Foundations of Data-parallel Skeletons

Fischer, Jörg; Gorlatch, Sergei; Bischof, Holger

doi:10.1007/978-1-4471-0097-3_1

Cited by 4 publications

(4 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The term algorithmic skeletons was coined by Cole [15] to denote high-level constructs encapsulating parallelism, communication, and synchronization, having an associated cost complexity. They originate from the need for programming languages designed from a fixed set of constructs capturing common computation patterns [6] and from the fact that many parallel applications share a common set of known interaction patterns like processor farms or pipelines [15,20]. The key idea behind algorithmic skeletons is that there exists a set of functions on structured data types called catamorphisms [43] that respects the way in which objects of their type is built.…”

Section: Exploiting Parallelismmentioning

confidence: 99%

“…merges two vectors element-wise while performing partial application on the functions "below" accordingly, and An important theorem in the theory of algorithmic skeletons, factorization [20,43], states that a function h is a catamorphism if it can be factorized like in Equation (2). As of this, the skeletons in Table 3, usually used in CPS design, are in fact patterns of map (i.e., the generic farm S ) and reduce (i.e., reduce S ).…”

Section: Skeleton Layer Definitionsmentioning

confidence: 99%

See 1 more Smart Citation

ForSyDe-Atom

Ungureanu

Medeiros

Sundström

et al. 2021

ACM Trans. Embed. Comput. Syst.

View full text Add to dashboard Cite

We present ForSyDe-Atom, a formal framework intended as an entry point for disciplined design of complex cyber-physical systems. This framework provides a set of rules for combining several domain-specific languages as structured, enclosing layers to orthogonalize the many aspects of system behavior, yet study their interaction in tandem . We define four layers: one for capturing timed interactions in heterogeneous systems, one for structured parallelism, one for modeling uncertainty, and one for describing component properties. This framework enables a systematic exploitation of design properties in a design flow by facilitating the stepwise projection of certain layers of interest, the isolated analysis and refinement on projections, and the seamless reconstruction of a system model by virtue of orthogonalization. We demonstrate the capabilities of this approach by providing a compact yet expressive model of an active electronically scanned array antenna and signal processing chain, simulate it, validate its conformity with the design specifications, refine it, synthesize a sub-system to VHDL and sequential code, and co-simulate the generated artifacts.

show abstract

Section: Exploiting Parallelismmentioning

confidence: 99%

Section: Skeleton Layer Definitionsmentioning

confidence: 99%

ForSyDe-Atom

Ungureanu

Medeiros

Sundström

et al. 2021

ACM Trans. Embed. Comput. Syst.

View full text Add to dashboard Cite

show abstract

“…If we consider signals wrapped in regular structures H(S) (short-hand S ), then we can formulate homomorphisms as atoms on those structures, taking processes as arguments. As a consequence those atoms will instantiate regular networks of processes which expose the inherent parallelism, concluding that the atoms specific to this layer are indeed algorithmic skeletons [12].…”

Section: E the Skeleton Layermentioning

confidence: 99%

“…3) Algorithmic skeletons: Cole coined this term [10] to denote high-level constructs encapsulating parallelism, communication and synchronization with an associated cost complexity. The key idea behind algorithmic skeletons is that there exists a set of functions on structured data types called homomorphisms which respect the way in which objects of their type are built [11], [12]. Intuitively, homomorphisms enable computation on separate parts of a large data structure to be performed in parallel.…”

Section: Introductionmentioning

confidence: 99%

A layered formal framework for modeling of cyber-physical systems

Ungureanu

Sander

2017

Design, Automation &Amp; Test in Europe Conference &Amp; Exhibition (DATE), 2017

View full text Add to dashboard Cite

Designing cyber-physical systems is highly challenging due to its manifold interdependent aspects such as composition, timing, synchronization and behavior. Several formal models exist for description and analysis of these aspects, but they focus mainly on a single or only a few system properties. We propose a formal composable framework which tackles these concerns in isolation, while capturing interaction between them as a single layered model. This yields a holistic, fine-grained, hierarchical and structured view of a cyber-physical system. We demonstrate the various benefits for modeling, analysis and synthesis through a typical example.

show abstract