Abstract Implementing complex algorithms for big data, artificial intelligence, and graph processing requires enormous effort. Succinct, declarative programs to solve complex problems that can be efficiently executed for batching and streaming data are in demand. This paper presents Nexus, a distributed Datalog evaluation system. It evaluates Datalog programs using thesemi-naivealgorithm for batch and streaming data using incremental and asynchronous iteration. Furthermore, we evaluate Datalog programs with aggregates to determine the advantages of implementing thesemi-naivealgorithm using incremental iteration on its performance. Our experimental results show that Nexus significantly outperforms acyclic dataflow-based systems.

Modern data analysis tasks often involve control flow statements, such as iterations. Common examples are PageRank and K-means. To achieve scalability, developers usually implement data analysis tasks in distributed dataflow systems, such as Spark and Flink. However, for tasks with control flow statements, these systems still either suffer from poor performance or are hard to use. For example, while Flink supports iterations and Spark provides ease-of-use, Flink is hard to use and Spark has poor performance for iterative tasks. As a result, developers typically have to implement different workarounds to run their jobs with control flow statements in an easy and efficient way.We propose Mitos, a system that achieves the best of both worlds: it achieves both high performance and ease-of-use. Mitos uses an intermediate representation that abstracts away specific control flow statements and is able to represent any imperative control flow. This facilitates building the dataflow graph and coordinating the distributed execution of control flow in a way that is not tied to specific control flow constructs. Our experimental evaluation shows that the performance of Mitos is more than one order of magnitude better than systems that launch new dataflow jobs for every iteration step. Remarkably, it is also up to 10.5 times faster than Flink, which has native iteration support, while matching the ease-of-use of Spark.

Many common data analysis tasks, such as performing hyperparameter optimization, processing a partitioned graph, and treating a matrix as a vector of vectors, offer natural opportunities for nested-parallel operations, i.e., launching parallel operations from inside other parallel operations. However, state-of-the-art dataflow engines, such as Spark and Flink, do not support nested parallelism. Users must implement workarounds, causing orders of magnitude slowdowns for their tasks, let alone the implementation effort.

The strong solutions of Nine Men's Morris and its variant, Lasker Morris, are well-known results (the starting positions are draws). We reexamined both of these games, and calculated extended strong solutions for them. By this, we mean the game-theoretic values of all possible game states that could be reached from certain starting positions where the number of stones to be placed by the players is different from the standard rules. These were also calculated for a previously unsolved third variant, Morabaraba, with interesting results: most of the starting positions where the players can place an equal number of stones (including the standard starting position) are wins for the first player (as opposed to the above games, where these are usually draws). We also developed a multivalued retrograde analysis, and used it as a basis for an algorithm for solving these games ultra-strongly. This means that when our program is playing against a fallible opponent, it has a greater chance of achieving a better result than the game-theoretic value, compared to randomly selecting between “just strongly” optimal moves. Previous attempts on ultrastrong solutions used local heuristics or learning during games, but we incorporated our algorithm into the retrograde analysis.