Introducing Kansas Lava

Gill, Andy; Bull, Tristan; Kimmell, Garrin; Perrins, Erik; Komp, Ed; Werling, Brett W.

doi:10.1007/978-3-642-16478-1_2

Cited by 34 publications

(24 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…. This is very much like the approach taken by Elliott [13], Gill et al [14], and Mainland and Morrisett [24]. A difference of our approach is that we maintain full type information of the embedded language in the term tree (not shown above to favour conciseness) and use type-preserving transformations in the frontend.…”

Section: The Surface Languagementioning

confidence: 85%

Accelerating Haskell array codes with multicore GPUs

Chakravarty

Keller

Lee

et al. 2011

Proceedings of the Sixth Workshop on Declarative Aspects of Multicore Programming

185

106

View full text Add to dashboard Cite

Current GPUs are massively parallel multicore processors optimised for workloads with a large degree of SIMD parallelism. Good performance requires highly idiomatic programs, whose development is work intensive and requires expert knowledge.To raise the level of abstraction, we propose a domain-specific high-level language of array computations that captures appropriate idioms in the form of collective array operations. We embed this purely functional array language in Haskell with an online code generator for NVIDIA's CUDA GPGPU programming environment. We regard the embedded language's collective array operations as algorithmic skeletons; our code generator instantiates CUDA implementations of those skeletons to execute embedded array programs.This paper outlines our embedding in Haskell, details the design and implementation of the dynamic code generator, and reports on initial benchmark results. These results suggest that we can compete with moderately optimised native CUDA code, while enabling much simpler source programs.

show abstract

Section: The Surface Languagementioning

confidence: 85%

Accelerating Haskell array codes with multicore GPUs

Chakravarty

Keller

Lee

et al. 2011

Proceedings of the Sixth Workshop on Declarative Aspects of Multicore Programming

185

106

View full text Add to dashboard Cite

show abstract

“…For instance, Chalmers Lava [24] introduces a library driven extension for Haskell, called Observable Sharing (ObS) which simplifies the detection of shared subcircuits in the design [25]. Finally, there is an experimental Lava setup working with Type Directed Observable Sharing (TD-ObS), which is commonly called Kansas Lava [26,27]. It is replacing the detection mechanism for shared subcircuits providing a more general technique to convert tree structures into graphs.…”

Section: A Brief Historymentioning

confidence: 97%

Processor design using a functional hardware description language

Schulze¹,

Sawitzki

2012

Microprocessors and Microsystems

View full text Add to dashboard Cite

“…Since 1998 a renewed interest has given birth to functional hardware description languages such as Lava [6] (with variants from Chalmers, Xilinx, York and in 2009 Kansas Lava [7,8]) and ForSyDe [9]. These examples are usually capable of generating hardware for many application domains.…”

Section: Related Workmentioning

confidence: 99%

Design space exploration for automatically generated cryptographic hardware using functional languages

Wolfs¹,

Aerts²,

Mentens³

2012

22nd International Conference on Field Programmable Logic and Applications (FPL)

View full text Add to dashboard Cite

This paper presents an EDA (Electronic Design Automation) tool that generates basic building blocks for cryptographic hardware in VHDL. The purpose of the tool is to decrease the design time of cryptographic hardware and to allow designers to make abstraction of both the arithmetic and design complexity. The tool generates multiple implementations for one arithmetic description and then benchmarks the implementations to find the most optimal, based upon design space parameters. These parameters consist of area and speed requirements. We present datapath and control logic results for a Xilinx Virtex-5 FPGA.The novelty in our approach lies in the fact that we exploit the higher-order features of functional languages to facilitate the design space exploration and that we take benefit from the strength of the third-party synthesis tool by generating VHDL code at an abstraction level that is higher than the gate level. Nevertheless, in this stage of the development of the tool, the different cryptographic architectures are hand-made and the selection of the most optimal solution, based upon user requirements, is done by exhaustive search. This means that the tool leaves room for improvement, but forms a solid base for further development. MOTIVATION AND CONTRIBUTIONIn storage and communication of digital data, security is an important issue. To provide confidentiality, authentication, secure key exchange, privacy, . . . cryptographic algorithms are needed. When requirements such as high speed, low area, low power consumption and/or high security level are an issue, these cryptographic algorithms are implemented in hardware coprocessors. Most of the times cryptographic hardware needs to be designed with as little overhead as possible because it does not contribute to the core functionality of the application. However, to achieve a sufficient level of security, cryptographic algorithms and in particular public * Davy Wolfs was funded by BOF project CREA/09/016 of the Katholieke Universiteit Leuven. key algorithms consist of rather complicated mathematical operations on large numbers. Consequently, the challenge for the design of cryptographic hardware is to find a suitable architecture for a rather complicated mathematical operation on large numbers that results in a restricted overhead in time, area and/or power consumption.Like almost all digital designs, a cryptographic hardware architecture consists of a datapath and control logic. In this paper, we explore the design space of both. We are specifically interested in the area and speed of the actual hardware after synthesis by both FPGA and ASIC (commercial) third-party tools. Therefore we make sure that the VHDL code generated by our Lava programs is platformindependent. We also shift from basic logic gates to formal FSMs for the description of the control path, as is common in VHDL programming. These state machines can be simulated in Haskell, and exported to native VHDL FSM code on the behavioral level, which gives more optimization possibilities to the synthesis ...

show abstract

Introducing Kansas Lava

Cited by 34 publications

References 12 publications

Accelerating Haskell array codes with multicore GPUs

Accelerating Haskell array codes with multicore GPUs

Processor design using a functional hardware description language

Design space exploration for automatically generated cryptographic hardware using functional languages

Contact Info

Product

Resources

About