Fritz Henglein

Futhark: purely functional GPU-programming with nested parallelism and in-place array updates

Futhark is a purely functional data-parallel array language that offers a machine-neutral programming model and an optimising compiler that generates OpenCL code for GPUs. This paper presents the design and implementation of three key features of Futhark that seek a suitable middle ground with imperative approaches. First, in order to express efficient code inside the parallel constructs, we introduce a simple type system for in-place updates that ensures referential transparency and supports equational reasoning. Second, we furnish Futhark with parallel operators capable of expressing efficient strength-reduced code, along with their fusion rules. Third, we present a flattening transformation aimed at enhancing the degree of parallelism that (i) builds on loop interchange and distribution but uses higher-order reasoning rather than array-dependence analysis, and (ii) still allows further locality-of-reference optimisations. Finally, an evaluation on 16 benchmarks demonstrates the impact of the language and compiler features and shows application-level performance competitive with hand-written GPU code.

Kleenex: compiling nondeterministic transducers to deterministic streaming transducers

We present and illustrate Kleenex, a language for expressing general nondeterministic finite transducers, and its novel compilation to streaming string transducers with essentially optimal streaming behavior, worst-case linear-time performance and sustained high throughput. Its underlying theory is based on transducer decomposition into oracle and action machines: the oracle machine performs streaming greedy disambiguation of the input; the action machine performs the output actions. In use cases Kleenex achieves consistently high throughput rates around the 1 Gbps range on stock hardware. It performs well, especially in complex use cases, in comparison to both specialized and related tools such as GNUawk, GNUsed, GNUgrep, RE2, Ragel and regular-expression libraries.

Infinitary Axiomatization of the Equational Theory of Context-Free Languages.

We give a natural complete infinitary axiomatization of the equational theory of the context-free languages, answering a question of Lei{ss} (1992).

FinPar: A Parallel Financial Benchmark

Commodity many-core hardware is now mainstream, but parallel programming models are still lagging behind in efficiently utilizing the application parallelism. There are (at least) two principal reasons for this. First, real-world programs often take the form of a deeply nested composition of parallel operators, but mapping the available parallelism to the hardware requires a set of transformations that are tedious to do by hand and beyond the capability of the common user. Second, the best optimization strategy, such as what to parallelize and what to efficiently sequentialize, is often sensitive to the input dataset and therefore requires multiple code versions that are optimized differently, which also raises maintainability problems.n This article presents three array-based applications from the financial domain that are suitable for gpgpu execution. Common benchmark-design practice has been to provide the same code for the sequential and parallel versions that are optimized for only one class of datasets. In comparison, we document (1) all available parallelism via nested map-reduce functional combinators, in a simple Haskell implementation that closely resembles the original code structure, (2) the invariants and code transformations that govern the main trade-offs of a data-sensitive optimization space, and (3) report target cpu and multiversion gpgpu code together with an evaluation that demonstrates optimization trade-offs and other difficulties. We believe that this work provides useful insight into the language constructs and compiler infrastructure capable of expressing and optimizing such applications, and we report in-progress work in this direction.

Relational Algebra by Way of Adjunctions

Bulk types such as sets, bags, and lists are monads, and therefore support a notation for database queries based on comprehensions. This fact is the basis of much work on database query languages. The monadic structure easily explains most of standard relational algebra---specifically, selections and projections---allowing for an elegant mathematical foundation for those aspects of database query language design. Most, but not all: monads do not immediately offer an explanation of relational join or grouping, and hence important foundations for those crucial aspects of relational algebra are missing. The best they can offer is cartesian product followed by selection. Adjunctions come to the rescue: like any monad, bulk types also arise from certain adjunctions; we show that by paying due attention to other important adjunctions, we can elegantly explain the rest of standard relational algebra. In particular, graded monads provide a mathematical foundation for indexing and grouping, which leads directly to an efficient implementation, even of joins.

Infinitary Axiomatization of the Equational Theory of Context-Free Languages

We give a natural complete infinitary axiomatization of the equational theory of the context-free languages, answering a question of Leis (1992).

PEG parsing in less space using progressive tabling and dynamic analysis

Tabular top-down parsing and its lazy variant, Packrat, are linear-time execution models for the TDPL family of recursive descent parsers with limited backtracking. Exponential work due to backtracking is avoided by tabulating the result of each (nonterminal, offset)-pair at the expense of always using space proportional to the product of the input length and grammar size. Current methods for limiting the space usage rely either on manual annotations or on static analyses that are sensitive to the syntactic structure of the grammar. We present progressive tabular parsing (PTP), a new execution model which progressively computes parse tables for longer prefixes of the input and simultaneously generates a leftmost expansion of the parts of the parse tree that can be resolved. Table columns can be discarded on-the-fly as the expansion progresses through the input string, providing best-case constant and worst-case linear memory use. Furthermore, semantic actions are scheduled before the parser has seen the end of the input. The scheduling is conservative in the sense that no action has to be undone in the case of backtracking. The time complexity is O(dmn) where m is the size of the parser specification, n is the size of the input string, and d is either a configured constant or the maximum parser stack depth. For common data exchange formats such as JSON, we demonstrate practically constant space usage.

2016 POPL impressions

The 43rd ACM Symposium on Principles of Programming Languages, POPL, took place in St. Petersburg, Florida, January 20th--22nd. POPL is the longest-running annual conference series in programming languages. It is one of SIGPLANs four flagship conferences and has throughout its history consistently been one of the most highly regarded programming language research conferences. Like its sibling conference, Programming Language Design and Implementation (PLDI), POPL covers both theoretical and experimental topics, but traditionally leans more to the left where the PLDI leans more to the right on the theory-practice spectrum. (Another difference is that POPL does not declare itself to be the premier forum, just a forum for programming language research.)

Modal Intersection Types, Two-Level Languages, and Staged Synthesis

A typed