Real-time algorithmic trading systems are widely used by pension funds, mutual funds, some hedge funds, market makers and other institutional traders, to manage market impact and risk, to provide liquidity to the market. The technologies of real-time information processing and high-performance computing, such as the parallel bridging model - SGL, are essential for such systems. However, many errors can be made with todays tools, for example, the distraction of developers because they must focus both on financial algorithms, parallel computing and coding, or compiler mis-optimization, etc. In this paper, we describe practical results with the software design of a real-time algorithmic trading prototype by undergraduate students within the CoSc 319 software engineering project course at the University of British Columbia's Okanagan campus (Canada) in collaboration with a PhD student from the University Paris-Est (France). The prototype can be modifi ed by end-users on the UML model level and then used with automatic Java code generation and execution within the Eclipse IDE. During the case study an advanced coding environment was developed for providing a visual and declarative approach to trading algorithms development so as to generate directly portable bitcode on Low-Level Virtual Machine (LLVM) from nancial speci cation of trading strategies. During the project, Canadian students collaborated with a research engineer from a hedge fund in Paris.
