Using Declarative Specification to Improve the Understanding, Extensibility, and Comparison of Model-Inference Algorithms

Abstract

Abstract-It is a staple development practice to log system behavior. Numerous powerful model-inference algorithms have been proposed to aid developers in log analysis and system understanding. Unfortunately, existing algorithms are typically declared procedurally, making them difficult to understand, extend, and compare. This paper presents InvariMint, an approach to specify modelinference algorithms declaratively. We applied the InvariMint declarative approach to two model-inference algorithms. The evaluation results illustrate that InvariMint (1) leads to new fundamental insights and better understanding of existing algorithms, (2) simplifies creation of new algorithms, including hybrids that combine or extend existing algorithms, and (3) makes it easy to compare and contrast previously published algorithms. InvariMint's declarative approach can outperform procedural implementations. For example, on a log of 50,000 events, InvariMint's declarative implementation of the kTails algorithm completes in 12 seconds, while a procedural implementation completes in 18 minutes. We also found that InvariMint's declarative version of the Synoptic algorithm can be over 170 times faster than the procedural implementation.

Synergizing Specification Miners through Model Fissions and Fusions (T)

“…Notice that the model inferred by traditional 2-tails (in Figure 1) accepts T1, T4, T7, but not T2, T3, T5, T6, and T8. This model is 1 Modal Transition System restricted to accepting traces that contain no repetitions of any methods. On the other hand, the CONTRACTOR++ model (in Figure 2) is less restrictive, as it accepts T1, T2, T4, T7, T8, but not T3, T5, T6.…”

“…Another two templates (i.e., never followed by, and never immediately followed by) were introduced by Beschastnikh et al [1] and have been demonstrated to be useful for describing FSA mining algorithms. The last two templates (i.e., always immediately followed by, and always immediately precedes) are newly introduced in this work.…”

“…In this work, we focus on techniques that extract a model in the form of a finite state automaton (FSA) by analyzing the execution traces of systems or libraries of interest. Krka et al [15], Beschastnikh et al [2], [1], Lo et al. [8], [23], Mariani et al [27], and many others have proposed FSA-based specification mining algorithms.…”

“…Both model fission and model fusion processes are completely automated. In this work, we use a set of 6 constraint templates to generate constraints, some of which were proposed by Dwyer et al [10] and Beschastnikh et al [1]. SpecForge checks whether one or more instances of these constraint templates are observed in a mined model.…”

Inferring Computational State Machine Models from Program Executions

“…Such approaches have traditionally been concerned with the inference of conventional Finite State Machines from traces [7], [8], [9], [10], [11], [12], [13]. Recently, however, several attempts have been made to infer Extended Finite State Machines (EFSMs [14]).…”