Tales from the Trenches

How in-memory helped SAP innovate their business. Now this enables us to help our customers achieve things that otherwise would have been impossible. Through different use cases I want to show you what In-Memory brought to our customers and what the value is they derived from this.

For six years, Swedish ISV Heads AB has been running their in-memory retail software suite at Gekås, a €500,000,000 “superstore” with 30,000 peak daily customers. For reference, this is three times the size of Walmart’s largest superstore. With only one 4-core server, Gekås manages the entire point of sale, accounting, ERP, inventory, warehouse and supply chain functions in real time. Number crunching reports that previously required hours or days and multiple staff to finish are now completed in seconds, literally at the push of a button. 

JacTravel is a global B2B travel provider, providing live availability for over 15,000 city hotels worldwide.   JacTravel Group, with headquarters in London and offices worldwide, have an annual turnover in excess of £350m and process upwards of 260 million API searches of hotel inventory each day.

The fallout from the financial crisis of 2008 continues to be felt in the financial markets domain. Looking back almost a decade later we can clearly observe some fundamental changes in the way in which risk is measured and managed. These changes have been driven both by new regulations and market conditions and may be summarised as a change in focus from the single payoff to a portfolio view taking into account both market and credit risk.

Java 8 introduced the Stream API - a modern, functional, and very powerful tool for performing bulk operations on collections of data. One of the main benefits of the Stream API is that it hides the details of iteration over the underlying data set, allowing for parallel processing within a single JVM, using a fork/join framework.

I will talk about the Stream API implementations for the development of parallel distributed across many machines and many JVMs programs on top Java 8 Stream API.

A real time fraud intelligence platform requires machine learning capabilities as well as customizability by industry, to be effective.  For example, in an online travel business, understanding where a flight ticket is purchased and where a user is going has an impact on determining fraud. But for a gaming company, the metrics for understanding fraud are somewhat different, as it's important to understand how a user buys an in-game credit and how soon thereafter that credit used.

The presentation will explain how GridGain IMDG is considered fit for purpose to increase the scalability and availability of the electronic banking offerings. The goal is to share experiences on how this technology fits into the technical landscape and which integration patterns are being tested and deployed.

Target audience engineers and architects thinking on using the same technology in similar patterns. Also technology providers to have feedback from the trenches.

My name is Bart Hansen, and as a Delivery Focused Architect, I am currently working at TUI Western-Europe.

The business context I support is all about planning and created holiday products, and selling these through Retail and Web channels. The footprint covered includes Belgium, France and The Netherlands reaching a total of almost 5 billion Euros.