Market Trade - Realtime Charts

Message Consumption

The endpoint handle the POST requests asynchronously via the java.util.concurrent.Callable class.

The TradeMessages are stored into MongoDB which is schema-less and easy to scale out.

In regards to performance, the endpoint was load tested with [Apache jMeter] (http://jmeter.apache.org/) making 1000 concurrent requests on a local env.

Message Processor

The TradeMessageController uses the StatisticService, which pull data from MongoDB and creates the statistics.

StatisticService sends data through websockets from the server to the client in realtime via 2 channels.

Channel '/init' broadcasts the initial data to render the Charts right after the connection of the client (browser) to the socket.

Channel 'topic/charts' broadcasts new Statistics right after the creation of every TradeMessage.

Message Frontend

The front-end draws the Charts built with Google Visualization API to display data in realtime.

Tech Stack

• Java 1.8
• Spring Boot /Spring MVC
• Mongo (noSql db)
• Gradle (build)
• Websocket/Stomp (realtime messages)
• Thymeleaf (template engine)
• Google Visualization API (realtime charts)
• TwitterBootstrap (css)
• jUnit Mockito (testing)

Prerequisites

• jdk 1.8
• mongo 3 - visit mongodb.org
• gradle 2.3 - visit gradle.org

Setup

• Clone the repo git clone https://github.com/pangio/market-trade.git
• Build ./gradlew clean compile
• Run ./gradlew bootRun

Endpoints / Http Methods

• Returns the list of all Trade Messages.

GET: http://localhost:8080/trades

• Finds a Trade Message by id.

GET: http://localhost:8080/trades/{tradeMessageId}

• Creates a new Trade Message.

POST: http://localhost:8080/trades

JSON body example:

{
    "userId" : "134256",
    "currencyFrom" : "EUR",
    "currencyTo" : "GBP",
    "amountSell" : 1000,
    "amountBuy" : 4338,
    "rate" : 5.43,
    "timePlaced" : "14-JAN-15 10:27:44",
    "originatingCountry" : "FR"
}

Tests

• Under the test source folder find the suite com.cf.markettrade.AllTests.java

Statistics - Realtime Charts

Just browse http://localhost:8080/statistics

Screenshots

NOTES

• Content Type should be set to application/json .
• Mongo sholud be running on localhost with the default values as port=27017 and dbpath=/data/db