Tuesday, 5 July 2016

Docker images creation in a development environment


Sorry, It´s been a while since my last post, a mixture of not having anything worth writing, holidays and some online training that drains most of my time.
Here is something I would like to share that might reduce the pain for those who want to dockerize their Spring-Boot applications (or any Java application) and then test and tune them while running inside a Docker container.

Former setup

  • A commit was pushed to Jenkins, who compiled and unit-tested it.
  • After some code analysis, a Docker image was created and pushed to a repository
    With set-up, testing was highly inefficient.
  • To test it in a server, I had to pull it and execute it and, while I was focused in the performance and memory consumption of my application, tuning the JVM parameters was essential... Unfortunately, they were given at compilation time (a Maven plugin) and could not be changed (or I did not know how to).
The drawbacks I see of this approach:
  • It was slow to test new changes, as the whole CI flow needed to be executed.
  • Unnecessary docker-related commits to the branch generating the Docker images.
The solution I found? Building the image myself before testing it.

Saturday, 2 April 2016

Testing & error handling with Spring + Reactor


As we saw in the previous post, the whole idea of reactive programming floats around two concepts: Data comes in asynchronous streams and those streams are immutable, so any modification results in new streams being created.
These facts makes the code not very intuitive and it is even worse when it comes to testing (even more if we are interacting with external resources, like a Web Service).
To help in the task, Spring and Reactor come with some handy tools that can be applied to simplify the development and testing.

First, let´s take a look at the scenario being tested:

Sequence diagram of the "production" code

Saturday, 12 March 2016

Intro to Reactive Programming with Reactor and Spring


What did I know last week about reactive programming? Nothing. That´s a fact. However, I started reading about and I find the topic really really interesting.

Unfortunately, there is not much documentation around (or at least, I had a hard time finding it) so I would like to share some resources with you along with an example that I have adapted to perform like our module Stokker (the module that used to fetch stock quotations from Yahoo Finance for us using Spring Integration).

What is Reactive Programming?

The best definition I could find was this amazing GitHub gist written by @andrestaltz that summarizes all the theoretical definitions to this line:

Reactive programming is programming with asynchronous data streams.

I really recommend to you its reading, but we can summarize it in:

  • You will work with immutable event streams to which you will have to subscribe.
  • Some operations will be provided to you, so you can transform these streams into new ones by filtering them, merging them o create new events from the previous ones.

The best way to understand this is, is to represent these operations graphically. For that purpose, we have this really cool web application implemented using JavaScript and RxObservables: RxMarbles (a must see!). For instance:

Two event streams: One with numbers, the other with each element multiplied by 10

The first line, represents an initial event stream. In the middle you can see the mapping function that is applied to every event in the stream and finally, the second stream with the results:

Note: Pay attention to the immutability here: Any transformation you apply to a stream creates a new one!

Let´s check out another transformation:

Two event streams: One with numbers, other with the average of the elements received

Look who is here! Our old friend, the moving average. Whenever a new event arrives in the original stream, a new value for the average will be calculated in the resulting stream.

Monday, 22 February 2016

Building a basic stock trading system with Drools (I)


In my previous post, I explained how can we calculate a number of indicators based on Stock prices and that, the value on those indicators is better exploited in algorithms rather that just plotted in charts. What kind of algorithm might that be?

I mentioned that one possibility is that each indicator has a "voice" or "opinion" and that the output of the algorithm is the sum of all those opinions, some sort of consensus. Then we can translate that consensus in actual actions: Buy and Sell orders.

How can we model those opinions? Basically, they are rules responding to certain mathematical expressions. One option to evaluate is a Java-based rule engine: Drools.

But first, take a look at the result of our calculations, so you can understand better. This graph shows a "score" given by our system on how bullish or bearish (that is, whether it has an upward/downward chance) a stock is. This score is calculated using simple indicators (Simple Moving Average, New MAX, New Min, etc.)
The price and indicators above, the SCORE indicator below

In this case, the rules said:

  • + 1 when the price is over the SMA(200).
  • + 1 when the price is making new MAX taking into account the last 200 days.
  • -1 when the price is under the SMA(200).
  • - 1 when the price is making new MIN taking into account the last 200 days.

So we can extrapolate that:

  • If the score was negative, and it has become positive; it´s likely for the price to go up.
  • If the score was +1 and becomes +2, it gets even better.
On the other hand:

  • If the score becomes negative or makes new lows, it´s time to close our position and avoid more losses (see right part of the graph).

Note: Please keep in mind that the main purpose here is learning the technology and have fun, don´t expect Warren-Buffet-like returns!

Sunday, 14 February 2016

Calculating and showing stock indicators with Spark & AngularJS


In a previous post, I showed how simple is to calculate a SMA (Simple Moving Average) indicator from your list of stock quotations. Today, let´s see how can we modify the code to calculate multiple indicators at once (for the moment, I could not implement it using several different sliding windows at once).

Also, let´s introduce a quick prototype coded in AngularJS in order to show the different indicators in a chart with the possibility of switching stocks and windows sizes live. This is just a debug feature as our aim is to use the indicators in our future home-made analysis algorithm.

Note: If you don´t know what SMAs are, you should read this before.

Live example?

This is much easier to understand if you see an actual example running, but unfortunately I'm having problems injecting an iframe here that shows the content properly. Instead, you can clone the repository and run the application with the test files included in it:
git clone https://github.com/victor-ferrer/sparkker
Note: The application requires a Java 8 JRE installed in your machine. Also, the JAR file is rather heavy as Spring Boot adds an embedded Tomcat Server and all the Spark dependencies...

Here is a screenshot of the application that will be accessible at http://localhost:8988/#/indicators

Chart showing the upward trend followed by one stock.
The different lines are the indicators calculated on a 200 days window.

Thursday, 4 February 2016

Video: Lambda architecture with Spring XD and Spark


Today I do not have any piece of code worth showing you. Instead, I would like to quickly comment on this YouTube video, which I have seen recently, that is coming from one the SpringOne2GX 2015 event (and later released to the public).

Note: The fact the these talks are published more and more often (like what happens with the DockerCon talks) kinda puts me off from paying the high premium involved in physically assisting to these events...

Sunday, 31 January 2016

Calculating Moving Averages with Spark MLLib


Since I began learning Spark, one of the things that given me the most pain is the Java API that wraps the Scala API used in Spark. So, I would like to share this problem I had using the Spark Machine Learning library in case is useful for anybody learning Spark. Also, it´s nice to take a look to the possibilities that the sliding windows give us when working on timed data.

The problem

As I wrote in my previous post, I created a new independent module to make studies on Stock data using Spark. One of the most interesting (and basic) studies one can make over Stock data, is to calculate the simple moving averages.

What is a simple moving average (SMA)?

If we take a window of time of length N, the SMA is:
Source Wikipedia: https://en.wikipedia.org/wiki/Moving_average
Wait, what? In plain English: For every element, we take the N precedent elements and calculate the average. The aim is to smooth the data in order to easily detect underlying patterns in the data.

One possible solution: Spark MLLib

Searching around in internet, I found that the Spark Machine Learning Libray has a sliding window operation over an RDD. Using this, it would be very easy to calculate the SMA as described in this StackOverflow answer.

Note: The RDD must be sorted before appliying the sliding function!

The caveat is that the previous code is written in Scala. How would it look like in Java?