Abstract- SequenceL is a concise, high-level language with a simple semantic that provides for the automatic derivation of many iterative and parallel control structures. The semantic repeatedly applies a “Normalize-Transpose ” operation to functions and operators until base cases are discovered, which include the grounding of variables and the application of built-in operators to operands of appropriate types. This paper introduces the language from an intuitive point of view, indicating the scalability of the normalize-transpose, the different forms of recursion supported by the language and its abstraction, and new shorthand methods for specifying additional iterative/recursive problem solutions.

Abstract. We present a high level language called SequenceL. The language is relationships between their inputs and outputs, rather than in terms of algorithms for carrying out computations. The semantics of the language are capable of automatically discovering and implementing the required algorithms, including iterative and parallel control structures in many cases. Current implementations are prototypes only, and do not produce code of comparable efficiency to that of a good human programmer. Current implementations can, however, be used as a tool to guide human programmers in discovering and comparing options for parallelizing their solutions. This paper describes the language and approach, and demonstrates the use of SequenceL to discover options for parallelizing matrix computations. 1 Introduction and Related Work SequenceL is a compact, Turing-complete, high-level language, in which algorithms for implementing a solution are automatically derived from a high level description of