No abstract

Abstract-Enterprise networks today host a wide variety of network services, which often depend on each other to provide and support network-based services and applications. Understanding such dependencies is essential for maintaining the well-being of an enterprise network and its applications, particularly in the presence of network attacks and failures. In a typical enterprise network, which is complex and dynamic in configuration, it is non-trivial to identify all these services and their dependencies. Several techniques have been developed to learn such dependencies automatically. However, they are either too complex to fine tune or cluttered with false positives and/or false negatives.In this paper, we propose a suite of novel techniques and develop a new tool named NSDMiner (which stands for Mining for Network Service Dependencies) to automatically discover the dependencies between network services from passively collected network traffic. NSDMiner is non-intrusive; it does not require any modification of existing software, or injection of network packets. More importantly, NSDMiner achieves higher accuracy than previous network-based approaches. Our experimental evaluation, which uses network traffic collected from our campus network, shows that NSDMiner outperforms the two best existing solutions significantly.

No abstract

Cyber attacks endanger physical, economic, social, and political security. There have been extensive efforts in government, academia, and industry to anticipate, forecast, and mitigate such cyber attacks. A common approach is time-series forecasting of cyber attacks based on data from network telescopes, honeypots, and automated intrusion detection/prevention systems. This research has uncovered key insights such as systematicity in cyber attacks. Here, we propose an alternate perspective of this problem by performing forecasting of attacks that are analyst-detected and -verified occurrences of malware. We call these instances of malware cyber event data. Specifically, our dataset was analyst-detected incidents from a large operational Computer Security Service Provider (CSSP) for the U.S. Department of Defense, which rarely relies only on automated systems. Our data set consists of weekly counts of cyber events over approximately seven years. This curated dataset has characteristics that distinguish it from most datasets used in prior research on cyber attacks. Since all cyber events were validated by analysts, our dataset is unlikely to have false positives which are often endemic in other sources of data. Further, the higher-quality data could be used for a number of important tasks for CSSPs such as resource allocation, estimation of security resources, and the development of effective risk-management strategies. To quantify bursts, we used a Markov model of state transitions. For forecasting, we used a Bayesian State Space Model and found that events one week ahead could be predicted with reasonable accuracy, with the exception of bursts. Our findings of systematicity in analyst-detected cyber attacks are consistent with previous work using cyber attack data from other sources. The advanced information provided by a forecast may help with threat awareness by providing a probable value and range for future cyber events one week ahead, similar to a weather forecast. Other potential applications for cyber event forecasting include proactive allocation of resources and capabilities for cyber defense (e.g., analyst staffing and sensor configuration) in CSSPs. Enhanced threat awareness may improve cybersecurity by helping to optimize human and technical capabilities for cyber defense.In contrast, nearly all prior research on modeling cyber attacks [4-10] lacked analyst detection and verification of computer security incidents with the exceptions of [11,12]. In these two exceptions, security incidents were verified by system administrators at a large university [11] or verified by analysts at a CSSP [12]. Thus in most earlier research, the sources for cyber attacks were processed data from network telescopes and honeypots [4,6,[8][9][10] and alerts from automated systems on real networks [5,7,13]. Compared to real networks, the majority of traffic to network telescopes (passive monitoring of unrequested network traffic to unused IP addresses) and honeypots (monitored and isolated systems that are designed to appear...