A provable time and space efficient implementation of NESL

Blelloch, Guy E.; Greiner, John

doi:10.1145/232629.232650

Cited by 57 publications

(46 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Recently, Minamide [1999] has proven that a standard CPS translation preserves the maximum amount of reachable space within a constant factor. He defines a spaceprofiling semantics for the simply typed lambda calculus that refines the work of Blelloch and Greiner [1996]. If we were to augment our semantics with this sort of space-profiling information, we may be able to prove a similar result for our translation.…”

Section: Theorem (Cps Type Preservation) If ·; · E : Int ∅ Then { }mentioning

confidence: 95%

Typed memory management via static capabilities

Walker

Crary

Morrisett

2000

ACM Trans. Program. Lang. Syst.

View full text Add to dashboard Cite

Region-based memory management is an alternative to standard tracing garbage collection that makes operations such as memory deallocation explicit but verifiably safe. In this article, we present a new compiler intermediate language, called the Capability Language (CL), that supports region-based memory management and enjoys a provably safe type system. Unlike previous regionbased type systems, region lifetimes need not be lexically scoped, and yet the language may be checked for safety without complex analyses. Therefore, our type system may be deployed in settings such as extensible operating systems where both the performance and safety of untrusted code is important. The central novelty of the language is the use of static capabilities to specify the permissibility of various operations, such as memory access and deallocation. In order to ensure capabilities are relinquished properly, the type system tracks aliasing information using a form of bounded quantification. Moreover, unlike previous work on region-based type systems, the proof of soundness of our type system is relatively simple, employing only standard syntactic techniques. In order to show how our language may be used in practice, we show how to translate a variant of Tofte and Talpin's high-level type-and-effects system for region-based memory management into our language. When combined with known region inference algorithms, this translation provides a way to compile source-level languages to CL.

show abstract

Section: Theorem (Cps Type Preservation) If ·; · E : Int ∅ Then { }mentioning

confidence: 95%

Typed memory management via static capabilities

Walker

Crary

Morrisett

2000

ACM Trans. Program. Lang. Syst.

View full text Add to dashboard Cite

show abstract

“…An abstract machine called a VCODE interpreter then executes these programs on physical hardware. Flattening transformations and their ability to transform nested (recursive) data parallelism into "flat" vector operations while preserving the asymptotic complexity of programs is a key contribution of NESL [8] and very recent work on using NESL-like nested data parallelism for GPUs [4] and multicore [3] has focused on extending flattening approaches to better match such hardware.…”

Section: Example: Nesl and Ndpcmentioning

confidence: 99%

A Case for Transforming Parallel Runtimes Into Operating System Kernels

Hale

Dinda

2015

Proceedings of the 24th International Symposium on High-Performance Parallel and Distributed Computing

View full text Add to dashboard Cite

The needs of parallel runtime systems and the increasingly sophisticated languages and compilers they support do not line up with the services provided by general-purpose OSes. Furthermore, the semantics available to the runtime are lost at the system-call boundary in such OSes. Finally, because a runtime executes at user-level in such an environment, it cannot leverage hardware features that require kernel-mode privileges-a large portion of the functionality of the machine is lost to it. These limitations warp the design, implementation, functionality, and performance of parallel runtimes. We summarize the case for eliminating these compromises by transforming parallel runtimes into OS kernels. We also demonstrate that it is feasible to do so. Our evidence comes from Nautilus, a prototype kernel framework that we built to support such transformations. After describing Nautilus, we report on our experiences using it to transform three very different runtimes into kernels.

show abstract

“…For control-flow languages, there is a large body of work on qualitative and quantitative reasoning of performance-related properties, e.g., WCET [12], cost semantics [13], resource bound certification [14], resource usage analysis [15], amortized resource analysis [16], and Energy Types [30]. RATE TYPES focuses on (data rates of) data-flow programming models, and their impact on energy consumption and CPU allocation.…”

Section: Related Workmentioning

confidence: 99%

“…At the core of this exploration is quantitative reasoning about performance-related properties, an active area of research (e.g. [12][13][14][15][16]). RATE TYPES follows a less explored path by focusing on quantitative reasoning over data-flow programming models.…”

Section: Introductionmentioning

confidence: 99%

Rate types for stream programs

Bartenstein

Liu

2014

SIGPLAN Not.

View full text Add to dashboard Cite

Abstract. We introduce RATE TYPES, a novel type system to reason about and optimize data-intensive programs. Built around stream languages, RATE TYPES performs static quantitative reasoning about stream rates -the frequency of data items in a stream being consumed, processed, and produced -a critical performance characteristic previously addressed by numerous experimental approaches but few foundational efforts. Even though streams are fundamentally dynamic, we find two essential concepts of stream rate control -throughput ratio and natural rate -are intimately related to the program structure itself and can be effectively reasoned about by a type system. RATE TYPES is proven sound over a time-aware and parallelism-aware operational semantics. The strong soundness result tolerates arbitrary schedules, and does not require any synchronization between stream filters. We further demonstrate the applications of RATE TYPES in energy-efficient computing and CPU allocation on multi-core architectures.

show abstract

A provable time and space efficient implementation of NESL

Abstract: In this paper we prove time and space bounds for the implementation of the programming language NESL on various parallel machine models. NESL is a sugared typed J-calculus with a set of array primitives and an explicit parallel map over arrays.Our

Cited by 57 publications

References 23 publications

Typed memory management via static capabilities

Typed memory management via static capabilities

A Case for Transforming Parallel Runtimes Into Operating System Kernels

Rate types for stream programs

Contact Info

Product

Resources

About