Biblio

To find security vulnerabilities, many research approaches and commercial tools use a static analysis and check constraints. Previous work compared using a benchmark several approaches where the static analysis and constraints are combined, and the evaluation focused on corner cases in the Java language. We extend the comparative evaluation of these approaches to include one approach that separates the constraints from the static analysis. We also extend the benchmark to cover more classes of security vulnerabilities. Approaches that combine the static analysis and constraints work well for vulnerabilities that are sensitive to the order in which statements are executed. The additional effort required to write separate constraints is rewarded by better recall in dealing with dataflow communication and better precision for callback methods that are common in applications built on frameworks such as Android. 

Self-protecting software systems are a class of autonomic systems capable of detecting and mitigating security threats at runtime. They are growing in importance, as the stovepipe static methods of securing software systems have been shown to be inadequate for the challenges posed by modern software systems. Self-protection, like other self-* properties, allows the system to adapt to the changing environment through autonomic means without much human intervention, and can thereby be responsive, agile, and cost effective. While existing research has made significant progress towards autonomic and adaptive security, gaps and challenges remain. This article presents a significant extension of our preliminary study in this area. In particular, unlike our preliminary study, here we have followed a systematic literature review process, which has broadened the scope of our study and strengthened the validity of our conclusions. By proposing and applying a comprehensive taxonomy to classify and characterize the state-of-the-art research in this area, we have identified key patterns, trends and challenges in the existing approaches, which reveals a number of opportunities that will shape the focus of future research efforts.

In highly configurable systems the configuration space is too big for (re-)certifying every configuration in isolation. In this project, we combine software analysis with network analysis to detect which configuration options interact and which have local effects. Instead of analyzing a system as Linux and SELinux for every combination of configuration settings one by one (>102000 even considering compile-time configurations only), we analyze the effect of each configuration option once for the entire configuration space. The analysis will guide us to designs separating interacting configuration options in a core system and isolating orthogonal and less trusted configuration options from this core. 

Security features are often hardwired into software applications, making it difficult to adapt security responses to reflect changes in runtime context and new attacks. In prior work, we proposed the idea of architecture-based self-protection as a way of separating adaptation logic from application logic and providing a global perspective for reasoning about security adaptations in the context of other business goals. In this paper, we present an approach, based on this idea, for combating denial-of-service (DoS) attacks. Our approach allows DoS-related tactics to be composed into more sophisticated mitigation strategies that encapsulate possible responses to a security problem. Then, utility-based reasoning can be used to consider different business contexts and qualities. We describe how this approach forms the underpinnings of a scientific approach to self-protection, allowing us to reason about how to make the best choice of mitigation at runtime. Moreover, we also show how formal analysis can be used to determine whether the mitigations cover the range of conditions the system is likely to encounter, and the effect of mitigations on other quality attributes of the system. We evaluate the approach using the Rainbow self-adaptive framework and show how Rainbow chooses DoS mitigation tactics that are sensitive to different business contexts.

We present an architecture for the Security Behavior Observatory (SBO), a client-server infrastructure designed to collect a wide array of data on user and computer behavior from hundreds of participants over several years. The SBO infrastructure had to be carefully designed to fulfill several requirements. First, the SBO must scale with the desired length, breadth, and depth of data collection. Second, we must take extraordinary care to ensure the security of the collected data, which will inevitably include intimate participant behavioral data. Third, the SBO must serve our research interests, which will inevitably change as collected data is analyzed and interpreted. This short paper summarizes some of our design and implementation benefits and discusses a few hurdles and trade-offs to consider when designing such a data collection system.

Sandboxes impose a security policy, isolating applications and their components from the rest of a system. While many sandboxing techniques exist, state of the art sandboxes generally perform their functions within the system that is being defended. As a result, when the sandbox fails or is bypassed, the security of the surrounding system can no longer be assured. We experiment with the idea of in-nimbo sandboxing, encapsulating untrusted computations away from the system we are trying to protect. The idea is to delegate computations that may be vulnerable or malicious to virtual machine instances in a cloud computing environment.

This may not reduce the possibility of an in-situ sandbox compromise, but it could significantly reduce the consequences should that possibility be realized. To achieve this advantage, there are additional requirements, including: (1) A regulated channel between the local and cloud environments that supports interaction with the encapsulated application, (2) Performance design that acceptably minimizes latencies in excess of the in-situ baseline.

To test the feasibility of the idea, we built an in-nimbo sandbox for Adobe Reader, an application that historically has been subject to significant attacks. We undertook a prototype deployment with PDF users in a large aerospace firm. In addition to thwarting several examples of existing PDF-based malware, we found that the added increment of latency, perhaps surprisingly, does not overly impair the user experience with respect to performance or usability.

Information system developers and administrators often overlook critical security requirements and best practices. This may be due to lack of tools and techniques that allow practitioners to tailor security knowledge to their particular context. In order to explore the impact of new security methods, we must improve our ability to study the impact of security tools and methods on software and system development. In this paper, we present early findings of an experiment to assess the extent to which the number and type of examples used in security training stimuli can impact security problem solving. To motivate this research, we formulate hypotheses from analogical transfer theory in psychology. The independent variables include number of problem surfaces and schemas, and the dependent variable is the answer accuracy. Our study results do not show a statistically significant difference in performance when the number and types of examples are varied. We discuss the limitations, threats to validity and opportunities for future studies in this area.

In highly configurable systems the configuration space is too big for (re-)certifying every configuration in isolation. In this project, we combine software analysis with network analysis to detect which configuration options interact and which have local effects. Instead of analyzing a system as Linux and SELinux for every combination of configuration settings one by one (>102000 even considering compile-time configurations only), we analyze the effect of each configuration option once for the entire configuration space. The analysis will guide us to designs separating interacting configuration options in a core system and isolating orthogonal and less trusted configuration options from this core. 

Security has become the Achilles’ heel of most modern software systems. Techniques ranging from the manual inspection to automated static and dynamic analyses are commonly employed to identify security vulnerabilities prior to the release of the software. However, these techniques are time consuming and cannot keep up with the complexity of ever-growing software repositories (e.g., Google Play and Apple App Store). In this paper, we aim to improve the status quo and increase the efficiency of static analysis by mining relevant information from vulnerabilities found in the categorized software repositories. The approach relies on the fact that many modern software systems are developed using rich application development frameworks (ADF), allowing us to raise the level of abstraction for detecting vulnerabilities and thereby making it possible to classify the types of vulnerabilities that are encountered in a given category of application. We used open-source software repositories comprising more than 7 million lines of code to demonstrate how our approach can improve the efficiency of static analysis, and in turn, vulnerability detection.

In many scientific fields, simulations and analyses require compositions of computational entities such as web-services, programs, and applications. In such fields, users may want various trade-offs between different qualities. Examples include: (i) performing a quick approximation vs. an accurate, but slower, experiment, (ii) using local slower execution environments vs. remote, but advanced, computing facilities, (iii) using quicker approximation algorithms vs. computationally expensive algorithms with smaller data. However, such trade-offs are difficult to make as many such decisions today are either (a) wired into a fixed configuration and cannot be changed, or (b) require detailed systems knowledge and experimentation to determine what configuration to use. In this paper we propose an approach that uses architectural models coupled with automated design space generation for making fidelity and timeliness trade-offs. We illustrate this approach through an example in the intelligence analysis domain.

Project documentation often contains security-relevant statements that are indicative of the security requirements of a system. However these statements may not be explicitly specified or straightforward to locate. At best, requirements analysts manually extract applicable security requirements from project documents. However, security requirements that are not explicitly stated may not be considered during implementation. The goal of this research is to aid requirements analysts in generating security requirements through identifying securityrelevant statements in project documentation and providing context-specific templates to generate security requirements. First, we identify the most prevalent security objectives from software security literature. To identify security-relevant statements in project documentation, we propose a tool-based process to classify statements as related to zero or more security objectives. We then develop a set of context-specific templates to help translate the security objectives of each statement into explicit sets of security functional requirements. We evaluate our process on six documents from the electronic healthcare software industry, identifying 46% of statements as implicitly or explicitly related to security. Our classification approach identified security objectives with a precision of .82 and recall of .79. From our total set of classified statements, we extracted 16 context-specific templates that identify 41 reusable security requirements.

As new market opportunities, technologies, platforms, and frameworks become available, systems require large-scale and systematic architectural restructuring to accommodate them. Today's architects have few techniques to help them plan this architecture evolution. In particular, they have little assistance in planning alternative evolution paths, trading off various aspects of the different paths, or knowing best practices for particular domains. In this paper, we describe an approach for planning and reasoning about architecture evolution. Our approach focuses on providing architects with the means to model prospective evolution paths and supporting analysis to select among these candidate paths. To demonstrate the usefulness of our approach, we show how it can be applied to an actual architecture evolution. In addition, we present some theoretical results about our evolution path constraint specification language.

A self-adaptive software system should be able to monitor and analyze its runtime behavior and make adaptation decisions accordingly to meet certain desirable objectives. Traditional software adaptation techniques and recent “models@runtime” approaches usually require an a priori model for a system’s dynamic behavior. Oftentimes the model is difficult to define and labor-intensive to maintain, and tends to get out of date due to adaptation and architecture decay. We propose an alternative approach that does not require defining the system’s behavior model beforehand, but instead involves mining software component interactions from system execution traces to build a probabilistic usage model, which is in turn used to analyze, plan, and execute adaptations. Our preliminary evaluation of the approach against an Emergency Deployment System shows that the associations mining model can be used to effectively address a variety of adaptation needs, including (1) safely applying dynamic changes to a running software system without creating inconsistencies, (2) identifying potentially malicious (abnormal) behavior for self-protection, and (3) our ongoing research on improving deployment of software components in a distributed setting for performance self-optimization.

Availability is an increasingly important quality for today's software-based systems and it has been successfully addressed by the use of closed-loop control systems in self-adaptive systems. Probes are inserted into a running system to obtain information and the information is fed to a controller that, through provided interfaces, acts on the system to alter its behavior. When a failure is detected, pinpointing the source of the failure is a critical step for a repair action. However, information obtained from a running system is commonly incomplete due to probing costs or unavailability of probes. In this paper we address the problem of fault localization in the presence of incomplete system monitoring. We may not be able to directly observe a component but we may be able to infer its health state. We provide formal criteria to determine when health states of unobservable components can be inferred and establish formal theoretical bounds for accuracy when using any spectrum-based fault localization algorithm.

Concurrent programs are prone to various classes of difficult-to-detect faults, of which data races are particularly prevalent. Prior work has attempted to increase the cost-effectiveness of approaches for testing for data races by employing race detection techniques, but to date, no work has considered cost-effective approaches for re-testing for races as programs evolve. In this paper we present SimRT, an automated regression testing framework for use in detecting races introduced by code modifications. SimRT employs a regression test selection technique, focused on sets of program elements related to race detection, to reduce the number of test cases that must be run on a changed program to detect races that occur due to code modifications, and it employs a test case prioritization technique to improve the rate at which such races are detected. Our empirical study of SimRT reveals that it is more efficient and effective for revealing races than other approaches, and that its constituent test selection and prioritization components each contribute to its performance.

Research shows that commonly accepted security requirements   are  not  generally  applied  in  practice.  Instead  of relying on requirements checklists, security experts rely on their expertise and background knowledge to identify security vulnerabilities.  To  understand  the  gap  between  available checklists  and  practice,  we  conducted  a  series  of  interviews  to encode   the   decision-making   process   of  security   experts   and novices during security requirements analysis. Participants were asked to analyze two types of artifacts: source code, and network diagrams  for  vulnerabilities  and  to  apply  a  requirements checklist to mitigate some of those vulnerabilities.  We framed our study using Situation Awareness—a cognitive theory from psychology—to   elicit  responses   that  we  later  analyzed   using coding theory and grounded analysis.  We report our preliminary results of analyzing two interviews that reveal possible decision- making patterns that could characterize how analysts perceive, comprehend   and  project  future  threats  which  leads  them  to decide upon requirements  and their specifications,  in addition, to how  experts  use  assumptions  to  overcome  ambiguity  in specifications.  Our goal is to build a model that researchers  can use to evaluate their security requirements methods against how experts transition through different situation awareness levels in their decision-making  process.

Both SAT and #SAT can represent difficult problems in seemingly dissimilar areas such as planning, verification,  and probabilistic  inference. Here, we examine an expressive new language, #∃SAT, that generalizes both of these languages.   #∃SAT problems require counting the number of satisfiable formulas in a concisely-describable  set of existentially quantified, propositional formulas. We characterize the expressiveness and worst-case difficulty of #∃SAT by proving it is complete for the complexity  class #P NP [1], and re- lating this class to more familiar complexity  classes. We also experiment with three new

general-purpose #∃SAT solvers on a battery  of problem distributions  including  a simple logistics domain. Our experiments show that, despite the formidable worst-case complex-

ity of #P NP [1], many of the instances can be solved efficiently  by noticing and exploiting a particular type of frequent structure.