User-level Resource-constrained Sandboxing

Fangzhe Chang, Ayal Itzkovitz, and Vijay Karamcheti
Department of Computer Science
Courant Institute of Mathematical Sciences
New York University
{fangzhe, ayali, vijayk}@cs.nyu.edu, https://cs.nyu.edu/pdsg

Abstract:

The popularity of mobile and networked applications has resulted in an increased demand for execution ``sandboxes''--environments that impose irrevocable restrictions on resource usage. Existing approaches rely on kernel modification for enforcing quantitative restrictions (e.g., limiting CPU utilization of an application to 25%). However, the general applicability of such approaches is constrained by the difficulty of modifying shrink-wrapped operating systems such as Windows NT.

 

This paper presents a user-level sandboxing approach for enforcing quantitative restrictions on resource usage of applications. Our approach actively monitors an application's interactions with the underlying system, proactively controlling them to enforce the desired behavior. Our approach leverages a core set of user-level mechanisms that are available in most modern operating systems: fine-grained timers, monitoring infrastructure, debugger processes, priority-based scheduling, and page-based memory protection. We describe implementation of a sandbox on Windows NT that imposes quantitative restrictions on CPU, memory, and network usage. Our results show that application usage of system resources can be restricted to within 3% of desired limits with minimal run-time overhead.