DETAILED DESCRIPTION OF THE EXEMPLARY EMBODIMENTS

[0023] The present invention supports the assessment of the security risks of a computing network by providing a precise means to calculate and compare the risks posed by security vulnerabilities. Specifically, the present invention allows a security auditing system to identify security vulnerabilities in various elements throughout a network. The security auditing system also can collect information about the function and importance of elements in a computing network. Using this information, the invention calculates a risk value for each security vulnerability that is identified. The risk value can be prioritized based on the ease with which the vulnerability can be repaired. Prioritizing risk values for a particular network element assists a user or network administrator in deciding which vulnerabilities to address first. The invention also supports the calculation of a security score for a network element that accumulates the risk values of each vulnerability associated with the element. For example, employing a band calculation method ensures that a large number of low-risk vulnerabilities does not produce a higher security score than a smaller number of high-risk vulnerabilities. Calculating a security score for each element with the band calculation method allows for a more meaningful comparison of elements across a network.

[0024] Although the exemplary embodiments will be generally described in the context of software modules running in a distributed computing environment, those skilled in the art will recognize that the present invention also can be implemented in conjunction with other program modules for other types of computers. In a distributed computing environment, program modules may be physically located in different local and remote memory storage devices. Execution of the program modules may occur locally in a stand-alone manner or remotely in a client/server manner. Examples of such distributed computing environments include local area networks of an office, enterprise-wide computer networks, and the global Internet.

[0025] The detailed description that follows is represented largely in terms of processes and symbolic representations of operations in a distributed computing environment by conventional computer components, including database servers, application servers, mail servers, routers, security devices, firewalls, clients, workstations, memory storage devices, display devices, and input devices. Each of these conventional distributed computing components is accessible via a communications network, such as a wide area network or local area network.

[0026] The processes and operations performed by the computer include the manipulation of signals by a client or server and the maintenance of these signals within data structures resident in one or more of the local or remote memory storage devices. Such data structures impose a physical organization upon the collection of data stored within a memory storage device and represent specific electrical or magnetic elements. These symbolic representations are the means used by those skilled in the art of computer programming and computer construction to most effectively convey teachings and discoveries to others skilled in the art.

[0027] The present invention also includes a computer program that embodies the functions described herein and illustrated in the appended flow charts. However, it should be apparent that there could be many different ways of implementing the invention in computer programming, and the invention should not be construed as limited to any one set of computer program instructions. Further, a skilled programmer would be able to write such a computer program to implement the disclosed invention based on the flow charts and associated description in the application text, for example. Therefore, disclosure of a particular set of program code instructions is not considered necessary for an adequate understanding of how to make and use the invention. The inventive functionality of the claimed computer program will be explained in more detail in the following description in conjunction with the remaining figures illustrating the program flow.

[0028] Referring now to the drawings, in which like numerals represent like elements throughout the several figures, aspects of the present invention and the preferred operating environment will be described.

[0029] FIG. 1 illustrates various aspects of an exemplary computing environment in which an embodiment of the present invention is designed to operate. Those skilled in the art will appreciate that FIG. 1 and the associated discussion are intended to provide a general description of representative computer network resources in an exemplary distributed computer environment including the inventive security audit system. The architecture comprises a console 105 and a security audit system 115 which are used to configure and schedule security audits of a network 110. The console 105 communicates information about the current security state of the network 110 to a user. The console 105 typically comprises a graphical user interface for presenting and managing data in a convenient format for the user. The console 105 is also operable for receiving information from the security audit system 115 and allowing control of the security audit system 115. The security audit system 115 comprises an active scan engine 120 and one or more other scan engines. In the exemplary embodiment illustrated in FIG. 1, the active scan engine 120 is coupled to an Internet scanning engine 130, a system scanning engine 150, and a database scanning engine 140. Each of these scan engines illustrated in FIG. 1 is coupled to a corresponding database.

[0030] The active scan engine's 120 primary task is acquiring and maintaining current data about the configuration and security posture of the network 110. The active scan engine 120 utilizes the subsidiary scan engines 130, 140 and 150 as a means for gathering information about the network 110. The network 110 typically comprises elements such as desktop computers, routers, and various servers. The active scan engine 120 is responsible for coordinating the configuration, scheduling, and running of scans of these elements found in the network 110. Typically, the active scan engine 120 is continuously running so that the scheduled scans can be run at their designated times, and the resultant data processed in a timely manner.

[0031] Referring to FIG. 2, an overview of an exemplary security auditing process 200 is shown. In alternative embodiments of the present invention, different auditing steps can be performed to collect information about a network that is used to compute a security score. In step 205, an active scan engine 120 configures scans that are to be run on a network 110. In step 210, the active scan engine schedules the times at which the various configured scans will be run on the network 110. An exemplary method for configuring and scheduling scans is described in greater detail in the U.S. non-provisional patent application filed concurrently herewith and referenced herein. When one of the scheduled times for running a scan is reached, the active scan engine 120 will run that scan on the network 110 in step 215. A scan may be run against particular hosts in the network or over the entire network 110. Typically, a scan collects information about the function of the hosts on the network and their respective vulnerabilities. This information is gathered, analyzed, and used to compute a security score for each host or element in the network in step 220. The security scores assist a user or network administrator in determining which vulnerabilities should be addressed first.

[0032] FIG. 3 illustrates an exemplary process for analyzing the results of a security scan, as referenced in step 220, for hosts located on the network 110. In step 305, the active scan engine 120 selects the ID of the first scan for which data are available. In step 310, the active scan engine 120 selects the ID of the first host for which data are available in the current scan job results. In step 315, the active scan engine 120 calculates an asset value for the current host. Asset values are calculated based on the operating system and services associated with the host. An asset value is a characteristic representing the importance of a particular host or element in the operation of the network. In step 320, the active scan engine computes a risk value for each vulnerability detected on the current host. The risk value quantifies the risk a particular vulnerability presents for a network. As will be discussed in connection with FIGS. 7 and 8A, each risk value calculated for a vulnerability is assigned to a band on the risk scale. Using the number of risk values assigned to each band, the active scan engine 120 computes a security score for the current host, in step 325.

[0033] In step 330, the active scan engine 120 determines if there are more hosts in the current job to be processed. If there are more hosts, the ID of the next host is retrieved in step 335, and the process 220 returns to step 315. In step 330, if there are no more hosts to process in the current job, the active scan engine 120 determines if there are more scan jobs to process in step 340. If there are more scan jobs, the ID of the next scan job is retrieved in step 340, and the process returns to step 310. In step 340, if the active scan engine 120 finds that there are no more hosts to process, the analysis is complete.

[0034] FIG. 4 illustrates an exemplary method for calculating an asset value as referred to in step 315 of FIG. 3. Predetermined asset values are generally assigned for the various operating systems, host services, and the vulnerabilities that are likely to be encountered in a scan of a network. These predetermined asset values can be chosen by the provider of the security audit system or the user. The host's operating system and services are identified during the scan performed in step 215. In step 405, the asset value assigned to the host's particular operating system is retrieved. In steps 410 and 415, respective asset values are retrieved for the host's services and relevant vulnerabilities. In the exemplary calculation illustrated herein, the highest of these three asset values is selected in step 420 and used as the asset value for the host in computing the risk values in FIG. 3. In alternative embodiments of the present invention, other processes, such as averaging the asset values, may be used to designate the significance of a particular host in a network 110.

[0035] FIG. 5 illustrates an exemplary method for calculating risk values for the vulnerabilities associated with a host, as referred to in step 320 of FIG. 3. In step 505 the active scan engine 120 selects the first vulnerability discovered on the current host. In step 510, the active scan engine 120 retrieves the asset value for the current host calculated in FIG. 4. The active scan engine 120 retrieves a value representing the probability of a successful exploit against the current vulnerability in step 515. Each vulnerability has one or more exploit methods associated with it. The likelihood of a successful exploit is related to the difficulty of the exploit method. For example, certain vulnerabilities are easily exploited through the use of a simple, publicly available script. An easily exploited vulnerability is assigned a relatively high exploit probability value. On the other hand, other vulnerabilities are very difficult to exploit and involve a complex attack, such as the use of a buffer overflow. Because a buffer overflow is a difficult exploit method to implement, it is assigned a relatively low probability value. To retrieve a vulnerability's exploit probability value, the active scan engine 120 first retrieves the vulnerability's exploit method. The active scan engine 120 then consults a table that maps exploit methods to probability values. As outlined above, these values are assigned on the basis of the difficulty of each method. If a vulnerability has multiple exploit methods, the method with the highest exploit probability is selected.

[0036] In step 520, the active scan engine 120 retrieves a predetermined severity value for the current vulnerability. This value represents the impact of a successful exploit of the vulnerability. In step 525, the risk is calculated as the product of the host asset value and the vulnerability's severity and exploit probability values. Alternative embodiments of the present invention may weigh the three factors differently or use less than all three values in calculating the risk value. In step 530, the active scan engine 120 adjusts the calculated risk value to take into account the difficulty of fixing the current vulnerability. The user can choose not to incorporate the fix difficulty factor and instead, use the “pure” risk value. In step 530, the count of vulnerabilities assigned to the risk band that contains the adjusted risk value is incremented. In step 540, the active scan engine 120 retrieves the next vulnerability detected on the current host. If there is such a vulnerability, the process 320 returns to step 515. Otherwise, the calculation and accumulation of risk values for the host is complete.

[0037] Referring to FIG. 6, an exemplary process is illustrated for adjusting a risk value as referred to in step 530 of FIG. 5. In step 605, the active scan engine 120 retrieves a fix difficulty value for the vulnerability. The fix difficulty value is a measure of the difficulty of remedying a vulnerability. Each vulnerability has associated with it one or more fix methods. To retrieve a fix difficulty value for a vulnerability, the active scan engine 120 first retrieves the vulnerability's fix method. The active scan engine 120 then consults a table that maps each fix method to a difficulty value. This mapping table is typically created by network security personnel for use by the security audit system 115. In step 610, the risk value is multiplied by the fix difficulty value to yield an adjusted risk value. In step 615, the adjusted risk value for the vulnerability can be compared to other adjusted risk values and ranked from highest to lowest. Using this exemplary method, when multiple vulnerabilities have the same computed risk, a user of the security audit system can address those that are easier to fix first. The adjusted risk value can also be used to calculate the security score so that an element's security score will reflect the difficulty with which its vulnerabilities can be fixed.

[0038] FIG. 7 illustrates an exemplary method for assigning a risk value to a risk band, and for incrementing the count of vulnerabilities assigned to the band, as referred to in step 535 of FIG. 5. A risk scale, as described in connection with FIG. 8, typically comprises several bands selected by the user. In step 705, the active scan engine 120 identifies the band i, such that r_i>=r(v)>r_i−1, where r_i denotes the maximum risk value of band i, r_i−1 denotes the maximum risk value of band i−1, and r(v) denotes the risk of the current vulnerability. In step 710, the count C_i of vulnerabilities assigned to band i is incremented by 1.

[0039] FIG. 8A and FIG. 8B illustrate an exemplary method for calculating a security score as referred to in Step 325 of FIG. 3. The illustrated exemplary method employs a logarithmic band calculation so that security scores can be accurately compared over the entire network 110. Using the logarithmic band calculation prevents a host with a large number of low risk values from achieving a higher security score than a host with a few higher risk values. In the illustration shown in FIG. 8A, risk values R₁, R₂, R₃, and R₄ are placed on a scale 800 that is divided into three bands. The number and width of the bands is arbitrary and can be varied to suit the needs and sensitivity of the network. The formula illustrated in FIG. 8A produces a security score that is a function of the band boundary values as opposed to the risk values. Using the exemplary risk values shown on the scale 800, r_i(max) is equal to 0.6, the upper boundary of the highest band containing a risk value. In the exemplary formula, 1$\mathrm{Security}\mathrm{Score}=r_{i\left(\max \right)-1}+\frac{r_{i\left(\max \right)}-r_{i\left(\max \right)-1}}{r_{i\left(\max \right)}+r_{i\left(\max \right)-1}}\sum _{k=1}^{R}\frac{r_{i\left(k\right)}+r_{i\left(k\right)-1}}{2^k}$

[0040] the first term, r_i(max)−1, is 0.2, the lower boundary of the highest band containing a risk value. The remainder of the formula is a fraction of the middle band calculated from the band values encompassing the risk values placed on the scale 800. The result of this calculation is that the security score will not exceed the upper boundary of the highest risk band, in this instance 0.6. The exemplary formula shown in FIG. 8A is merely one way of calculating a security score and alternative embodiments of the present invention may employ other formulas for the calculation.

[0041] Referring to FIG. 8B, an exemplary method for calculating a security score using the band calculation approach is illustrated. This exemplary method shows the calculation of a security score for a particular host in a network. This exemplary method can also be adapted to calculate a security score for a group of hosts within a network. Beginning with step 805, the active scan engine 120 queries for the existence of vulnerabilities for the current host. If there are no vulnerabilities associated with the host, the security score is set to zero in step 810. If there are vulnerabilities, the “yes” branch is followed to step 815, where the active scan engine 120 determines the index of the highest-risk band that has at least one vulnerability's risk value assigned to it. As illustrated in FIG. 8A, the lowest band contains one risk value and the second band contains three risk values. In step 815, the highest risk band which contains at least one risk value is identified as band b_i(max). In the example shown in FIG. 8A, band b_i(max) is the middle band bounded by 0.2 and 0.6 on the risk scale. As noted in FIG. 8A, an embodiment of the current invention implements a simplified version of the band calculation formula. This simplified calculation is accomplished in steps 820 and 825. In step 820, the simplified summation calculation yields preliminary score s. In step 825, the security score is computed by multiplying the summation value by the fraction (r_i(max)−r_i(max)−1)/(r_i(max)+r_i(max)−1), and adding r_i(max)−1, as noted in the formula above. In alternative embodiments of the present invention other methods may be used to sum the risk values to get an accumulated security score.

[0042] FIG. 9 illustrates an exemplary method for determining the highest-risk band in which a risk value is present. In the example shown in FIG. 8A, the second band is band b_i(max) because there are no risk values in the upper region between 0.6 and 1.0. In step 905, the index is set to the highest band risk. In step 910, if the index is equal to b_min, the lowest risk band, the process ends. Otherwise, in step 915 the active scan engine 120 determines if at least one risk value has been assigned to the current band. If so, the current band index is the desired one and the process is complete. Otherwise, the current index is decremented in step 920 and the process 815 returns to step 910.

[0043] FIG. 10 illustrates an exemplary process for summing the banded vulnerability risk values as referenced in step 820. In step 1005, the calculation is initialized by setting the current band index i equal to b_i(max). The preliminary security score s is set to 0.0. The value PreviousCount, which represents the number of risk values that have already been processed, is set to 0. The value StartIndex, which represents the index of the first risk value in the current band, is set to 1. In step 1010, if the current band is band b_min, the process is complete. Otherwise, control passes to step 1015, where a value n is set equal to the count of risks C_i assigned to the current band plus the value PreviousCount. The preliminary score s is incremented by (r_i+r_i−1)*(2^{n−Startndex+1}−1.0)/2ⁿ. The values StarIndex and PreviousCount are then incremented by C_i. The current band index i is then decremented by 1, and the process 820 returns to step 1010.

[0044] In conclusion, the present invention enables and supports security auditing of a distributed computing network by providing a useful numerical value of the risk associated with a vulnerability or group of vulnerabilities. The security audit system can collect information about the elements of a network and compute a risk value for vulnerabilities detected therein. The risk value can be based on the importance of the network element, the likelihood of exploit, and the potential for damage to the network in the event the vulnerability is exploited. The risk value can also be adjusted to reflect the difficulty of remedying the vulnerability. The security audit system can also collect risk values for a particular element and compute a total security score for the element. The security audit system uses a banded calculation method to ensure that a host with several low risk values does not have a higher security score than a host with a few high risk values. Security scores are useful for comparing individual elements or groups of elements on the network.

[0045] It will be appreciated that the present invention fulfills the needs of the prior art described herein and meets the above-stated objects. While there has been shown and described the preferred embodiment of the invention, it will be evident to those skilled in the art that various modifications and changes may be made thereto without departing from the spirit and the scope of the invention as set forth in the appended claims and equivalence thereof. Although the present invention has been described as operating on a local area network, it should be understood that the invention can be applied to other types of distributed computing environments. Furthermore, it should be readily apparent that portions of the calculation can be varied in order modify the results without departing from the scope of the invention.