GeoPulse is a specialized server application designed to handle real-time geolocation data from clients via the UDP protocol. The server captures location coordinates and processes them for various uses, such as real-time tracking and historical data retrieval. GeoPulse is optimized utilizing GeoJSON and Swoole to handle numerous simultaneous UDP connections. It integrates seamlessly with Laravel queue system, allowing you to capture tracking events using Laravel jobs.
- Client Sends HTTP Request: The client sends periodic HTTP POST requests to update location.
- HTTP Request is Sent to Server: The request is transmitted over HTTP.
- Server Receives HTTP Request: The server processes the incoming HTTP request.
- Server Processes Request (Overhead): The server processes the request with potential delays.
- Server Sends HTTP Response Back to Client: The server sends a response back to the client.
- Client Receives HTTP Response (OK).
- Client Sends UDP Packet (Continuous): The client continuously sends UDP packets.
- GeoPulse Server Processes UDP Packet Immediately: Data is processed immediately.
- No Response Needed (Data Delivered Fast): No acknowledgment (ACK) is required from the server.
GeoPulse uses a JSON format to transmit data packets over UDP for real-time location tracking. The structure is designed to include all necessary information such as the application ID, client ID, and location data, making it easy to parse and process on the server side.
{
"appId": "yourAppId123",
"clientId": "client456",
"data": {
"type": "Point",
"coordinates": [102.0, 0.5]
}
}For bandwidth efficiency, you might consider compressing the JSON payload using MessagePack. GeoPulse already supports MessagePack as an alternative to JSON for smaller payload sizes.
- Kernel: x86_64 Linux 6.8.0-41-generic / Intel Core i5-4590T @ 4x 3GHz / 16GB of ram / Local network
- Tools: Apache Jmeter with udp plugin / number of threads 5000 in 1 second
- pulse-config: open_cpu_affinity = true , 'cpu_affinity_ignore' => [0,1], 'worker_num' => 6
- server responds with "OK" , in real life scenario the server will not responde to the client.
docker pull laggounewalid/geopulse:1.0docker run -d -p 9505:9505/udp -v ./pulse-config:/var/www/html/config laggounewalid/geopulse:1.0The pulse-config/ folder must contain a pulse.php config file.
- Queue server supported by
illuminate/queue - Database supported by
illuminate/database(Oracle Database not supported by GeoPulse)
CREATE TABLE `pulse_coordinates` (
`id` int(10) unsigned NOT NULL AUTO_INCREMENT,
`created_at` timestamp NOT NULL DEFAULT current_timestamp(),
`appId` varchar(255) DEFAULT NULL,
`clientId` varchar(255) DEFAULT NULL,
`coordinate` point DEFAULT NULL,
`updated_at` timestamp DEFAULT NULL,
PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4;Here is a boilerplate PHP code for configuration you can use in pulse-config/pulse.php:
<?php
return [
/*
* Configuration: appId
*
* The unique application identifier used to authenticate clients.
* This App ID allows the server to verify that a client is authorized to send UDP packets to GeoPulse.
* Every UDP packet sent by a client must include this App ID for validation purposes.
*/
'appId' => 'your_app_id_here',
/*
* Configuration: port
*
* The application port number for the UDP server.
* This port is used to receive incoming packets containing coordinate data.
* Ensure that the specified port (default: 9505) is open and not blocked by a firewall.
*/
'port' => 9505,
/*
* Configuration: use-msgPack
*
* A boolean flag to enable or disable MessagePack for data serialization and deserialization.
* When set to true, the server will use MessagePack to unpack received data packets.
* If set to false, the server will process data as raw strings.
*/
'use-msgPack' => true,
'swoole' => [
'debug_mode' => true,
'display_errors' => true,
'worker_num' => 4,
'enable_coroutine' => true,
'open_eof_check' => true,
'package_eof' => "\r\n",
'dispatch_mode' => 1,
],
/*
* Configuration: enable-queue
*
* Determines whether the server should use queues to process packets.
*/
'enable-queue' => true,
/*
* Configuration: queue-pool-size
*
* Specifies the number of queue worker connections created and managed in the Swoole connection pool.
* A larger pool size can help manage a higher volume of queue jobs concurrently, but it also increases memory and resource usage.
* Default value: 10
*/
'queue-pool-size' => 10,
/*
* Configuration: queue-connection
*
* Defines the connection settings for the queue driver. Follows the same configuration structure used by Laravel's queue system.
*/
'queue-connection' => [
'driver' => 'redis',
'connection' => 'default',
'queue' => 'geopulse',
'retry_after' => 90,
'block_for' => null,
'after_commit' => false,
],
'redis' => [
'options' => [
'cluster' => 'redis',
'prefix' => 'YOUR_APP_NAME_database_',
],
'default' => [
'url' => '',
'host' => '',
'username' => '',
'password' => '',
'port' => 6379,
'database' => '0',
],
],
/*
* Configuration: enable-database
*
* Determines whether the server should use a database for storing location records.
*/
'enable-database' => true,
/*
* Configuration: db-pool-size
*
* Defines the number of database connections created and added to the Swoole connection pool.
* Increasing the pool size can improve performance when dealing with many concurrent database operations but will also increase resource consumption.
* Default value: 10
*/
'db-pool-size' => 10,
/*
* Configuration: database-connection
*
* Defines the database connection settings, following the same structure as Laravel's database configuration file.
*/
'table-name' => 'pulse_coordinates',
'database-connection' => [
'driver' => 'mariadb',
'url' => null,
'host' => 'YOUR_DATABASE_HOST',
'port' => '3306',
'database' => 'DB_NAME',
'username' => 'DB_USER',
'password' => 'DB_PASSWORD',
'unix_socket' => null,
'charset' => 'utf8mb4',
'collation' => 'utf8mb4_unicode_ci',
'prefix' => '',
'prefix_indexes' => true,
'strict' => false,
'engine' => null,
],
];<?php
use Swoole\Coroutine\Client;
use function Swoole\Coroutine\run;
run(function () {
$client = new Client(SWOOLE_SOCK_UDP);
if (!$client->connect('0.0.0.0', 9505, 0.5)) {
echo "Connect failed. Error: {$client->errCode}\n";
}
$data = ['appId' => 'your_app_id_here', 'clientId' => '22f8e456-93f2-4173-8f2d-8a010abcceb1', 'data' => ['type' => 'Point', 'coordinates' => [1, 1]]];
$data = msgpack_pack($data);
$client->send($data);
$client->close();
});<?php
namespace App\Jobs;
use Illuminate\Bus\Queueable;
use Illuminate\Contracts\Queue\ShouldQueue;
use Illuminate\Foundation\Bus\Dispatchable;
use Illuminate\Queue\InteractsWithQueue;
use Illuminate\Queue\SerializesModels;
class PulseLocationUpdatedJob implements ShouldQueue
{
use Dispatchable, InteractsWithQueue, Queueable, SerializesModels;
/**
* Create a new job instance.
*/
public function __construct()
{
//
}
/**
* Execute the job.
*/
public function handle($job, array $data): void
{
// Data example:
// [
// "point" => [
// "type" => "Point",
// "coordinates" => [1, 1]
// ],
// "appId" => "your_app_id_here",
// "clientId" => "22f8e456-93f2-4173-8f2d-8a010abcceb1"
// ]
$job->delete
();
}
}To implement your own broadcaster (e.g., Kafka or MongoDB), add the broadcaster in bin/bootstrap.php. Your broadcaster needs to be located in src/app/Actions/ and must implement PacketActionContract.
<?php
namespace Pulse\Actions;
use Illuminate\Queue\Capsule\Manager as Queue;
use Pulse\Contracts\Action\PacketActionContract;
use Pulse\Contracts\PacketParser\Packet;
class PublishToKafkaTopic implements PacketActionContract
{
public function handle(Packet $packet): void
{
// Implement your logic here
}
}$container->add(BroadcastPacketService::class, function () use ($config) {
$broadcaster = new BroadcastPacketService;
$broadcaster->addAction(new PublishToKafkaTopic);
return $broadcaster;
});