Parametric inference of memory requirements for garbage collected languages

Abstract

The accurate prediction of program's memory requirements is a critical component in software development. Existing heap space analyses either do not take deallocation into account or adopt specific models of garbage collectors which do not necessarily correspond to the actual memory usage. We present a novel approach to inferring upper bounds on memory requirements of Java-like programs which is parametric on the notion of object lifetime, i.e., on when objects become collectible. If objects lifetimes are inferred by a reachability analysis, then our analysis infers accurate upper bounds on the memory consumption for a reachability-based garbage collector. Interestingly, if objects lifetimes are inferred by a heap liveness analysis, then we approximate the program minimal memory requirement, i.e., the peak memory usage when using an optimal garbage collector which frees objects as soon as they become dead. The key idea is to integrate information on objects lifetimes into the process of generating the recurrence equations which capture the memory usage at the different program states. If the heap size limit is set to the memory requirement inferred by our analysis, it is ensured that execution will not exceed the memory limit with the only assumption that garbage collection works when the limit is reached. Experiments on Java bytecode programs provide evidence of the feasibility and accuracy of our analysis.

A multi-domain incremental analysis engine and its application to incremental resource analysis

“…The modifications (described in [6] in detail) are performed in order to overcome some limitations inherent to the size analysis and the UB solver of costa and are not related to the incremental algorithm presented in this paper. Furthermore, we have used as benchmarks the following programs borrowed from the Apache-Commons Project [28]: StringEncrypt and ParseTarHeader from Apache-Commons-Compress, and TestOrthogonal and TestDistance from Apache-Commons-Math.…”

Making resource analysis practical for real-time Java

“…COSTA implements different garbage collection models: a user can select 'scope', 'reachability', 'liveness', or none at all [4]. This functionality is retained in ResAna.…”

Live Heap Space Bounds for Real-Time Systems

“…Solving the ILP problem of the example in Appendix A took about 25 seconds, using lp_solve version 5.5.2.0. 3 …”

“…As already mentioned, for each task we borrow from [2,3], the rule-based representation of programs along with semantics, which we could however simplify due to special treatment of state variables and lack of mutation. We also adopt from their work the step of inferring size relations.…”

“…Recent years have seen a respectable development in techniques for analysis of live heap space usage of programs [3,14,9,18]. The common goal of this line of research is to obtain an a priori upper bound on the size of heap memory reachable from various points in a program, expressed as a function of its input data.…”