Abstract We describe a new automatic static analysis for determining upper-bound functions on the... more Abstract We describe a new automatic static analysis for determining upper-bound functions on the use of quantitative resources for strict, higher-order, polymorphic, recursive programs dealing with possibly-aliased data. Our analysis is a variant of Tarjan's manual amortised cost analysis technique. We use a type-based approach, exploiting linearity to allow inference, and place a new emphasis on the number of references to a data object. The bounds we infer depend on the sizes of the various inputs to a program.
Parallel functional programming has a relatively long history. Burge was one of the first to sugg... more Parallel functional programming has a relatively long history. Burge was one of the first to suggest the basic technique of evaluating function arguments in parallel, with the possibility of functions absorbing unevaluated arguments and perhaps also exploiting speculative evaluation 22]. Berkling also considered the application of functional languages to parallel processing 17].
This paper uses Template Haskell to automatically select appropriate skeleton implementations in ... more This paper uses Template Haskell to automatically select appropriate skeleton implementations in the Eden parallel dialect of Haskell. The approach allows implementation parameters to be statically tuned according to architectural cost models based on source analyses. This permits us to target a range of parallel architecture classes from a single source specification. A major advantage of the approach is that cost models are user-definable and can be readily extended to new data or computation structures etc.
This paper introduces the design of SymGrid, a new Grid framework that will, for the first time, ... more This paper introduces the design of SymGrid, a new Grid framework that will, for the first time, allow multiple invocations of symbolic computing applications to interact via the Grid. SymGrid is designed to support the specific needs of symbolic computation, including computational steering (greater interactivity), complex data structures, and domain-specific computational patterns (for irregular parallelism).
This paper introduces HaskSkel: an implementation of algorithmic skeletons for the non-strict fun... more This paper introduces HaskSkel: an implementation of algorithmic skeletons for the non-strict functional language Haskell. The implementation is based on the evaluation strategy mechanism for Glasgow Parallel Haskell (GpH), and has been tested in both simulated and real parallel settings.
This paper describes CASH (the Computer Algebra SHell), a new interface that allows Haskell progr... more This paper describes CASH (the Computer Algebra SHell), a new interface that allows Haskell programmers to access the complete functionality of a number of computer algebra systems directly and interactively. Using CASH, Haskell programmers can access previously-unavailable mathematical software. Additionally, users of computer algebra systems can exploit the rapidly growing Haskell code base and its rich set of libraries.
Abstract This chapter gives an overview of the amortised analysis approach that we have used for ... more Abstract This chapter gives an overview of the amortised analysis approach that we have used for Hume. We introduce an analogy with carbon emissions, describe amortised analysis in general terms, outline how we will develop an automatic analysis and introduce the Hume programming language that will be used as the basis for undertaking our analysis.
It is widely claimed that functional languages are particularly suitable for programming parallel... more It is widely claimed that functional languages are particularly suitable for programming parallel computers. A claimed advantage is that the programmer is not burdened with details of task creation, placement, scheduling, and synchronisation, these decisions being taken by the system instead. Leaving aside the question of whether a pure functional language is expressive enough to encompass all the parallel algorithms we might wish to program, there remains the question of how e ectively the compiler and run-time system map the program onto a real parallel system, a task usually carried out mostly by the programmer. This is the question we address in this paper.
Abstract Modern grid and Web technologies provide straightforward access to applications and serv... more Abstract Modern grid and Web technologies provide straightforward access to applications and services running on remote resources. The user (which may be a software developer or even an application) should be able to discover such a service and dynamically load an interface to interact with it. While some software systems support either fully or partially automatic creation of client software to access Web services, fewer support grid services in the same way.
This paper defines a dynamic semantics for the functional programming language Haskell. The seman... more This paper defines a dynamic semantics for the functional programming language Haskell. The semantics defines the meaning of expressions, modules and programs, including all nonoptionalI/O requests. All constructs in the Haskell language are considered. We use thesame natural semantics style as the definition of Standard ML. For brevity, we assume that the translations specified by the Wadler/Peyton Jones staticsemantics have been performed, and base our source language on the target language of thestatic semantics.
Abstract Over 20 implementations of different functional languages are benchmarked using the same... more Abstract Over 20 implementations of different functional languages are benchmarked using the same program, a floating-point intensive application taken from molecular biology. The principal aspects studied are compile time and execution time for the varying implementations. An important consideration is how the program can be modified and tuned to obtain maximal performance on each language implementation.
Abstract We present the results of the first four years of the European research project SCIEnce ... more Abstract We present the results of the first four years of the European research project SCIEnce (www. symbolic-computation. org), which aims to provide key infrastructure for symbolic computation research. A primary outcome of the project is that we have developed a new way of combining computer algebra systems using the Symbolic Computation Software Composability Protocol (SCSCP), in which both protocol messages and data are encoded in the OpenMath format.
To take advantage of distributed-memory parallel machines it is essential to have good control of... more To take advantage of distributed-memory parallel machines it is essential to have good control of task granularity. This paper describes a fairly accurate parallel simulator for Haskell, based on the Glasgow compiler, and complementary tools for visualising task granularities. Together these tools allow us to study the e ects of various annotations on task granularity on a variety of simulated parallel architectures. They also provide a more precise tool for the study of parallel execution than has previously been available for Haskell programs.
Abstract We investigate the claim that functional languages o er low-cost parallelism in the cont... more Abstract We investigate the claim that functional languages o er low-cost parallelism in the context of symbolic programs on modest parallel architectures. We base our investigation on several large programs that we have written or parallelised in Glasgow Parallel Haskell (GpH). The programs cover a range of application areas, use several parallel programming paradigms, and are measured on two very di erent parallel architectures.
The GRIP architecture allows efficient execution of functional programs on a multi-processor buil... more The GRIP architecture allows efficient execution of functional programs on a multi-processor built from standard hardware components. State-of-the-art compilation techniques are combined with sophisticated runtime resource-control to give good parallel performance. This paper reports the results of running GRIP on an application which is apparently unsuited to the basic functional model: a database transaction manager incorporating updates as well as lookup transactions.
Abstract This paper studies the runtime behaviour of various parallel divideand-conquer algorithm... more Abstract This paper studies the runtime behaviour of various parallel divideand-conquer algorithms written in a non-strict functional language, when three common granularity control mechanisms are used: a simple cut-o, a priority thread creation and a priority scheduling mechanism. These mechanisms use granularity information that is currently provided via annotations to improve the performance of the parallel programs.
Abstract In principle, pure functional languages promise straightforward architecture-independent... more Abstract In principle, pure functional languages promise straightforward architecture-independent parallelism. We investigate the validity of this claim in the context of our highly-portable implementation of an implicitly-parallel functional language: the GUM implementation of Glasgow Parallel Haskell (GpH). We discuss architecture independence at two levels: lowlevel (ie the implementation) and high-level (ie the programmer).
Abstract: We investigate two similar but contrasting parallel functional language designs: Eden a... more Abstract: We investigate two similar but contrasting parallel functional language designs: Eden and GPH. Both languages use the non-strict functional language Haskell as a core expression language, both are implemented as extensions of the high performance Glasgow Haskell Compiler (GHC), and both implementations are available on the same distributed architecture: a Beowulf cluster. This allows an exceptionally pure comparison of the language design characteristics and their impact on parallel performance.
School of Computer Science, University of St Andrews, St Andrews, Scotland. б а зКзиЙ в К Кй. Ь а... more School of Computer Science, University of St Andrews, St Andrews, Scotland. б а зКзиЙ в К Кй. Ь а З ЙНПП Й ПО Н, м З ЙНПП Й ПО зиж иК This paper introduces a high level abstract machine design for the domain-specific functional language Hume, demonstrating how strong cost models can be constructed for a restricted subset of the language, providing good time and space behaviour, while maintaining good language abstractions and providing formal translations. l Introduction
Abstract We describe a new automatic static analysis for determining upper-bound functions on the... more Abstract We describe a new automatic static analysis for determining upper-bound functions on the use of quantitative resources for strict, higher-order, polymorphic, recursive programs dealing with possibly-aliased data. Our analysis is a variant of Tarjan's manual amortised cost analysis technique. We use a type-based approach, exploiting linearity to allow inference, and place a new emphasis on the number of references to a data object. The bounds we infer depend on the sizes of the various inputs to a program.
Parallel functional programming has a relatively long history. Burge was one of the first to sugg... more Parallel functional programming has a relatively long history. Burge was one of the first to suggest the basic technique of evaluating function arguments in parallel, with the possibility of functions absorbing unevaluated arguments and perhaps also exploiting speculative evaluation 22]. Berkling also considered the application of functional languages to parallel processing 17].
This paper uses Template Haskell to automatically select appropriate skeleton implementations in ... more This paper uses Template Haskell to automatically select appropriate skeleton implementations in the Eden parallel dialect of Haskell. The approach allows implementation parameters to be statically tuned according to architectural cost models based on source analyses. This permits us to target a range of parallel architecture classes from a single source specification. A major advantage of the approach is that cost models are user-definable and can be readily extended to new data or computation structures etc.
This paper introduces the design of SymGrid, a new Grid framework that will, for the first time, ... more This paper introduces the design of SymGrid, a new Grid framework that will, for the first time, allow multiple invocations of symbolic computing applications to interact via the Grid. SymGrid is designed to support the specific needs of symbolic computation, including computational steering (greater interactivity), complex data structures, and domain-specific computational patterns (for irregular parallelism).
This paper introduces HaskSkel: an implementation of algorithmic skeletons for the non-strict fun... more This paper introduces HaskSkel: an implementation of algorithmic skeletons for the non-strict functional language Haskell. The implementation is based on the evaluation strategy mechanism for Glasgow Parallel Haskell (GpH), and has been tested in both simulated and real parallel settings.
This paper describes CASH (the Computer Algebra SHell), a new interface that allows Haskell progr... more This paper describes CASH (the Computer Algebra SHell), a new interface that allows Haskell programmers to access the complete functionality of a number of computer algebra systems directly and interactively. Using CASH, Haskell programmers can access previously-unavailable mathematical software. Additionally, users of computer algebra systems can exploit the rapidly growing Haskell code base and its rich set of libraries.
Abstract This chapter gives an overview of the amortised analysis approach that we have used for ... more Abstract This chapter gives an overview of the amortised analysis approach that we have used for Hume. We introduce an analogy with carbon emissions, describe amortised analysis in general terms, outline how we will develop an automatic analysis and introduce the Hume programming language that will be used as the basis for undertaking our analysis.
It is widely claimed that functional languages are particularly suitable for programming parallel... more It is widely claimed that functional languages are particularly suitable for programming parallel computers. A claimed advantage is that the programmer is not burdened with details of task creation, placement, scheduling, and synchronisation, these decisions being taken by the system instead. Leaving aside the question of whether a pure functional language is expressive enough to encompass all the parallel algorithms we might wish to program, there remains the question of how e ectively the compiler and run-time system map the program onto a real parallel system, a task usually carried out mostly by the programmer. This is the question we address in this paper.
Abstract Modern grid and Web technologies provide straightforward access to applications and serv... more Abstract Modern grid and Web technologies provide straightforward access to applications and services running on remote resources. The user (which may be a software developer or even an application) should be able to discover such a service and dynamically load an interface to interact with it. While some software systems support either fully or partially automatic creation of client software to access Web services, fewer support grid services in the same way.
This paper defines a dynamic semantics for the functional programming language Haskell. The seman... more This paper defines a dynamic semantics for the functional programming language Haskell. The semantics defines the meaning of expressions, modules and programs, including all nonoptionalI/O requests. All constructs in the Haskell language are considered. We use thesame natural semantics style as the definition of Standard ML. For brevity, we assume that the translations specified by the Wadler/Peyton Jones staticsemantics have been performed, and base our source language on the target language of thestatic semantics.
Abstract Over 20 implementations of different functional languages are benchmarked using the same... more Abstract Over 20 implementations of different functional languages are benchmarked using the same program, a floating-point intensive application taken from molecular biology. The principal aspects studied are compile time and execution time for the varying implementations. An important consideration is how the program can be modified and tuned to obtain maximal performance on each language implementation.
Abstract We present the results of the first four years of the European research project SCIEnce ... more Abstract We present the results of the first four years of the European research project SCIEnce (www. symbolic-computation. org), which aims to provide key infrastructure for symbolic computation research. A primary outcome of the project is that we have developed a new way of combining computer algebra systems using the Symbolic Computation Software Composability Protocol (SCSCP), in which both protocol messages and data are encoded in the OpenMath format.
To take advantage of distributed-memory parallel machines it is essential to have good control of... more To take advantage of distributed-memory parallel machines it is essential to have good control of task granularity. This paper describes a fairly accurate parallel simulator for Haskell, based on the Glasgow compiler, and complementary tools for visualising task granularities. Together these tools allow us to study the e ects of various annotations on task granularity on a variety of simulated parallel architectures. They also provide a more precise tool for the study of parallel execution than has previously been available for Haskell programs.
Abstract We investigate the claim that functional languages o er low-cost parallelism in the cont... more Abstract We investigate the claim that functional languages o er low-cost parallelism in the context of symbolic programs on modest parallel architectures. We base our investigation on several large programs that we have written or parallelised in Glasgow Parallel Haskell (GpH). The programs cover a range of application areas, use several parallel programming paradigms, and are measured on two very di erent parallel architectures.
The GRIP architecture allows efficient execution of functional programs on a multi-processor buil... more The GRIP architecture allows efficient execution of functional programs on a multi-processor built from standard hardware components. State-of-the-art compilation techniques are combined with sophisticated runtime resource-control to give good parallel performance. This paper reports the results of running GRIP on an application which is apparently unsuited to the basic functional model: a database transaction manager incorporating updates as well as lookup transactions.
Abstract This paper studies the runtime behaviour of various parallel divideand-conquer algorithm... more Abstract This paper studies the runtime behaviour of various parallel divideand-conquer algorithms written in a non-strict functional language, when three common granularity control mechanisms are used: a simple cut-o, a priority thread creation and a priority scheduling mechanism. These mechanisms use granularity information that is currently provided via annotations to improve the performance of the parallel programs.
Abstract In principle, pure functional languages promise straightforward architecture-independent... more Abstract In principle, pure functional languages promise straightforward architecture-independent parallelism. We investigate the validity of this claim in the context of our highly-portable implementation of an implicitly-parallel functional language: the GUM implementation of Glasgow Parallel Haskell (GpH). We discuss architecture independence at two levels: lowlevel (ie the implementation) and high-level (ie the programmer).
Abstract: We investigate two similar but contrasting parallel functional language designs: Eden a... more Abstract: We investigate two similar but contrasting parallel functional language designs: Eden and GPH. Both languages use the non-strict functional language Haskell as a core expression language, both are implemented as extensions of the high performance Glasgow Haskell Compiler (GHC), and both implementations are available on the same distributed architecture: a Beowulf cluster. This allows an exceptionally pure comparison of the language design characteristics and their impact on parallel performance.
School of Computer Science, University of St Andrews, St Andrews, Scotland. б а зКзиЙ в К Кй. Ь а... more School of Computer Science, University of St Andrews, St Andrews, Scotland. б а зКзиЙ в К Кй. Ь а З ЙНПП Й ПО Н, м З ЙНПП Й ПО зиж иК This paper introduces a high level abstract machine design for the domain-specific functional language Hume, demonstrating how strong cost models can be constructed for a restricted subset of the language, providing good time and space behaviour, while maintaining good language abstractions and providing formal translations. l Introduction
Uploads
Papers by Kevin Hammond