gexecutor

KO | EN

gitlite — search

#python #hacktoberfest #ai #golang #go #android #rust #deep-learning #linux #macos #windows #awesome #kubernetes #openai

(★ 13)

synchronous and asynchronous event based c++ executor libray

gexecutor 최신버젼 다운로드

최종 버전 다운로드 (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

#C++ GExecutor 1.1#

#Introduction#
GExecutor provides a unified way to handle synchronous and asynchronous tasks by abstracting them from underlying thread and process architecture. It is a C++ library which offers similar capabilities like Executor framework in Java and Twisted in Python.

One of the well studied patterns for implementing a High performance I/O bound applications is to use asynchronous processing of client requests. Multiple client requests can be served without the cost of context switching thus achieving better performance at lower CPU and memory requirements. Such a design pattern is also referred as reactor pattern. Libraries like libevent or boost::asio make it easy to implement reactor patterns built on lower level APIs like kqueue, epoll, select etc.

Reactor pattern has two limitations  
1. **Synchronous Apis processing**: Any synchronous API calls would block processing of other events in the event loop. This would result in poor performance as application cannot process other requests while it is waiting for synchronous API to return. A very common solution is to use threads pools to offload such tasks. However, there is no simple way to handle such tasks in C++.  
2. **Multi CPU limitation**: Typical event loops can use only a single CPU effectively. This can be a significant practical limitation as one can easily find more than 8 logical CPUs in any off-the-shelf server. It is non trival to use multiple CPUs with reactor pattern in C++/C.

Multi-threaded or multi-process performance can outperform a single threaded systems by using concurrent data structures or shared memory in spite of added context switches and locking overheads in a reactor pattern. However, such systems may not perform well due to bad locking design (coarse), deadlocks, and complex data structures based on shared memory. More details on this discussion can be found at [C10K Problem](http://www.kegel.com/c10k.html).

GExecutor offers a hybrid solution to take advantage of both paradigms and make it simple to use appropriate pattern for different aspects of application without loss of performance. It is a hybrid event loop based task processing framework which handles and routes tasks between async event loops and worker threads for processing synchronous tasks. It is designed as a utility that is layered above libevent or asio and can be plugged into any application. It does not create
a wrapper over the standard workflows or interfaces provided by existing libraries.

GExecutor is inspired by Java Executor based on [SEDA](http://www.eecs.harvard.edu/~mdw/proj/seda/) and [Twisted](http://twistedmatrix.com/trac/wiki) for python. It currently has option to use libevent or boost::asio based event loop for implementation for Linux.

#Architecture
GExector design is based on two important elements. Every executor has a task queue that it uses to receive Tasks. Each Task implements a virtual method *Execute()* that is invoked by the executor. Tasks contain pointer to the response Queue that it can use to send back results to the originating Executor.

A DeferredTask template is a simple interface to use callback functions instead of creating a lower level task interface.

A DeferredTask (deferred_task.h) supports three kinds of callbacks  
1. task_fn: a function pointer implementing the task which is deferred onto a different execution context.  
2. callback: Optional Task Callback is called with the arguments as the result returned by the above task_fn.  
3. errback: Task Errback called when either during task execution or during task callback execution.

#Example#
Here is an hello world example of a simple server based on GExecutor. Please see samples/simple_http_server for full reference:

/**
    * Hello world task that is executed in a different thread.
    */
    void print_hello() {
       std::cout << "Hello World\n";
    }
    /**
    * Callback called in the originator thread to indicate Hello World task was done and indeed it was successful in saying Hello!
    */
    void print_hello_done() {
       std::cout << "Said Hello to the world\n";
    }
    /**
    * Callback called if there was an exception in executing the Hello World task.
    */
    void print_hello_failed() {
        std::cout << "Could not say Hello\n";
    }  
    
    // Main thread that instantiates the GExecutor Service and sets up async and sync executors.
    
    int main(int argc, const char* argv[]) {
        // Creates a GExecutor reactor with default async executor loop
        GExecutorService executor_svc(true);
        // Creates a pool of synchronous workers that accepts tasks on taskq sync_executor->taskq();
        GExecutorSharedPtr sync_executor = executor_svc.CreateSyncExecutor("sync", num_sync_workers);
        // run the default reactor.
        executor_svc.run();
        // alternatively if event_base was created outside one can do executor_svc(false) and call 
        // event_dispatch outside.
    }  
    
    // Example to create an async task for the default executor
    
    void create_async_task() {
        //Example to add print "Hello" Tasks
        GExecutorSharedPtr async_executor = executor_svc.gexecutor(
            executor_svc.kDefaultExecutorId);
        GTaskQSharedPtr taskq = async_executor->taskq();
        boost::shared_ptr<DeferredTask<void>> d(
            new DeferredTask<void>(taskq, print_hello);
        // attach callback when task print_hello was successful
        d.set_callback(print_hello_done);
        // attach callback when task print_hello failed.
        d.set_errback(print_hello_failed);
        async_executor->EnQueueTask(d);
    }

// Example to create an sync task for the default executor

void create_sync_task() {
        //Example to add print "Hello" Tasks
        GExecutorSharedPtr sync_executor = executor_svc.gexecutor(
            "sync");
        GExecutorSharedPtr async_executor = executor_svc.gexecutor(
            executor_svc.kDefaultExecutorId);
        // Get Taskq to receive the reponse from the sync executor
        // and set it in the deferred task.
        GTaskQSharedPtr resp_taskq = async_executor->taskq();
        boost::shared_ptr<DeferredTask<void>> d(
            new DeferredTask<void>(resp_taskq, print_hello);
        // attach callback when task print_hello was successful
        d.set_callback(print_hello_done);
        // attach callback when task print_hello failed.
        d.set_errback(print_hello_failed);
        sync_executor->EnQueueTask(d);
    }

Here are some of the example reactor paradigms that can be easily implemented using GExecutor
##Single Async loop with synchronous worker pools##
![alt text](https://github.com/cppexecutor/gexecutor/blob/dev/1async1sync.jpg "Single Async loop with synchronous worker pools")

There is one default async executor and one pool of synch workers. This is perhaps most common option for the I/O bound applications. CPU bound application would either need to break away tasks for workers or create multiple async execution blocks.

##Multiple Async loop using single synchronous worker pool##
This is useful in the cases like HTTP server needs more than one CPU to just handle the TCP connection accepts. These accepts can be handled in a distributed executors. However, the request can then be sent to a common Executor/thread that is used for implementing the application logic using reactor design. This mechanism avoids need to use shared memory for sharing information like configuration data between two threads/processes.
![alt text](https://github.com/cppexecutor/gexecutor/blob/dev/multi-async-1-sync.jpg "Multiple Async loop using single synchronous worker pool")

##Multiple Async loop with multiple synchronous worker pools##
This is useful in implementing multi-stage SEDA style processing blocks where each stage is also based on reactor processing. Each stage can have its own synchronous processing pools with their own worker sizes.
![alt text](https://github.com/cppexecutor/gexecutor/blob/dev/multi-async-multi-sync.jpg "Multiple Async loop with multiple synchronous worker pools")

#Design considerations#
One important aspect of the GExecutor interface is that it accepts taskq as a parameter for the async executors. This is because the taskq needs to be created before the Async threads are created. Otherwise the underlying FD and pipe interface used for notifications would not be visible outside the async thread created.

#Working with the Code#
##Requirements##
GExecutor is only supported on Linux. There is not much Linux specific implementation other than the fact that I have not had time to make it full platform dependent.

##Dependencies##
It is uses boost_system, libevent, GTest (unit tests), cmake (build). Script *setup.sh* can be used for simple installation.

## Boost vs libevent
GExecutor can work with either libevent or boost::asio. It uses the reactor pattern based architecture.

###Libevent based libraries and header files
**library**: libgexecutor-common.so and libgexecutor.so
**header files**: gexecutor.h gexecutor_service.h gtaskq.h gexecutor_service_base.h
				gexecutor_common.h deferred_task.h

###boost::asio based libraries and header files
**Library**: libgexecutor-common.so and libgexecutor-asio.so
**header files**: gexecutor.h gtaskq.h gexecutor_service_base.h
				gexecutor_common.h deferred_task.h gexecutor_service_asio.h gsync_executor_asio.h 
        		gasync_executor_asio.h

##Installation##
Linux installation requires following steps:

git checkout https://github.com/cppexecutor/gexecutor.git
    ./setup.sh
    mkdir build
    cd build
    cmake ../src
    make
    sudo make install
    make test
  
#Licensing#
All the software provided under gexecutor under the MIT License

#Contact#
Gaurav Rastogi
https://www.linkedin.com/in/grrastogi
contact: gr.rastogi@gmail.com

gexecutor

gexecutor 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

gexecutor

gexecutor 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Download README