gRPC & High Performance APIs · 강의

클라이언트 스트리밍 이해

클라이언트가 일련의 메시지를 서버로 보낼 수 있도록 클라이언트 측 스트리밍을 구현하는 방법을 배웁니다.

클라이언트 스트리밍 이해은(는) CoddyKit의 무료 gRPC & High Performance APIs 강의입니다. 이것은 4개 중 2번째 강의입니다. 아래에서 전체 강의를 무료로 읽을 수 있으며, 내장 코드 에디터와 24/7 AI 튜터와 함께 브라우저에서 직접 실습할 수 있습니다. 이 강의는 gRPC & High Performance APIs 학습 경로의 일부이며, 진행 상황이 웹과 CoddyKit 앱에 동기화됩니다. gRPC & High Performance APIs 강의에는 총 4개의 강의가 포함되어 있습니다.

이 강의의 일부는 아직 번역되지 않았으며 영어로 표시됩니다.

What is Client Streaming?

Welcome to client streaming! In gRPC, client streaming is a communication pattern where the client sends a sequence of messages to the server.

Unlike a simple unary RPC (request-response), the client doesn't just send one message. Instead, it sends a stream of messages, and the server processes them, then sends back a single response at the end.

How Client Streaming Works

Imagine uploading a large file by sending it in many small chunks. The server collects all chunks, rebuilds the file, and then sends a single "upload complete" confirmation.

The client initiates the RPC.
The client sends multiple messages asynchronously.
The server receives and processes these messages.
Once the client finishes sending (signals completion), the server sends back a single response.

Defining Client Stream in Protobuf

To define a client-streaming method in your .proto file, you use the stream keyword for the request type, but not for the response type.

Here's an example for a log upload service:

syntax = "proto3";

package client_streaming;

service LogService {
  rpc UploadLogs (stream LogEntry) returns (UploadSummary);
}

message LogEntry {
  string message = 1;
  int64 timestamp = 2;
}

message UploadSummary {
  int32 uploaded_count = 1;
  string status_message = 2;
}

Server: Handling the Client Stream

On the server side, your method will receive a StreamObserver for the client's incoming messages and will use another StreamObserver to send its single response.

The server's StreamObserver will have onNext() for each incoming message, onError() for errors, and onCompleted() when the client finishes sending.

import io.grpc.stub.StreamObserver;
import io.grpc.Server;
import io.grpc.ServerBuilder;
import client_streaming.LogEntry;
import client_streaming.LogServiceGrpc;
import client_streaming.UploadSummary;

public class LogServer {
    private Server server;

    private void start() throws Exception {
        int port = 50051;
        server = ServerBuilder.forPort(port)
            .addService(new LogServiceImpl())
            .build()
            .start();
        System.out.println("Server started, listening on " + port);
        Runtime.getRuntime().addShutdownHook(new Thread(() -> {
            System.err.println("*** shutting down gRPC server since JVM is shutting down");
            LogServer.this.stop();
            System.err.println("*** server shut down");
        }));
    }

    private void stop() {
        if (server != null) {
            server.shutdown();
        }
    }

    private void blockUntilShutdown() throws InterruptedException {
        if (server != null) {
            server.awaitTermination();
        }
    }

    public static void main(String[] args) throws Exception {
        final LogServer logServer = new LogServer();
        logServer.start();
        logServer.blockUntilShutdown();
    }

    static class LogServiceImpl extends LogServiceGrpc.LogServiceImplBase {
        @Override
        public StreamObserver<LogEntry> uploadLogs(StreamObserver<UploadSummary> responseObserver) {
            return new StreamObserver<LogEntry>() {
                private int logCount = 0;

                @Override
                public void onNext(LogEntry log) {
                    // Process each log entry as it arrives
                    System.out.println("Received log: " + log.getMessage() + " at " + log.getTimestamp());
                    logCount++;
                }

                @Override
                public void onError(Throwable t) {
                    System.err.println("UploadLogs cancelled or failed: " + t.getMessage());
                    responseObserver.onError(t);
                }

                @Override
                public void onCompleted() {
                    // After all logs are received, send a single summary response
                    UploadSummary summary = UploadSummary.newBuilder()
                        .setUploadedCount(logCount)
                        .setStatusMessage("Successfully processed " + logCount + " log entries.")
                        .build();
                    responseObserver.onNext(summary);
                    responseObserver.onCompleted();
                    System.out.println("Finished processing client stream. Sent summary.");
                }
            };
        }
    }
}

Client: Sending the Stream

On the client side, you'll get a StreamObserver to send your messages. You call onNext() for each message you want to send and finally onCompleted() to signal the end of the stream.

The server's single response will be handled by a separate StreamObserver you provide.

import io.grpc.ManagedChannel;
import io.grpc.ManagedChannelBuilder;
import io.grpc.stub.StreamObserver;
import client_streaming.LogEntry;
import client_streaming.LogServiceGrpc;
import client_streaming.UploadSummary;

import java.util.concurrent.TimeUnit;

public class LogClient {
    private final LogServiceGrpc.LogServiceStub asyncStub;
    private final ManagedChannel channel;

    public LogClient(String host, int port) {
        channel = ManagedChannelBuilder.forAddress(host, port)
            .usePlaintext() // For demonstration, use plaintext
            .build();
        asyncStub = LogServiceGrpc.newStub(channel);
    }

    public void shutdown() throws InterruptedException {
        channel.shutdown().awaitTermination(5, TimeUnit.SECONDS);
    }

    public void uploadMultipleLogs() throws InterruptedException {
        StreamObserver<UploadSummary> responseObserver = new StreamObserver<UploadSummary>() {
            @Override
            public void onNext(UploadSummary summary) {
                System.out.println("Server Response: " + summary.getStatusMessage() + " (" + summary.getUploadedCount() + " logs)");
            }

            @Override
            public void onError(Throwable t) {
                System.err.println("UploadLogs failed: " + t.getMessage());
            }

            @Override
            public void onCompleted() {
                System.out.println("Server has completed processing.");
            }
        };

        StreamObserver<LogEntry> requestObserver = asyncStub.uploadLogs(responseObserver);

        try {
            // Send multiple log entries
            LogEntry log1 = LogEntry.newBuilder().setMessage("User login attempt").setTimestamp(System.currentTimeMillis()).build();
            LogEntry log2 = LogEntry.newBuilder().setMessage("Database query executed").setTimestamp(System.currentTimeMillis() + 100).build();
            LogEntry log3 = LogEntry.newBuilder().setMessage("API call completed").setTimestamp(System.currentTimeMillis() + 200).build();

            requestObserver.onNext(log1);
            System.out.println("Client sent log 1");
            Thread.sleep(100); // Simulate some delay
            requestObserver.onNext(log2);
            System.out.println("Client sent log 2");
            Thread.sleep(100);
            requestObserver.onNext(log3);
            System.out.println("Client sent log 3");

            // Mark the end of the client stream
            requestObserver.onCompleted();
            System.out.println("Client finished sending logs.");

            // Wait for server response (handled by responseObserver)
            Thread.sleep(1000); // Give time for server to respond
        } catch (RuntimeException e) {
            requestObserver.onError(e);
            throw e;
        }
    }

    public static void main(String[] args) throws Exception {
        LogClient client = new LogClient("localhost", 50051);
        try {
            client.uploadMultipleLogs();
        } finally {
            client.shutdown();
        }
    }
}

Running the Example

To see client streaming in action:

First, compile your .proto file to generate the necessary Java classes.
Run the LogServer application. It will start listening for requests.
Then, run the LogClient application. It will send three log entries and wait for the server's summary.

Observe the console output from both the client and server to understand the flow of messages.

Key StreamObserver Methods

The StreamObserver interface is crucial for handling streaming RPCs. Both the client and server use implementations of this interface.

onNext(T value): Called for each message received in the stream. The client uses this to send messages, the server uses it to receive.
onError(Throwable t): Called if an RPC fails or is cancelled.
onCompleted(): Called when the stream has finished. The client calls this after sending all messages; the server calls it after sending its final response.

When to Use Client Streaming

Client streaming is ideal for scenarios where a client needs to send a large amount of data or a series of related messages to a server, and only cares about a single final result.

Large File Uploads: Sending a file chunk by chunk.
Log Aggregation: A client sending many log entries to a central logging service.
Batch Operations: Sending a list of items to be processed as a single batch, receiving a summary.
Sensor Data Collection: Continuously sending readings from a sensor device.

Quick Check

You're designing a gRPC service where a client needs to send a series of sensor readings to a server, and the server will process them and return a single summary report.

Recap: Client Streaming

You've learned about client-side streaming in gRPC!

Client streaming allows a client to send a sequence of messages.
The server processes these messages and sends a single response back.
In Protobuf, you define it by using the stream keyword for the request type.
Both client and server use StreamObserver to manage the flow of messages (onNext(), onError(), onCompleted()).
It's great for tasks like uploading large data or sending continuous log entries.

Next, we'll explore server-side streaming!

자주 묻는 질문

“클라이언트 스트리밍 이해” 강의는 무료인가요?

네 — “클라이언트 스트리밍 이해” 전체 내용을 이 웹사이트에서 무료로 읽을 수 있습니다. 인터랙티브하게 실습하려면(내장 코드 에디터와 24/7 AI 튜터), CoddyKit PRO로 업그레이드하면 gRPC & High Performance APIs 강의 전체를 잠금 해제할 수 있습니다. gRPC & High Performance APIs 강의에는 총 4개의 강의가 포함되어 있습니다.

“클라이언트 스트리밍 이해”에서 뭘 배우나요?

클라이언트가 일련의 메시지를 서버로 보낼 수 있도록 클라이언트 측 스트리밍을 구현하는 방법을 배웁니다. 브라우저에서 직접 실행하는 실습 코드로 gRPC & High Performance APIs을(를) 배우며, 24/7 AI 튜터가 강의를 진행하면서 질문에 답변해줍니다.

gRPC & High Performance APIs을(를) 시작하는 데 경험이 필요한가요?

사전 경험은 필요하지 않습니다. CoddyKit의 gRPC & High Performance APIs은(는) 초급자부터 고급 학습자까지를 위해 구성되어 있으므로, 여기서 시작하거나 처음부터 시작할 수 있으며 자신의 속도대로 진행할 수 있습니다. 이것은 4개 중 2번째 강의입니다.

“클라이언트 스트리밍 이해” 강의는 얼마나 걸리나요?

대부분의 CoddyKit 강의는 약 5~10분이 소요됩니다. 각 강의는 간결하고 인터랙티브하여 꾸준한 진행이 가능하며, 웹과 앱에서 중단한 부분부터 바로 시작할 수 있습니다.

이 gRPC & High Performance APIs 강의에서 코드를 작성하고 실행할 수 있나요?

네. 모든 gRPC & High Performance APIs 강의에는 내장 코드 에디터가 포함되어 있으므로, 브라우저에서 바로 실제 코드를 작성하고 실행한 후 즉시 AI 피드백을 받을 수 있습니다 — 로컬 설정이 필요 없습니다.

이 강의의 모든 강의

← gRPC & High Performance APIs(으)로 돌아가기