Scalable Processing of Contemporary Semi-Structured Data on Commodity Parallel Processors - A Compilation-based Approach

Abstract

JSON (JavaScript Object Notation) and its derivatives are essential in the modern computing infrastructure. However, existing software often fails to process such types of data in a scalable way, mainly for two reasons: (i) the processing often requires to build a memory-consuming parse tree; (ii) there exist inherent dependences in processing the data stream, preventing any data-level parallelization. Facing the challenges, developers often have to construct ad-hoc pre-parsers to split the data stream in order to reduce the memory consumption and increase the data parallelism. However, this strategy requires more programming efforts. Moreover, the pre-parsing itself is non-trivial to parallelize, thus introducing a new serial bottleneck. To solve the dilemma, this work introduces a scalable yet fully automatic solution-a compilation system, namely JPStream, that compiles standard JSONPath queries into parallel executables with bounded memory footprints. First, JPStream adopts a stream processing design that combines the querying and parsing into one pass, without generating any in-memory parse tree. To achieve this, JPStream uses a novel joint compilation technique that compiles the queries and the JSON syntax together into a single automaton. Furthermore, JPStream leverages the "enumerability" of automaton to break the dependences and reason about the transition rules to prune infeasible cases. It also features a module that learns data constraints from the input data to enhance the pruning. Evaluation on real-world JSON datasets with standard JSONPath queries shows that JPStream can reduce the memory consumption significantly, by up to 95%, meanwhile achieving near-linear speedup on multicore and manycore processors.

Challenging Sequential Bitstream Processing via Principled Bitwise Speculation

“…In addition, there are also proposals of programming language constructs [35], speculation design for irregular applications [34], as well as compilers with speculative execution supports [38]. Some recent works also target speculative parallelization of semi-structured data stream processing, but at the byte level rather than bit level, like speculative HTML parsing [51] and speculative path query processing of XML/JSON streams [19,20].…”

CENTAURUS: A Dynamic Parser Generator for Parallel Ad Hoc Data Extraction

“…XML streaming [10], [14], [17] is an instance on XML processing. In particular, it has recently been actively investigated for JSON [9], [15], [18], [19], [25]. Although these studies dealt with JSON parsing, their problem settings differed depending on concerned data models and queries.…”

“…Grammar-based enhancement of parallel processing [9], [10] aimed to compile queries with grammatical constraints. Our aim with Centaurus is grammar-based programming.…”

JSONSki: streaming semi-structured data with bit-parallel fast-forwarding

“…To demonstrate its efficiency, we compared JSONSki with several existing JSON tools using real-world datasets and standard path queries. The results have shown that JSONSki outperforms JPStream [35], a state-of-the-art streaming library, and simdjson [40], a popular SIMD-based parser substantially, in both the large and small record processing scenarios. • First, it systematically reveals the fast-forward opportunities in the semi-structured data streaming (Section 3).…”

“…• Streaming scheme can naturally avoid the above issues by immediately consuming the semi-structured data stream, locating and extracting the data of interests on the fly. For example, JsonSurfer [13] and JPStream [35] evaluate path queries dynamically as they traverse the data stream without generating any in-memory trees. Thus, they only need to scan the data in one pass with a small memory footprint.…”