Michael J. Steindorfer

Optimizing hash-array mapped tries for fast and lean immutable JVM collections

The data structures under-pinning collection API (e.g. lists, sets, maps) in the standard libraries of programming languages are used intensively in many applications. The standard libraries of recent Java Virtual Machine languages, such as Clojure or Scala, contain scalable and well-performing immutable collection data structures that are implemented as Hash-Array Mapped Tries (HAMTs). HAMTs already feature efficient lookup, insert, and delete operations, however due to their tree-based nature their memory footprints and the runtime performance of iteration and equality checking lag behind array-based counterparts. This particularly prohibits their application in programs which process larger data sets. In this paper, we propose changes to the HAMT design that increase the overall performance of immutable sets and maps. The resulting general purpose design increases cache locality and features a canonical representation. It outperforms Scala’s and Clojure’s data structure implementations in terms of memory footprint and runtime efficiency of iteration (1.3–6.7x) and equality checking (3–25.4x).

Code specialization for memory efficient hash tries (short paper)

The hash trie data structure is a common part in standard collection libraries of JVM programming languages such as Clojure and Scala. It enables fast immutable implementations of maps, sets, and vectors, but it requires considerably more memory than an equivalent array-based data structure. This hinders the scalability of functional programs and the further adoption of this otherwise attractive style of programming. In this paper we present a product family of hash tries. We generate Java source code to specialize them using knowledge of JVM object memory layout. The number of possible specializations is exponential. The optimization challenge is thus to find a minimal set of variants which lead to a maximal loss in memory footprint on any given data. Using a set of experiments we measured the distribution of internal tree node sizes in hash tries. We used the results as a guidance to decide which variants of the family to generate and which variants should be left to the generic implementation. A preliminary validating experiment on the implementation of sets and maps shows that this technique leads to a median decrease of 55% in memory footprint for maps (and 78% for sets), while still maintaining comparable performance. Our combination of data analysis and code specialization proved to be effective.

M3: A general model for code analytics in rascal

This short paper introduces M3, a simple and extensible model for capturing facts about source code for future analysis. M3 is a core part of the standard library of the Rascal meta programming language.We motivate it, position it to related work and detail the key design aspects.

Towards a software product line of trie-based collections

Collection data structures in standard libraries of programming languages are designed to excel for the average case by carefully balancing memory footprint and runtime performance. These implicit design decisions and hard-coded trade-offs do constrain users from using an optimal variant for a given problem. Although a wide range of specialized collections is available for the Java Virtual Machine (JVM), they introduce yet another dependency and complicate user adoption by requiring specific Application Program Interfaces (APIs) incompatible with the standard library. A product line for collection data structures would relieve library designers from optimizing for the general case. Furthermore, a product line allows evolving the potentially large code base of a collection family efficiently. The challenge is to find a small core framework for collection data structures which covers all variations without exhaustively listing them, while supporting good performance at the same time. We claim that the concept of Array Mapped Tries (AMTs) embodies a high degree of commonality in the sub-domain of immutable collection data structures. AMTs are flexible enough to cover most of the variability, while minimizing code bloat in the generator and the generated code. We implemented a Data Structure Code Generator (DSCG) that emits immutable collections based on an AMT skeleton foundation. The generated data structures outperform competitive hand-optimized implementations, and the generator still allows for customization towards specific workloads.

PIE: A Domain-Specific Language for Interactive Software Development Pipelines

Context. Software development pipelines are used for automating essential parts of software engineering processes, such as build automation and continuous integration testing. In particular, interactive pipelines, which process events in a live environment such as an IDE, require timely results for low-latency feedback, and persistence to retain low-latency feedback between restarts. Inquiry. Developing an incrementalized and persistent version of a pipeline is one way to reduce feedback latency, but requires dependency tracking, cache invalidation, and other complicated techniques. Therefore, interactivity complicates pipeline development if timeliness and persistence become responsibilities of the pipeline programmer, rather than being supported by the underlying system. Approach. We develop Pipelines for Interactive Environments (PIE), a Domain-Specific Language (DSL), API, and runtime for developing interactive software development pipelines, where ease of development is a focus. Knowledge. PIE provides a straightforward programming model that enables direct and concise expression of pipelines without boilerplate, reducing the development and maintenance effort of pipelines. Compiled pipeline programs can be embedded into interactive environments such as code editors and IDEs, enabling timely feedback at a low cost. Grounding. Compared to the state of the art, PIE reduces the code required to express an interactive pipeline by a factor of 6 in a case study on syntax-aware editors. Furthermore, we evaluate PIE in two case studies of complex interactive software development scenarios, demonstrating that PIE can handle complex interactive pipelines in a straightforward and concise way.

Declarative specification of indentation rules: a tooling perspective on parsing and pretty-printing layout-sensitive languages

In layout-sensitive languages, the indentation of an expression or statement can influence how a program is parsed. While some of these languages (e.g., Haskell and Python) have been widely adopted, there is little support for software language engineers in building tools for layout-sensitive languages. As a result, parsers, pretty-printers, program analyses, and refactoring tools often need to be handwritten, which decreases the maintainability and extensibility of these tools. Even state-of-the-art language workbenches have little support for layout-sensitive languages, restricting the development and prototyping of such languages. In this paper, we introduce a novel approach to declarative specification of layout-sensitive languages using layout declarations. Layout declarations are high-level specifications of indentation rules that abstract from low-level technicalities. We show how to derive an efficient layout-sensitive generalized parser and a corresponding pretty-printer automatically from a language specification with layout declarations. We validate our approach in a case-study using a syntax definition for the Haskell programming language, investigating the performance of the generated parser and the correctness of the generated pretty-printer against 22191 Haskell files.

To-many or to-one? all-in-one! efficient purely functional multi-maps with type-heterogeneous hash-tries

An immutable multi-map is a many-to-many map data structure with expected fast insert and lookup operations. This data structure is used for applications processing graphs or many-to-many relations as applied in compilers, runtimes of programming languages, or in static analysis of object-oriented systems. Collection data structures are assumed to carefully balance execution time of operations with memory consumption characteristics and need to scale gracefully from a few elements to multiple gigabytes at least. When processing larger in-memory data sets the overhead of the data structure encoding itself becomes a memory usage bottleneck, dominating the overall performance. In this paper we propose AXIOM, a novel hash-trie data structure that allows for a highly efficient and type-safe multi-map encoding by distinguishing inlined values of singleton sets from nested sets of multi-mappings. AXIOM strictly generalizes over previous hash-trie data structures by supporting the processing of fine-grained type-heterogeneous content on the implementation level (while API and language support for type-heterogeneity are not scope of this paper). We detail the design and optimizations of AXIOM and further compare it against state-of-the-art immutable maps and multi-maps in Java, Scala and Clojure. We isolate key differences using microbenchmarks and validate the resulting conclusions on a case study in static analysis. AXIOM reduces the key-value storage overhead by 1.87x; with specializing and inlining across collection boundaries it improves by 5.1x.

Towards Zero-Overhead Disambiguation of Deep Priority Conflicts

**Context** Context-free grammars are widely used for language prototyping and implementation. They allow formalizing the syntax of domain-specific or general-purpose programming languages concisely and declaratively. However, the natural and concise way of writing a context-free grammar is often ambiguous. Therefore, grammar formalisms support extensions in the form of *declarative disambiguation rules* to specify operator precedence and associativity, solving ambiguities that are caused by the subset of the grammar that corresponds to expressions. **Inquiry** Implementing support for declarative disambiguation within a parser typically comes with one or more of the following limitations in practice: a lack of parsing performance, or a lack of modularity (i.e., disallowing the composition of grammar fragments of potentially different languages). The latter subject is generally addressed by scannerless generalized parsers. We aim to equip scannerless generalized parsers with novel disambiguation methods that are inherently performant, without compromising the concerns of modularity and language composition. **Approach** In this paper, we present a novel low-overhead implementation technique for disambiguating deep associativity and priority conflicts in scannerless generalized parsers with lightweight data-dependency. **Knowledge** Ambiguities with respect to operator precedence and associativity arise from combining the various operators of a language. While *shallow conflicts* can be resolved efficiently by one-level tree patterns, *deep conflicts* require more elaborate techniques, because they can occur arbitrarily nested in a tree. Current state-of-the-art approaches to solving deep priority conflicts come with a severe performance overhead. **Grounding** We evaluated our new approach against state-of-the-art declarative disambiguation mechanisms. By parsing a corpus of popular open-source repositories written in Java and OCaml, we found that our approach yields speedups of up to 1.73x over a grammar rewriting technique when parsing programs with deep priority conflicts--with a modest overhead of 1-2 % when parsing programs without deep conflicts. **Importance** A recent empirical study shows that deep priority conflicts are indeed wide-spread in real-world programs. The study shows that in a corpus of popular OCaml projects on Github, up to 17 % of the source files contain deep priority conflicts. However, there is no solution in the literature that addresses efficient disambiguation of deep priority conflicts, with support for modular and composable syntax definitions.