Abstract. Effective risk management is a key to ensure that our nuclear power plants, medical equipment, and power grids are dependable; and it is often required by law. Fault Tree Analysis (FTA) is a widely used methodology here, computing important dependability measures like system reliability. This paper presents DFTCalc, a powerful tool for FTA, providing (1) efficient fault tree modelling via compact representations; (2) effective analysis, allowing a wide range of dependability properties to be analysed (3) efficient analysis, via state-of-the-art stochastic techniques; and (4) a flexible and extensible framework, where gates can easily be changed or added. Technically, DFTCalc is realised via stochastic model checking, an innovative technique offering a wide plethora of powerful analysis techniques, including aggressive compression techniques to keep the underlying state space small.

Stateful model checking creates numerous states which need to be stored and checked if already visited. One option for such storage is a hash map and this has been used in many model checkers. In particular, we are interested in the performance of concurrent hash maps for use in multi-core model checkers with a variable state vector size. Previous research claimed that open addressing was the best performing method for the parallel speedup of concurrent hash maps. However, here we demonstrate that chaining lends itself perfectly for use in a concurrent setting.We implemented 12 hash map variants, all aiming at multicore efficiency. 8 of our implementations support variable-length key-value pairs. We compare our implementations and 22 other hash maps by means of an extensive test suite. Of these 34 hash maps, we show the representative performance of 11 hash maps.Our implementations not only support state vectors of variable length, but also feature superior scalability compared with competing hash maps. Our benchmarks show that on 96 cores, our best hash map is between 1.3 and 2.6 times faster than competing hash maps, for a load factor under 1. For higher load factors, it is an order of magnitude faster.

High-performance software typically uses dynamic memory allocations and multi-threading to leverage multi-core CPUs. Model checking such software not only has to deal with state space explosion, but also with variable-length states due to dynamic allocations. Moreover, changes between states are typically small, calling for incremental updates. Many model checkers, although efficiently dealing with the latter, only support fixed-length state vectors. In this paper, we introduce dtree, a concurrent compression tree data structure that compactly stores variable-length states while allowing partial state reconstruction and incremental updates without reconstructing states. We implemented dtree in the dmc multi-core model checker. We show that, for models with states of varying length, dtree is up to 2.9 times faster and uses on average 29% less memory than state-of-the-art tools.

No abstract