpcCallback.get();
            if (controller.failed()) {
                throw new IOException(controller.errorText());
            }
            Pair<List<S>, Long> pair = new Pair<>(new ArrayList<>(), 0L);
            if (response.getFirstPartCount() == 0) {
                return pair;
            }
            List<S> list = new ArrayList<>();
            for (int i = 0; i < response.getFirstPartCount(); i++) {
                ByteString b = response.getFirstPart(i);
                T t = getParsedGenericInstance(ci.getClass(), 4, b);
                S s = ci.getPromotedValueFromProto(t);
                list.add(s);
            }
            pair.setFirst(list);
            ByteBuffer bb = ByteBuffer.allocate(8).put(getBytesFromResponse(response.getSecondPart()));
            bb.rewind();
            pair.setSecond(bb.getLong());
            return pair;
        }
    }, stdCallback);
    return stdCallback.getStdParams();
}
Also used : AggregateRequest(org.apache.hadoop.hbase.protobuf.generated.AggregateProtos.AggregateRequest) ByteString(com.google.protobuf.ByteString) ArrayList(java.util.ArrayList) AggregateResponse(org.apache.hadoop.hbase.protobuf.generated.AggregateProtos.AggregateResponse) ArrayList(java.util.ArrayList) List(java.util.List) Pair(org.apache.hadoop.hbase.util.Pair) IOException(java.io.IOException) ByteBuffer(java.nio.ByteBuffer) RpcController(com.google.protobuf.RpcController) RpcCallback(com.google.protobuf.RpcCallback) CoprocessorRpcUtils(org.apache.hadoop.hbase.ipc.CoprocessorRpcUtils) AtomicLong(java.util.concurrent.atomic.AtomicLong) AggregateService(org.apache.hadoop.hbase.protobuf.generated.AggregateProtos.AggregateService)

Example 9 with AggregateResponse

use of org.apache.hadoop.hbase.protobuf.generated.AggregateProtos.AggregateResponse in project hbase by apache.

the class AggregationClient method sum.

/**
   * It sums up the value returned from various regions. In case qualifier is
   * null, summation of all the column qualifiers in the given family is done.
   * @param table
   * @param ci
   * @param scan
   * @return sum &lt;S&gt;
   * @throws Throwable
   */
public <R, S, P extends Message, Q extends Message, T extends Message> S sum(final Table table, final ColumnInterpreter<R, S, P, Q, T> ci, final Scan scan) throws Throwable {
    final AggregateRequest requestArg = validateArgAndGetPB(scan, ci, false);
    class SumCallBack implements Batch.Callback<S> {

        S sumVal = null;

        public S getSumResult() {
            return sumVal;
        }

        @Override
        public synchronized void update(byte[] region, byte[] row, S result) {
            sumVal = ci.add(sumVal, result);
        }
    }
    SumCallBack sumCallBack = new SumCallBack();
    table.coprocessorService(AggregateService.class, scan.getStartRow(), scan.getStopRow(), new Batch.Call<AggregateService, S>() {

        @Override
        public S call(AggregateService instance) throws IOException {
            RpcController controller = new AggregationClientRpcController();
            // Not sure what is going on here why I have to do these casts. TODO.
            CoprocessorRpcUtils.BlockingRpcCallback<AggregateResponse> rpcCallback = new CoprocessorRpcUtils.BlockingRpcCallback<>();
            instance.getSum(controller, requestArg, rpcCallback);
            AggregateResponse response = rpcCallback.get();
            if (controller.failed()) {
                throw new IOException(controller.errorText());
            }
            if (response.getFirstPartCount() == 0) {
                return null;
            }
            ByteString b = response.getFirstPart(0);
            T t = getParsedGenericInstance(ci.getClass(), 4, b);
            S s = ci.getPromotedValueFromProto(t);
            return s;
        }
    }, sumCallBack);
    return sumCallBack.getSumResult();
}
Also used : AggregateRequest(org.apache.hadoop.hbase.protobuf.generated.AggregateProtos.AggregateRequest) ByteString(com.google.protobuf.ByteString) AggregateResponse(org.apache.hadoop.hbase.protobuf.generated.AggregateProtos.AggregateResponse) IOException(java.io.IOException) RpcController(com.google.protobuf.RpcController) RpcCallback(com.google.protobuf.RpcCallback) CoprocessorRpcUtils(org.apache.hadoop.hbase.ipc.CoprocessorRpcUtils) AggregateService(org.apache.hadoop.hbase.protobuf.generated.AggregateProtos.AggregateService)

Example 10 with AggregateResponse

use of org.apache.hadoop.hbase.protobuf.generated.AggregateProtos.AggregateResponse in project hbase by apache.

the class AggregateImplementation method getMin.

/**
   * Gives the minimum for a given combination of column qualifier and column
   * family, in the given row range as defined in the Scan object. In its
   * current implementation, it takes one column family and one column qualifier
   * (if provided). In case of null column qualifier, minimum value for the
   * entire column family will be returned.
   */
@Override
public void getMin(RpcController controller, AggregateRequest request, RpcCallback<AggregateResponse> done) {
    AggregateResponse response = null;
    InternalScanner scanner = null;
    T min = null;
    try {
        ColumnInterpreter<T, S, P, Q, R> ci = constructColumnInterpreterFromRequest(request);
        T temp;
        Scan scan = ProtobufUtil.toScan(request.getScan());
        scanner = env.getRegion().getScanner(scan);
        List<Cell> results = new ArrayList<>();
        byte[] colFamily = scan.getFamilies()[0];
        NavigableSet<byte[]> qualifiers = scan.getFamilyMap().get(colFamily);
        byte[] qualifier = null;
        if (qualifiers != null && !qualifiers.isEmpty()) {
            qualifier = qualifiers.pollFirst();
        }
        boolean hasMoreRows = false;
        do {
            hasMoreRows = scanner.next(results);
            int listSize = results.size();
            for (int i = 0; i < listSize; i++) {
                temp = ci.getValue(colFamily, qualifier, results.get(i));
                min = (min == null || (temp != null && ci.compare(temp, min) < 0)) ? temp : min;
            }
            results.clear();
        } while (hasMoreRows);
        if (min != null) {
            response = AggregateResponse.newBuilder().addFirstPart(ci.getProtoForCellType(min).toByteString()).build();
        }
    } catch (IOException e) {
        CoprocessorRpcUtils.setControllerException(controller, e);
    } finally {
        if (scanner != null) {
            try {
                scanner.close();
            } catch (IOException ignored) {
            }
        }
    }
    log.info("Minimum from this region is " + env.getRegion().getRegionInfo().getRegionNameAsString() + ": " + min);
    done.run(response);
}
Also used : InternalScanner(org.apache.hadoop.hbase.regionserver.InternalScanner) AggregateResponse(org.apache.hadoop.hbase.protobuf.generated.AggregateProtos.AggregateResponse) ArrayList(java.util.ArrayList) IOException(java.io.IOException) Scan(org.apache.hadoop.hbase.client.Scan) Cell(org.apache.hadoop.hbase.Cell)

Aggregations

IOException (java.io.IOException)21 AggregateResponse (org.apache.hadoop.hbase.protobuf.generated.AggregateProtos.AggregateResponse)21 AggregateRequest (org.apache.hadoop.hbase.protobuf.generated.AggregateProtos.AggregateRequest)14 ByteString (com.google.protobuf.ByteString)9 ArrayList (java.util.ArrayList)9 RpcCallback (com.google.protobuf.RpcCallback)7 RpcController (com.google.protobuf.RpcController)7 CompletableFuture (java.util.concurrent.CompletableFuture)7 Cell (org.apache.hadoop.hbase.Cell)7 HRegionInfo (org.apache.hadoop.hbase.HRegionInfo)7 Scan (org.apache.hadoop.hbase.client.Scan)7 CoprocessorRpcUtils (org.apache.hadoop.hbase.ipc.CoprocessorRpcUtils)7 AggregateService (org.apache.hadoop.hbase.protobuf.generated.AggregateProtos.AggregateService)7 InternalScanner (org.apache.hadoop.hbase.regionserver.InternalScanner)7 ByteBuffer (java.nio.ByteBuffer)6 AtomicLong (java.util.concurrent.atomic.AtomicLong)3 Pair (org.apache.hadoop.hbase.util.Pair)3 List (java.util.List)2 NavigableMap (java.util.NavigableMap)2 TreeMap (java.util.TreeMap)1