Exploit detection system

Abstract

According to one embodiment, a virtualized malware detection system is integrated with a virtual machine host including a plurality of virtual machines and a security virtual machine. Logic within the virtual machines are configured to perform a dynamic analysis of an object and monitor for the occurrence of a triggering event. Upon detection of a triggering event within a virtual machine, the logic within the virtual machine provides the security virtual machine with information associated with the triggering event for further analysis. Based on the further analysis, the object may then be classified as “non-malicious,” or “malicious.”

Description

FIELD

Embodiments of the disclosure relate to the field of cyber security. More specifically, embodiments of the disclosure relate to a system for detecting anomalous, or more specifically, malicious behavior through successive intercommunications between virtual machines.

GENERAL BACKGROUND

Over the last decade, malicious software has become a pervasive problem for Internet users as many networked resources include vulnerabilities that are subject to attack. For instance, over the past few years, more and more vulnerabilities are being discovered in software that is loaded onto endpoint devices present on the network. These vulnerabilities may be exploited by allowing a third-party, e.g., through computer software, to gain access to one or more areas within the network not typically accessible. For example, a third-party may exploit a vulnerability to gain unauthorized access to email accounts and/or data files.

While some vulnerabilities continue to be addressed through software patches, prior to the release of such software patches, network devices will continue to be targeted for attack by exploits, namely malicious computer code that attempts to acquire sensitive information, adversely influence, or attack normal operations of the network device or the entire enterprise network by taking advantage of a vulnerability in computer software.

Currently, a datacenter may employ a plurality of virtual machines to simulate one or more endpoint devices in order to detect attacks by exploits and/or malware. However, the performance of advanced security measures on each virtual machine within the datacenter may lead to duplication of exploit detection efforts. This duplication of exploit detection efforts results in increased overhead of the plurality of virtual machines.

Currently, a virtual machine may perform a dynamic analysis of an object, e.g., one or more files included within received network traffic and/or files stored in a local or external storage device. In addition, each of a plurality of virtual machines may be configured with different software images thereby simulating various network device deployments. In some instances, a virtual machine may include a software configuration to simulate a specific network device within a particular enterprise network, e.g., configured with one or more specific applications to perform execution of particular objects.

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the invention are illustrated by way of example and not by way of limitation in the figures of the accompanying drawings, in which like references indicate similar elements and in which:

FIG. 1 is a block diagram of an exemplary virtualized malware detection system.

FIG. 2 is a block diagram of an exemplary virtual machine host of a virtualized malware detection system communicatively coupled to a network security appliance, an endpoint security appliance and cloud services.

FIG. 3 is an in-depth block diagram of an exemplary virtual machine host of a virtualized malware detection system communicatively coupled to a network security appliance, an endpoint security appliance and cloud services.

FIG. 4A is a flowchart illustrating a first exemplary method of detecting malware and/or exploits using a virtualized malware detection system.

FIG. 4B is a flowchart illustrating a second exemplary method of detecting malware and/or exploits using a virtualized malware detection system.

FIGS. 5A and 5B are flowcharts illustrating an exemplary method for detecting malware and/or exploits associated with a uniform resource locator (URL) opened within a browser using a virtualized malware detection system.

FIG. 6 is an exemplary block diagram of logic associated with the detection system 110₁ of FIG. 1.

DETAILED DESCRIPTION

Various embodiments of the disclosure relate to a virtualized malware detection system that improves exploit detection and/or visual representation of the detection of the suspected exploit and/or malware. The virtualized malware detection system may include one or more virtual machine (VM) hosts, a controller, and optionally, a storage device. Each VM host may include one or more VMs, which perform dynamic analysis of an object, a “security” VM, which performs further analysis for each of the VMs within the VM host, and a hypervisor, which manages communications between the VMs and the security VM. It is contemplated that the security VM may be a secondary VM or another type of logic that runs in a virtual environment.

In one embodiment, the virtualized malware detection system may be deployed in a datacenter wherein the one or more VM hosts that include one or more VMs, a security VM and a hypervisor provide for savings on hardware costs. In one such embodiment, the VM hosts may be segmented (e.g., partitioned) according to enterprise and/or individual user (e.g., a solo entrepreneur) wherein each of the VMs and the security VM within a VM host are utilized by, for example, a single enterprise or solo entrepreneur. In a second embodiment, a datacenter deployment may include hardware comprising, at least, a first VM host including 500 VMs, a security VM and a hypervisor, and a second VM host including 1,000 VMs, a security VM and a hypervisor. In such an embodiment, a first enterprise may utilize the first VM including all 500 VMs and the security VM included therein, while a second enterprise may utilize the second VM including all 1,000 VMs and the security VM included therein.

In an alternative embodiment, the VMs within a VM host may be segmented according to enterprise and/or solo entrepreneur. For example, a datacenter deployment may include hardware comprising a VM host including 1,000 VMs, a security VM and a hypervisor. A first enterprise may utilize 900 VMs and a solo entrepreneur may utilize 100 VMs wherein all 1,000 VMs utilize the same security VM. The communication between the VMs and the security VM is coordinated by the hypervisor; therefore, data of the first enterprise and data of the entrepreneur will not be commingled.

In one embodiment of the disclosure, the virtualized malware detection system determines whether an object is associated with a malicious attack through a dynamic analysis of the object within a virtual run-time environment. The virtual run-time environment features one or more VMs, which may be provisioned with a guest image associated with a prescribed software profile. Each guest image may include a software application and/or an operating system (OS). The guest image may further include one or more monitors, namely software components that are configured to observe and capture run-time behavior of an object under analysis during processing within the virtual machine. In another embodiment, the virtualized malware detection system may perform pre-processing, which may include signature checks and/or comparisons of information associated with the object with content within the one or more entries of a blacklist and/or a whitelist. Herein, the term “signature” may designate an indicator of a set of characteristics and/or behaviors exhibited by one or more exploits that may not be unique to those exploit(s). Thus, a match of the signature may indicate to some level of probability that a portion of received data constitutes an exploit. Subsequently, depending on the results of the pre-processing, the virtualized malware detection system may conduct dynamic analysis of the object.

In one embodiment, each of the VMs within a VM host may be configured with a guest image to simulate a particular endpoint device which may be based on the specific needs of a corporation, for example. Specifically, each VM may be configured with different operating systems, different applications, different versions of a common operating system and/or different versions of a common application. In one embodiment, the virtualized malware detection system may receive an object via a network connection based on an action by a user. For example, a user may remotely control actions on a VM within the VM host such that the user may download an object from an Internet browsing application and launch the object. Upon the user launching the object remotely in the VM, the VM within the VM host performs an analysis to determine whether the object is associated with a malicious attack. Specifically, the VM may provide the security VM with information associated with the object in order to enable the security VM to perform pre-processing thereon. When the pre-processing does not result in a determination that the object is non-malicious (e.g., the object matches an entry on a whitelist), the VM continues a dynamic analysis of the object within the virtual run-time environment. Herein, the term “launch” (and other tenses) represents performance of one or more events that initiates activation of an object under analysis. When the VM detects the occurrence of a triggering event, the VM provides information associated with the triggering event to the security VM for further analysis. The triggering event may be an event that, through experiential knowledge and/or machine learning techniques, has been determined to have an association with a malicious attack but signifies further analysis should be performed.

In one embodiment, as the security VM is performing further analysis on the information associated with the triggering event, the VM continues the dynamic analysis. Furthermore, the VM may provide the security VM with post-processing information associated with the triggering event throughout the dynamic analysis. In a second embodiment, the VM may pause processing until the security VM requests additional information associated with processing of the object subsequent to the occurrence of the triggering event. Therefore, duplication of work performed on each of the VMs is avoided by off-loading the further analysis of information associated with the occurrence of a triggering event to a security VM.

Embodiments of the invention may be employed by or take the form of a server device or apparatus implementing a virtualized malware detection system, where the virtualized malware detection system includes a plurality of VM hosts, which each include a plurality of VMs and a security VM. According to one embodiment of the disclosure, an endpoint device, management system or cloud computing service transmits an object to the virtualized malware detection system for analysis, wherein a user remotely launches the object within a VM of the detection system 110₁. Depending on the analysis by the virtualized malware detection system, the object may be labeled as either “non-malicious” or “malicious.” Throughout the specification, claims and figures, the term “network traffic” will be used in the discussion but any form of incoming data may be substituted.

I. Terminology

In the following description, certain terminology is used to describe features of the invention. For example, in certain situations, both terms “logic” and “engine” are representative of hardware, firmware and/or software that is configured to perform one or more functions. As hardware, logic (or engine) may include circuitry having data processing or storage functionality. Examples of such circuitry may include, but are not limited or restricted to a microprocessor, one or more processor cores, a programmable gate array, a microcontroller, a controller, an application specific integrated circuit, wireless receiver, transmitter and/or transceiver circuitry, semiconductor memory, or combinatorial logic.

Logic (or engine) may be software in the form of one or more software modules, such as executable code in the form of an executable application, an application programming interface (API), a subroutine, a function, a procedure, an applet, a servlet, a routine, source code, object code, a shared library/dynamic load library, or one or more instructions. These software modules may be stored in any type of a suitable non-transitory storage medium, or transitory storage medium (e.g., electrical, optical, acoustical or other form of propagated signals such as carrier waves, infrared signals, or digital signals). Examples of non-transitory storage medium may include, but are not limited or restricted to a programmable circuit; a semiconductor memory; non-persistent storage such as volatile memory (e.g., any type of random access memory “RAM”); persistent storage such as non-volatile memory (e.g., read-only memory “ROM”, power-backed RAM, flash memory, phase-change memory, etc.), a solid-state drive, hard disk drive, an optical disc drive, or a portable memory device. As firmware, the executable code is stored in persistent storage.

An “exploit” may be construed broadly as information (e.g., executable code, data, command(s), etc.) that attempts to take advantage of a vulnerability in software and/or an action by a person gaining unauthorized access to one or more areas of a network, a computer and/or an electronic device. For instance, a “vulnerability” is a coding error or artifact of software (e.g., computer program) that allows an attacker to alter legitimate control flow during processing of the software (computer program) by a network device, and thus, causes the network device to experience undesirable or anomalous behaviors. The undesired or anomalous behaviors may include a communication-based anomaly or an execution-based anomaly, which, for example, could (1) alter the functionality of an network device executing application software in an atypical manner (a file is opened by a first process where the file is configured to be opened by a second process and not the first process); (2) alter the functionality of the network device executing that application software without any malicious intent; and/or (3) provide unwanted functionality which may be generally acceptable in another context. As an illustrative example, a computer program may be considered as a state machine, where all valid states (and transitions between states) are managed and defined by the program, in which case an exploit may be viewed as seeking to alter one or more of the states (or transitions) from those defined by the program. The term “anomalous behavior” should be understood to include either (i) a first event that is an atypical occurrence or a malicious occurrence, or (ii) a relationship identifying that the first event is based on a second event, the relationship being an atypical relationship between the first and second event or a relationship between the first and second events that is malicious to the network, electronic device on which the relationship appears, or to one or more users of the electronic device or of the network.

According to one embodiment, the term “malware” may be construed broadly as computer code that executes an exploit to take advantage of a vulnerability, for example, to harm or co-opt operation of a network device or misappropriate, modify or delete data. Conventionally, malware is often said to be designed with malicious intent.

The term “transmission medium” is a physical or logical communication path between two or more network devices (e.g., any devices with data processing and network connectivity such as, for example, a security appliance, a server, a mainframe, a computer such as a desktop or laptop, netbook, tablet, firewall, smart phone, router, switch, bridge, etc.). For instance, the communication path may include wired and/or wireless segments. Examples of wired and/or wireless segments include electrical wiring, optical fiber, cable, bus trace, or a wireless channel using infrared, radio frequency (RF), or any other wired/wireless signaling mechanism.

The term “object” generally refers to a collection of data, whether in transit (e.g., over a network) or at rest (e.g., stored), often having a logical structure or organization that enables it to be classified for purposes of analysis. During analysis, for example, the object may exhibit a set of expected characteristics and, during processing, a set of expected behaviors. The object may also exhibit a set of unexpected characteristics and a set of unexpected behaviors that may evidence an exploit and potentially allow the object to be classified as an exploit.

Examples of objects may include one or more flows or a self-contained element within a flow itself. A “flow” generally refers to related packets that are received, transmitted, or exchanged within a communication session. For convenience, a packet is broadly referred to as a series of bits or bytes having a prescribed format, which may include packets, frames, or cells.

As an illustrative example, an “object” may include a set of flows such as a sequence of transmissions in accordance with a particular communication protocol (e.g., Transmission Control Protocol (TCP), Hypertext Transfer Protocol (HTTP), etc.), or inter-process communications (e.g., Remote Procedure Call “RPC” or analogous processes, etc.). Similar, as another illustrative example, the object may be a self-contained element, where different types of such objects may include an executable file, non-executable file (such as a document or a dynamically link library), a Portable Document Format (PDF) file, a JavaScript file, Zip file, a Flash file, a document (for example, a Microsoft Office® document), an electronic mail (email), downloaded web page, an instant messaging element in accordance with Session Initiation Protocol (SIP) or another messaging protocol, or the like.

The terms “potentially suspicious,” “suspicious” and “malicious” may each be represent a probability (or level of confidence) that the object is associated with a malicious attack. For instance, the probability may be based, at least in part, on (i) pattern matches; (ii) analyzed deviations in messaging practices set forth in applicable communication protocols e.g., HTTP, TCP, etc.); (iii) analyzed compliance with certain message formats established for the protocol (e.g., out-of-order commands); (iv) analyzed header or payload parameters to determine compliance, (v) attempts to communicate with external servers during dynamic processing, and/or (vi) attempts to access predetermined (e.g., secure) locations in memory during dynamic processing.

Lastly, the terms “or” and “and/or” as used herein are to be interpreted as inclusive or meaning any one or any combination. Therefore, “A, B or C” or “A, B and/or C” mean “any of the following: A; B; C; A and B; A and C; B and C; A, B and C.” An exception to this definition will occur only when a combination of elements, functions, steps or acts are in some way inherently mutually exclusive.

The invention may be utilized for displaying an interactive infection visualization detailing detection, verification and/or prioritization of malicious content. As this invention is susceptible to embodiments of many different forms, it is intended that the present disclosure is to be considered as an example of the principles of the invention and not intended to limit the invention to the specific embodiments shown and described.

II. Virtualized Malware Detection System

Referring to FIG. 1, a block diagram of a malware and/or exploit detection system 100 deploying a plurality of exemplary virtualized malware detection systems (referred to hereinafter as “detection systems”) 110₁-110_D (D≥1, e.g., D=3) communicatively coupled to a management system 120 via a network 140 is shown. The malware and/or exploit detection system 100 may be deployed within an enterprise network, where the detection system 110₁ is communicatively coupled for network connectivity, e.g., to communicate with one or more client devices 190₁-190_C.

Additionally, the detection system 110₁ may be communicatively coupled to the management system 120 and one or more detection systems 110₂-110₃ via the network 140. In general, the management system 120 may be adapted to manage detection systems 110₁-110₃, in part, through propagation of updates of, for example, a whitelist, a blacklist, correlation rules, guest images for the VMs, and/or VM hosts. For example, updates may include, but are not limited or restricted to, modifying a current entry in the whitelist or the blacklist, modifying a current correlation rule, adding a new entry to the whitelist or the blacklist, adding a new correlation rule, removing a current entry in the whitelist or the blacklist, and/or removing a current correlation rule. Also, the detection system 110₁ may be communicatively coupled to a network security appliance 150 and/or an endpoint security appliance 160. The coupling to the network security appliance 150 and/or the endpoint security appliance 160 may be through a wired connection or via a wireless network, such as the network 140. In addition, the detection system 110₁ may be communicatively coupled, e.g., indirectly via the endpoint security appliance 160, to the cloud services 170. Finally, the detection system 110₁ may be communicatively coupled to a storage device 130.

As further shown in FIG. 1, the detection system 110₁ may include one or more VM hosts 111₁-111_P (P≥1) and a controller 112. The VM host 111₁ may include one or more VMs, a security VM, and a hypervisor. In a first embodiment, the VM host 111₁ may be a hardware appliance hosting several VMs wherein the VM host 111₁ may receive instructions, commands and/or updates via the controller 112. According to this embodiment, the VM hosts 111₁-111_P may reside in a single location on the same hardware (e.g., an enterprise server). In a second embodiment, the VM hosts 111₁-111_P may reside on separate hardware devices wherein each VM host 111₁-111_P is communicatively coupled to the controller 112. In the second embodiment, the separate hardware devices housing one or more VM hosts 111₁-111_P may be located at a central location (e.g., a datacenter) or located at various remote locations and communicatively coupled to the controller 112 and the storage 130 (e.g., via a network). The storage 130 may include storage space (e.g., a non-transitory storage medium) for each of the VMs included in the detection system 110₁.

As will be discussed in detail below, each VM_i (1≤i≤M, wherein M=3 for this example) deploying as part of the VM host 111₁ may model a separate endpoint device (e.g., acting as a virtual representation of the endpoint device including running a specified operating system and one or more specified applications). Alternatively, one or more VMs (e.g., VM₁-VM_M) of the VM host 111₁ may model a single endpoint device wherein various versions of operating systems and/or applications are run according to configurations of the endpoint device. For example, VM₁ of the VM host 111₁ may run the Microsoft® Windows® 8.1 operating system and Adobe® Acrobat® version 10.0 while VM₂ of the VM host 111₁ may run the Microsoft® Windows® 8.1 operating system and Adobe® Acrobat® version 11.0 in order to simulate the endpoint devices as needed by, for example, a corporation utilizing the detection system 110₁. Additionally, the use of various configurations may also aid in determining the whether malware and/or exploits are triggered in a specified version of an application.

The security VM₁ of the VM host 111₁ serves as a VM dedicated to providing further analysis of an object identified in one or more of the VM₁-VM_M as being “potentially suspicious.” As will be discussed below, each of the VM₁-VM_M implemented within the VM host 111₁ provides a virtualized run-time environment in which to execute an object. The detection system 110₁ also enables a determination as to whether one or more triggering events (e.g., events that signify the object may be suspicious and/or malicious) are detected during the execution of the object. Once a triggering event is detected by at least one VM of VM₁-VM_M (for this example, VM₁ will be used), the object is determined to be “potentially suspicious,” and information associated with the object and the detected event is transmitted to the security VM₁ of the VM host 111₁ wherein further analysis is done to determine whether the object is “suspicious.” If the security VM₁ determines the object is suspicious, a request for further information associated with the detected event is transmitted to the VM₁. Subsequently, the VM₁ may provide the security VM₁ with information associated with the event and information associated with the object, wherein the information provided is based on processing conducted subsequent to the detection of the event (e.g., information detailing one or more of changes to Registry files, the operating system and/or configuration files, attempts to access locations in memory, generated files, etc.). The security VM₁ performs subsequent analysis based on one or more portions of information received from the VM₁ and determines whether the object is malicious.

The hypervisor₁ included in the VM host 111₁ manages communications between VM₁-VM_M and the security VM₁. In addition, the hypervisor₁ may include logic to manage communications between the VM₁-VM_M and the security VM₁ as well as data storage.

In a first embodiment, each of the VMs within the VM hosts 111₁-111_P correspond to endpoint devices within a separate enterprise network. For example, each of VM₁-VM_M deployed within the VM host 111₁ may correspond to endpoint devices within an enterprise network of a first company and each of VM₁-VM_N deployed within the VM host 111₂ may correspond to endpoint devices within an enterprise network of a second company. In a second embodiment, each VM₁-VM_Q of the VM host 111_P may correspond to endpoint devices of a plurality of enterprise networks of a plurality of companies. For example, VM₁ and VM₂ of the VM host 111_P may correspond to endpoint devices of an enterprise network of a first company and the VM₃ and the VM_Q of the VM host 111_P may correspond to endpoint devices of an enterprise network of a second company.

A. Virtual Machine Host Component

Referring now to FIG. 2, a block diagram of an exemplary the virtual machine (VM) host 111₁ of the virtualized malware detection system 110₁ that is communicatively coupled to a network security appliance 150, an endpoint security appliance 160 and cloud services 170 is shown. In particular, FIG. 2 provides a detailed illustration of the interactions within the VM host 111₁. As shown, the each of VM₁-VM_M communicates with the security VM₁ using the hypervisor₁ included in the VM host 111₁. In addition, FIG. 2 illustrates the hypervisor₁ may include an exploit queue storage 210 and a VM routing logic 220.

The exploit queue storage 210 may store triggering events detected by one or more of VM₁-VM_M for example in a queue format, until the security VM₁ is available for processing. In one embodiment, the security VM₁ may be performing analysis on information received from the VM₁ when the VM₂ detects a triggering event during processing. In such a situation, the exploit queue storage 210 may store the information associated with the object being processed and the triggering event detected by the VM₂ until the security VM₁ is available (e.g., has finished processing the information provided by the VM₁ and/or information in the queue ahead of the information provided by the VM₂).

Included in the hypervisor₁, the VM routing logic 220 may include logic to determine what information in the queue to provide to the security VM₁ as well as control communications from the security VM₁ to each of VM₁-VM_M. In addition, the VM routing logic 220 may maintain, e.g., via a table stored in the exploit queue storage 210, information such as (i) the VM from which information was provided and/or (ii) the process identification (ID) of the process in which the triggering event was detected.

Now referring to FIG. 3, an in-depth block diagram of a portion of the VM host 111₁ of the detection system 110₁ communicatively coupled to a network security appliance 150, an endpoint security appliance 160 and cloud services 170 is shown. The portion of the VM host 111₁ that is illustrated includes the VM_M, the security VM₁ and the hypervisor₁. Of course, as illustrated previously, the VM host 111₁ may include a plurality of VMs as seen in FIG. 1.

As illustrated, the VM_M includes an exploit event detection engine 301, a post-exploit event detection engine 302, a dynamic protection engine 303 and a forensics assistant 304. The exploit event detection engine 301 monitors the processing of the object within the VM_M in order to detect an occurrence of a triggering event. Upon detecting the triggering event, the exploit event detection engine 301 collects information associated with the detected triggering event as well as information of the object. Examples of the collected information may include, but is not limited or restricted to, a type of triggering event, the application processing the object that caused the triggering event (e.g., application name and/or version number), input to an application that triggered the event (e.g., simulated user input and/or a document that was used as input), the operating system name and version, etc. The collected information is then provided to the security VM₁ for further analysis. Following the detection of an occurrence of the triggering event, the post-exploit event detection engine 302 monitors the processing of the object with the VM_M.

The collection of triggering events may be pre-defined and pre-stored, e.g., in the storage 130. Alternatively, although not illustrated, the triggering events may be pre-stored in a storage device located within the detection system 110₁, in the VM host 111₁, or in each of the VM₁-VM_M. The collection of triggering events may be updated via the management system 120 (e.g., due to one or more detections of an exploit and/or malware by detection systems 110₂—detection systems 110₃). Additionally, the collection of triggering events may be updated by the network security appliance 150, the endpoint security appliance 160 and/or the cloud services 170, wherein such an update may be facilitated by the security VM_M.

Upon detection of a triggering event, the post-exploit event detection engine 302 monitors the processing of the object in the VM_M. In contrast, the exploit detection engine 301 monitors the processing prior to detection of a triggering event. Specifically, the monitoring conducted by the post-exploit event detection engine 302 is to detect any changes made to the environment within the VM_M (e.g., change in Registry files, attempted changes to the operating system, etc.) and/or actions performed as a result of the triggering event (e.g., a successful or unsuccessful attempt to delete a file/directory, create a file/directory, establish communication with an external server, password protect a file/directory, encrypt a file/directory, etc.). Upon request, the post-exploit event detection engine 302 provides the security VM₁ with the collected information associated with the actions and changes subsequent to the detection of the occurrence of the triggering event.

The dynamic protection engine 303 determines identifying information of the object (e.g., URL, file name of the object, name of attachment in which the object was received email address of the sender of the object, etc.) and provides the initial identifying information to a pre-processing engine 312 implemented within the security VM₁ as described below. Upon analysis by the pre-processing engine 312, to be discussed below, the dynamic protection engine 303 receives a notification from the pre-processing engine 312 based on the analysis. Based on the notification from the pre-processing engine 312, the dynamic protection engine 303 may allow the VM_M to continue processing or stop processing of the object based on a determination as to whether the object is malicious or non-malicious. Additionally, the dynamic protection engine 303 may save the state information of the VM_M when a triggering event is detected (hereinafter referred to as “taking a snapshot”). Subsequently, once processing of the object has been completed and/or the object has been identified as malicious or non-malicious, the dynamic protection engine 303 may revert the VM_M back to the state captured in the snapshot.

The forensics assistant 304 collects state information of the VM_M upon the detection of the occurrence of a triggering event. Examples of information collected by the forensics assistant 304 include, but are not limited or restricted to, registry key, registry values, metadata of specific storage locations (e.g., files and/or folders that may be interacted with during processing of an object), a disk image (e.g., a snapshot of a storage device), network activity occurring during processing of an object and/or requests of user interaction occurring during processing of an object.

The security VM₁ includes an analysis engine 311, the pre-processing engine 312 and a forensics collection engine 313. The analysis engine 311 receives information associated with a triggering event and the object being processed from the VM_M. Specifically, the analysis engine 311 receives information associated with the triggering event and information associated with the object being processed from the exploit event detection engine 301 as discussed above. The analysis engine 311 performs an analysis on the information received from the exploit event detection engine 301 to determine whether the object being processed in “suspicious.” The analysis performed by the analysis engine 311 is based on one or more correlation rules. The correlation rules provide a basis to determine whether the object is suspicious by assessing whether the information associated with the detected triggering event and the object being processed results in a rule violation.

For example, one correlation rule may state a prohibition of a particular application writing to a predetermined storage location without permission. A violation of the correlation rule would occur when the particular application attempts to write data to the predetermined storage without permission (whether the write attempt was successful or unsuccessful). Additional correlation rules may relate to occurrences of anomalous behaviors and may be defined for particular applications and/or may be defined for all applications. Additional examples of correlation rules may include, but are not limited or restricted to, a prohibition on copying data stored in a predetermined storage location, a prohibition on deleting data stored in a predetermined storage location, a prohibition on creating a file in a predetermined location in a file system, a prohibition on encrypting data in a predetermined storage location, or the like.

The analysis engine 311 may include a single set of correlation rules that are applied to all information associated with a detected triggering event. Alternatively, the analysis engine 311 may include multiple sets of correlation rules wherein intelligence (e.g., a filtering mechanism) may determine one or more sets of correlation rules to apply to the received information. For example, one or more sets of correlation rules may be applied based on, among other features, the detected triggering event, the operating system of the VM_M and/or the application/process that triggered the occurrence of the triggering event.

When the analysis based on the correlation rules results in a determination that the object is “suspicious,” the analysis engine 311 requests, from the post-exploit event detection engine 302 of the VM_M, additional information pertaining to any changes to and/or actions taken within the environment of the VM_M. Based upon the received additional information provided by the post-exploit event detection engine 302, the analysis engine 311 performs further analysis based on correlation rules (e.g., the same as above, a separate set of correlation rules, or a combination) to determine whether the suspicious object is malicious. In one embodiment, the correlation rules may be applied to both the initial information provided by the exploit event detection engine 301 as well as the additional information provided by the post-exploit event detection engine 302. As discussed above, a single set of correlation rules may be applied to the additional information (and optionally the initial information) or one or more sets of correlation rules may be applied based on intelligence in the analysis engine 311. Based on the analysis of at least the additional information provided by the post-exploit event detection engine 302, the analysis engine 311 may determine that the object is “malicious.”

Furthermore, the analysis of the additional information may occur repeatedly when an analysis of the additional information does not initially result in a determination that the object is malicious. When the analysis of the additional information does not result in a determination that the object is malicious and when the processing of the object in the VM_M is not yet complete, the analysis engine 311 may requests more additional information (e.g., information pertaining to changes to and/or actions taken within the environment of the VM_M since the first set of additional information was provided). The reoccurring process is illustrated in FIGS. 4A and 4B and will be discussed below.

The pre-processing engine 312 receives identifying information from the dynamic protection engine 303 and performs an analysis to determine whether the identifying information of the object matches an entry on either a whitelist or a blacklist. When the identifying information of the object matches an entry on the blacklist (e.g., signifying that the object has previously been determined to be malicious), the pre-processing engine 312 may (i) notify the dynamic protection engine 303 that processing of the object should cease as the object is malicious, or (ii) notify the dynamic protection engine 303 of the match between the identifying information of the object and an entry on the blacklist but that processing in the VM_M should continue in order to determine and/or verify any malicious effects as a result of processing the object.

When the identifying information of the object matches an entry on the whitelist (e.g., signifying that the object has previously been determined to be non-malicious), the pre-processing engine 312 may (i) notify the dynamic protection engine 303 that processing of the object should cease as the object is non-malicious, or (ii) notify the dynamic protection engine 303 of the match between the identifying information of the object and an entry on the whitelist but that processing in the VM_M should continue in order to verify the object is non-malicious based on updated correlation rules and/or other information collected/generated since the object was placed on the whitelist. When the object does not appear on either a whitelist or a blacklist, the dynamic protection engine 303 is notified to continue processing so that a determination of malicious/non-malicious can be made.

As is illustrated in FIG. 3, the analysis engine 311 is communicatively coupled to the endpoint security appliance 160 and the cloud services 170. The result of the analysis performed by the analysis engine 311 may be supplied to the endpoint security appliance 160 for the purpose of alerting a user, network administrator and/or expert network analyst and/or enabling a user, network administrator and/or expert network analyst to view the results. Examples of information that may be displayed for viewing include, but are not limited or restricted to, one or more of (i) how the object entered the network, (ii) which application(s) triggered the occurrence of the triggering event and/or (iii) any malicious effects from the processing of the object. Additionally, the object may be processed by the endpoint security appliance 160 using a two-step analysis of a static analysis and a dynamic analysis. An illustrative example of the two-step analysis including a static analysis and a dynamic analysis may be found in a prior U.S. patent application entitled “Dynamically Remote Tuning of Malware Content Detection System,” U.S. patent application Ser. No. 14/231,216 filed Mar. 31, 2014, the contents of which are incorporated by reference. Furthermore, the results of the analysis engine 311 may be uploaded to the cloud services 170 for future reference by the detection systems 110_1-3 and/or client devices.

As is further illustrated, the pre-processing engine 312 is communicatively coupled to the network security appliance 150. The result of the analysis performed by the pre-processing engine 312 may be supplied to the network security appliance 150 for the purpose of alerting a user, network administrator and/or expert network analyst and/or enabling a user, network administrator and/or expert network analyst to view the results. The viewing may include, among other things, the whitelist/blacklist on which the object appears as well as information associated with the object collected during analysis and/or processing. Additionally, the object may be processed by the network security appliance 170 using a two-step analysis of a static analysis and a dynamic analysis.

The forensics collection engine 313 collects and analyzes the information (e.g., forensics data) collected by the forensics assistants 304 of each VM₁-VM_M within the VM Host 111₁. The forensics collection engine 313 may collect the information from the forensics assistants 304 at (i) scheduled intervals and/or (ii) upon detection of a triggering event. The forensics collection engine 313 may be configured to query the forensics assistants 304 of each VM₁-VM_M at scheduled intervals. Additionally, or in the alternative, forensics collection engine 313 may be configured to query the forensics assistants 304 of the VM_M in which a triggering event was detected upon the detection.

Upon receiving a response to one or more queries, the forensics collection engine 313 correlates the collected information against a predetermined rule set (e.g., one or more rules that may recite expected behavior by an object and/or application, and/or one or more rules that may recite anomalous behavior by an object and/or application). The rule set may identify abnormalities of the processing environment of the VM_M. For example, the correlation may identify deviations from expected behaviors of an object of a particular object-type or application as a result of processing an object. In one embodiment, one or more of the deviations identified by the rule set may not rise to a suspicious level of a triggering event; however, the forensics collection engine 313 may determine one or more rules of the rule set that were violated, and therefore an alert to, for example, the endpoint security appliance 160 for further processing may be provided by the security VM. Alternatively, one or more rules of the rule set may have different weightings that are used in determining whether an alert should be issued. Furthermore, the detection system 110₁ may also provide alerts to the user remotely controlling the VM_M from which the information violating the one or more rules of a rule set was collected, a network administration and/or an expert malware analyst. The correlation results and/or information collected by the forensics collection engine 313 may be stored in the storage 130, for example.

Similarly, the forensics collection engine of the security VM₂ collects and analyzes the information collected by the forensics assistants of each VM₁-VM_N within the VM Host 111₂ while the forensics collection engine of the security VM₃ collects and analyzes the information collected by the forensics assistants of each VM₁-VM_Q within the VM Host 111_P.

B. Exemplary Flow of a Virtualized Malware Detection System

Referring to FIGS. 4A and 4B, flowcharts illustrating an exemplary method for detecting malware and/or exploits using the detection system 110₁ are shown. Referring to FIG. 4A, a flowchart illustrating a first exemplary method of detecting malware and/or exploits using detection system 110₁ is shown. Each block illustrated in FIG. 4A represents an operation performed in the method 400A of detecting malware and/or exploits based on the use of detection system 110₁ wherein the first VM (e.g., the VM_M as described above) and the second VM (e.g., the security VM₁ as described above) process the object concurrently (at least partially overlapping at the same time).

At block 401, an object is received within network traffic and at block 402, the object undergoes pre-processing to determine whether the object is malicious. In this embodiment, the pre-processing may include a comparison of information identifying the object with an entry on a whitelist and/or a blacklist. A match between the identifying information of the object and an entry on the blacklist may signify the object is malicious. When the object is determined to be malicious based on pre-processing (yes at block 402), actions are performed to handle the malicious object (block 403). Examples of actions performed to handle a malicious object include, but are not limited or restricted to, (i) generating an alert to notify a user, (ii) a network administrator and/or an expert network analyst, uploading information associated with the malicious object to the cloud services and/or (iii) preventing the object from being received and/or processed by a client device, if possible.

When pre-processing does not result in a determination that the object is malicious (no at block 402), the first VM launches the object and begins processing (block 405). At block 406, at least one of a set of triggering events is detected that identifies the object as being “potentially suspicious.” When an object is identified as being potentially suspicious, information associated with the object and information associated with the detected triggering event are provided to the second VM for analysis based on one or more correlation rules (block 407). Based on the analysis performed in block 407, a determination is based as to whether the object is suspicious (block 408). When the object is not determined to be suspicious (no at block 408), a determination is made as to whether the processing of the object is complete (e.g., the determination may be made by either the exploit event detection engine 301 or the post-exploit event detection engine 302) (block 409). When the processing of the object is not complete (no at block 409), the first VM continues processing the object while monitoring the processing for the occurrence of at least one or a predefined set of triggering event. When the processing of the object is complete (yes at block 409), the object is determined to be non-malicious (block 410), as no triggering event that led to a determination the object was suspicious was detected.

When the object is determined to be suspicious (yes at block 408), the second VM requests information associated with the post-event processing of the suspicious object from the first VM (block 412). In one embodiment, the first VM continues to process the object upon detection of the occurrence of the triggering event (e.g., the processing of the first VM and the analysis by the second VM occur concurrently). Therefore, when the second VM requests information associated with the post-event processing, the first VM supplies the information associated with the post-event processing without having to first continue processing of the object.

The second VM performs further analysis of the information associated with the post-event processing (and, optionally, the information pertaining to the triggering event) based on correlation rules (block 413). Following the further analysis by the second VM, a determination as to whether the object is malicious is made (block 414). When the object is determined to be malicious (yes at block 414), actions for handling the malicious object, as discussed above, are performed (block 415).

When the object is not determined to be malicious (no at block 414), a determination is made as to whether the processing of the object is complete (block 417). When the processing is complete (yes at block 417), the object is determined to be non-malicious (block 419). Additionally, as described above, the results of the analyses performed by the first VM and/or the second VM may be provided to a user, a network administrator and/or an export network analyst via one or more security appliances.

When the processing of the object is not complete (not at block 417), the first VM provides the second VM with additional information associated with the post-event processing (e.g., information associated with post-event processing since the second VM was previously provided with information associated with the post-event processing) (block 418) and the method 400A returns to block 413.

Referring to FIG. 4B, a flowchart illustrating a second exemplary method of detecting malware and/or exploits using the detection system 110₁ is shown. Each block illustrated in FIG. 4B represents an operation performed in the method 400B of detecting malware and/or exploits based on the use of the detection system 110₁ wherein the first VM and the second VM do not process the object concurrently.

At block 401, an object is received within network traffic and at block 402, the object undergoes pre-processing to determine whether the object is malicious. In this embodiment, the pre-processing may include a comparison of information identifying the object with an entry on a whitelist and/or a blacklist. A match between the identifying information of the object and an entry on the blacklist may signify the object is malicious. When the object is determined to be malicious based on pre-processing (yes at block 402), actions are performed to handle the malicious object (block 403). Examples of actions performed to handle a malicious object include, but are not limited or restricted to, (i) generating an alert to notify a user, (ii) a network administrator and/or an expert network analyst, uploading information associated with the malicious object to the cloud services and/or (iii) preventing the object from being received and/or processed by a client device, if possible.

When pre-processing does not result in a determination that the object is malicious (no at block 402), the first VM launches the object and begins processing the object (block 405). At block 406, at least one of a set of triggering events is detected that identifies the object as being “potentially suspicious.” In one embodiment, when an object is identified as being potentially suspicious, the first VM pauses processing and provides information associated with the object and the detected triggering event is provided to the second VM for analysis based on correlation rules (block 421). The second VM performs an analysis of the event information using correlation rules at block 407. Based on the analysis performed in block 407, a determination is based as to whether the object is suspicious (block 408). When the object is not determined to be suspicious at block 408, a determination is made as to whether the processing of the object is complete (e.g., the determination may be made by either the exploit event detection engine 301 or the post-exploit event detection engine 302) (block 409). When the processing of the object is not complete (no at block 409), the first VM continues processing the object, monitoring the processing for the occurrence of a triggering event. When the processing of the object is complete (yes at block 409), the object is determined to be non-malicious (block 410) as no triggering event that led to a determination the object was even suspicious was detected. Subsequently, the process may end at block 411.

When pre-processing does not result in a determination that the object is malicious (no at block 402), the first VM launches the object and begins processing (block 405). At block 406, at least one of a set of triggering events is detected that identifies the object as being “potentially suspicious.” When an object is identified as being potentially suspicious, the first VM pauses processing and provides information associated with the object and the detected triggering event is provided to the second VM for analysis based on correlation rules (block 421). The second VM performs an analysis of the event information using correlation rules at block 407. Based on the analysis performed in block 407, a determination is based as to whether the object is suspicious (block 408). When the object is not determined to be suspicious at block 408, a determination is made as to whether the processing of the object is complete (e.g., the determination may be made by either the exploit event detection engine 301 or the post-exploit event detection engine 302) (block 409). When the processing of the object is not complete (no at block 409), the first VM continues processing the object while monitoring the processing for the occurrence of at least one of a predefined set of triggering event. When the processing of the object is complete (yes at block 409), the object is determined to be non-malicious (block 410), as no triggering event that led to a determination the object was suspicious was detected.

When the object is determined to be suspicious (yes at block 408), the second VM notifies the first VM to (i) continue processing the object and (ii) provide information associated with the post-event processing of the suspicious object (block 422). In one embodiment, the first VM continues to process the object upon detection of the triggering event (e.g., the processing of the first VM and the analysis by the second VM occur concurrently). Therefore, when the second VM requests information associated with the post-event processing, the first VM supplies the information associated with the post-event processing without having to first continue processing of the object.

The first VM pauses processing and the second VM performs further analysis of the information associated with the post-event processing (and, optionally, the information pertaining to the triggering event) based on correlation rules (block 413). Following the further analysis by the second VM, a determination as to whether the object is malicious is made (block 414). When the object is determined to be malicious (yes at block 414), actions for handling the malicious object are performed, as discussed above (block 415).

When the object is not determined to be malicious (no at block 414), a determination is made as to whether the processing of the object is complete (block 417). When the processing is complete (yes at block 417), the object is determined to be non-malicious (block 419). Additionally, as described above, the results of the analyses performed by the first VM and/or the second VM may be provided to a user, a network administrator and/or an export network analyst via one or more security appliances.

When the processing of the object is not complete (not at block 417), the first VM continues processing and subsequently provides the second VM with additional information associated with the post-event processing (block 418) and the method 400B returns to block 413.

Referring to FIGS. 5A and 5B, flowcharts illustrating an exemplary method for detecting malware and/or exploits associated with a URL opened within a browser using the detection system 110₁ are shown. Each block illustrated in FIGS. 5A and 5B represents an operation performed in the method 500 of detecting malware and/or exploits based on the use of the detection system 110₁ wherein a URL is processed in a browser application. Referring to FIG. 5A, at block 501, a URL is opened in a browser application within a first VM. At block 502, information identifying the URL, in this example, the URL itself, is sent to the second VM for pre-processing to determine whether the URL matches an entry on a blacklist or a whitelist. When the URL is determined to be malicious, e.g., matches an entry on a blacklist (yes at block 503), actions to handle the malicious URL are performed at block 504. Examples of actions to handle the malicious URL include, but are not limited or restricted to, (i) generating an alert to notify a user, (ii) a network administrator and/or an expert network analyst uploading information associated with the malicious object to the cloud services and/or (iii) preventing the object from being received and/or processed by a client device, if possible. In one embodiment, an alert for user notification may be generated at block 505.

When the URL is not determined to be malicious (no at block 503), the first VM processes the URL in the browser application (block 507). At block 508, a determination is made as to whether a triggering event that identifies the URL as “potentially suspicious” has been detected prior to the completion of the processing of the URL. When a triggering event has not been detected (no at block 508), the URL is determined to be non-malicious at block 509. At optional block 510, the second VM may add information associated with the non-malicious URL to a whitelist for future reference and the process may subsequently end at block 511.

When a triggering event has been detected (yes at block 508), the URL is determined to be “potentially suspicious,” at block 512 and information associated with the detected triggering event and the information associated with the URL are provided to the second VM at block 513. At block 514, the second VM analyzes the information associated with the detected triggering event and the information associated with the URL using one or more correlation rules wherein a first score is assigned to the URL based on the analysis.

The method 500 continues in FIG. 5B at block 515 wherein a determination is made as to whether the first score is above a first threshold. When the first score is not above the first threshold (no at block 515), the first VM continues processing the URL in the browser application (block 516) and the method 500 returns to block 507.

When the first score is determined to be above the first threshold (yes at block 515), the URL is determined to be “suspicious” (block 517). At block 518, the first VM provides the second VM with information associated with the post-event processing associated with the URL (e.g., changes to or actions conducted within the virtual run-time environment of the first VM since the triggering event was detected). At block 519, the second VM performs further analysis on, at least, the information associated with the post-event processing (may optionally include the information associated with the triggering event and the information associated with the URL) based on one or more correlation rules, which may be the same as the correlation rules used above in association with block 514, may be separate correlation rules, or may be a combination thereof. At block 520, a second score is assigned to the URL based on the processing of, at least, the information associated with the post-event processing. A determination is made as to whether the second score exceeds a second threshold at block 521. When the second score is not above the second threshold (no at block 521), the first VM continues processing the URL in the browser application (block 522) and the method 500 returns to block 507.

When the second score is determined to be above the second threshold (yes at block 521), the URL is determined to be “malicious” (block 523). Actions may then be taken to handle the malicious URL. For example, (i) at optional block 524, the information associated with the URL (e.g., identifying information, metadata, information signifying how the URL was received by the detection system 110₁, etc.) may be provided to the second VM for inclusion on a blacklist, (ii) at optional block 525, an alert may be generated for user notification (in one embodiment, using an endpoint security appliance 160 and/or a network security appliance 150), and/or (iii) at optional block 526, the information associated with the URL may be uploaded to cloud services 170 for reference by one or more other detection systems and/or by one or more client devices.

Referring now to FIG. 6, an exemplary block diagram of logic associated with the detection system 110₁ of FIG. 1 is shown. The detection system 110₁ includes one or more processors 600 that are coupled to the communication interface logic 610 via a first transmission medium 620. Communication interface logic 610 enables communication with other detection systems 110₂-110₃ and management system 120 of FIG. 1. According to one embodiment of the disclosure, the communication interface logic 610 may be implemented as a physical interface including one or more ports for wired connectors. Additionally, or in the alternative, communication interface logic 610 may be implemented with one or more radio units for supporting wireless communications with other network devices.

The one or more processors 600 are further coupled to the persistent storage 630 via the transmission medium 625. According to one embodiment of the disclosure, the persistent storage 630 may include (i) the VM_M including the exploit event detection engine 301, the post exploit event detection engine 302, the dynamic protection engine 303 and the forensics assistant 304; and (ii) the security VM₁ including the analysis engine 311, the pre-processing engine 312 and the forensics collection engine 313. Of course, when implemented as hardware (such as circuitry and/or programmable logic arrays), one or more of these logic units could be implemented separately from each other. In addition, one or more of these logic units may be implemented in hardware while one or more logic units may be implemented as software.

In the foregoing description, the invention is described with reference to specific exemplary embodiments thereof. It will, however, be evident that various modifications and changes may be made thereto without departing from the broader spirit and scope of the invention as set forth in the appended claims.

Claims

1. A system comprising: one or more hardware processors; anda storage module communicatively coupled to the one or more hardware processors, the storage module comprising logic that, upon execution by the one or more hardware processors, performs operations comprising: providing, by a first virtual machine, identifying information corresponding to an object to a security virtual machine for analysis of the object, the first virtual machine being one of a plurality of virtual machines and the first virtual machine being different than the security virtual machine;performing, by the security virtual machine, a pre-processing based on the identifying information to determine whether the object is malicious;responsive to the pre-processing resulting in a determination of suspicious, potentially suspicious or non-malicious, processing the object in the first virtual machine;upon detection within the first virtual machine of a triggering event, providing, by the first virtual machine, information associated with the triggering event to the security virtual machine, wherein the detection occurs during the processing of the object in the first virtual machine; anddetermining, within the security virtual machine, whether the object is malicious based upon an analysis of the information associated with the triggering event using one or more correlation rules, wherein a hypervisor coordinates communication between the security virtual machine and the plurality of virtual machines to prevent comingling of data between the plurality of virtual machines, and wherein the information associated with the triggering event is different than the identifying information.

2. The system of claim 1, wherein the pre-processing includes a signature check of the identifying information of the object.

3. The system of claim 1, wherein the information associated with the triggering event is an event that, through at least one of experiential knowledge or machine learning techniques, has been determined to have an association with a malicious attack.

4. The system of claim 1, wherein the triggering event is an attempt to perform one of: (i) deleting a first file or a first directory, (ii) creating a second file or second directory, (iii) establishing communication with an external server, (iv) protecting a file or directory with a password, or (v) encrypting a third file or a third directory.

5. The system of claim 1, wherein the first virtual machine simulates an endpoint device through configuration of the first virtual machine using a prescribed software profile.

6. The system of claim 1, wherein logic within the security virtual machine receives information associated with one or more triggering events from each of a plurality of first virtual machines included in the system.

7. The system of claim 1 further comprising: upon determining that the object is malicious, generating an alert within the security virtual machine to notify one or more of a user of an endpoint device, a network administrator or an expert network analyst.

8. The system of claim 7, wherein the alert is provided to the one or more of a user of an endpoint device, a network administrator or an expert network analyst through a security appliance.

9. The system of claim 1 further comprising: upon determining that the object is malicious, adding information associated with the object to a blacklist.

10. The system of claim 1 further comprising: upon determining that the object is malicious, uploading information associated with the object to cloud services.

11. The system of claim 1 further comprising: receiving, by logic of the security virtual machine, (a) one or more updates to one or more of (i) a blacklist, (ii) a whitelist, or (iii) a correlation rule of one or more correlation rules, or (b) a set of correlation rules including updates to the one or more correlation rules from a security appliance.

12. The system of claim 1, wherein the information associated with the triggering event is provided from the first virtual machine to the security virtual machine via the hypervisor communicatively coupled to the first virtual machine and the security virtual machine.

13. The system of claim 1, wherein a first set of the plurality of virtual machines is configured for utilization by a first entity and a second set of the plurality of virtual machines is configured for utilization by a second entity, and wherein the first entity and the second entity are each one of an enterprise or an entrepreneur.

14. The system of claim 1, wherein the plurality of virtual machines are contained within a virtual machine host.

15. A non-transitory computer readable medium having stored thereon logic that, upon execution by one or more processors, performs operations comprising: providing, by a first virtual machine, identifying information corresponding to an object to a security virtual machine for analysis of the object, wherein the first virtual machine being one of a plurality of virtual machines and the first virtual machine being different than the security virtual machine;performing, by the security virtual machine, a pre-processing based on the identifying information to determine whether the object is malicious;responsive to the pre-processing resulting in a determination of suspicious, potentially suspicious or non-malicious, processing the object in the first virtual machine;upon detection, by the first virtual machine, of a triggering event, providing, by logic of the first virtual machine, information associated with the triggering event to the security virtual machine, wherein the detection of the triggering event occurs during the processing of the object in the first virtual machine; anddetermining, within the security virtual machine, that the object is malicious based upon an analysis of information associated with the triggering event and additional information associated with processing of the object after the detection of the triggering event using one or more correlation rules, wherein a hypervisor coordinates communication between the security virtual machine and the plurality of virtual machines to prevent comingling of data between the plurality of virtual machines, and wherein the information associated with the triggering event, the additional information, and the identifying information are different from each other.

16. The non-transitory computer readable medium of claim 15, wherein the logic that, upon execution by the one or more processors, performs operations further comprising: determining, within the security virtual machine, that the object is suspicious based upon an analysis of the information associated with the triggering event using one or more correlation rules; andupon determination the object is suspicious based on the information associated with the triggering event, requesting, by logic within the security virtual machine, the additional information associated with processing of the object after the detection of the triggering event.

17. The non-transitory computer readable medium of claim 15, wherein the pre-processing includes a signature check of the identifying information of the object.

18. The non-transitory computer readable medium of claim 15, wherein the information associated with the triggering event is an event that, through at least one of experiential knowledge or machine learning techniques, has been determined to have an association with a malicious attack.

19. The non-transitory computer readable medium of claim 15, wherein the triggering event is an attempt to perform one of: (i) deleting a first file or a first directory, (ii) creating a second file or second directory, (iii) establishing communication with an external server, (iv) protecting a file or directory with a password, or (v) encrypting a third file or a third directory.

20. The non-transitory computer readable medium of claim 15, wherein the information associated with the triggering event is provided from the first virtual machine to the security virtual machine via the hypervisor communicatively coupled to the first virtual machine and the security virtual machine.

21. The non-transitory computer readable medium of claim 15, wherein a first set of the plurality of virtual machines is configured for utilization by a first entity and a second set of the plurality of virtual machines is configured for utilization by a second entity, and wherein the first entity and the second entity are each one of an enterprise or an entrepreneur.

22. The non-transitory computer readable medium of claim 15, wherein the plurality of virtual machines are contained within a virtual machine host.

23. The non-transitory computer readable medium of claim 15, wherein the logic that, upon execution by the one or more processors, performs operations further comprising: providing, by the logic of the first virtual machine, additional information associated with processing of the object after the detection of the triggering event, to the security virtual machine; anddetermining, within the security virtual machine, that the object is malicious based upon an analysis of the additional information associated with processing of the object after the detection of the triggering event using one or more correlation rules, wherein a hypervisor coordinates communication between the security virtual machine and the plurality of virtual machines to prevent comingling of data between the plurality of virtual machines.

24. A computerized method comprising: providing, by a first virtual machine, identifying information corresponding to an object to a security virtual machine for analysis of the object, wherein the first virtual machine being one of a plurality of virtual machines and the first virtual machine being different than the security virtual machine;performing, by the security virtual machine, a pre-processing based on the identifying information to determine whether the object is malicious;responsive to the pre-processing resulting in a determination of suspicious, potentially suspicious or non-malicious, processing the object in the first virtual machine;upon detection, within the first virtual machine, of a triggering event, providing, by logic within the first virtual machine, information associated with the triggering event to the security virtual machine, wherein the detection occurs during the processing of the object in the first virtual machine; anddetermining, within the security virtual machine, that the object is malicious based upon an analysis of the information associated with the triggering event using one or more correlation rules, and wherein a hypervisor coordinates communication between the security virtual machine and the plurality of virtual machines to prevent comingling of data between the plurality of virtual machines, wherein the information associated with the triggering event is different than the identifying information.

25. The computerized method of claim 24, wherein the triggering event is an attempt to perform one of: (i) deleting a first file or a first directory, (ii) creating a second file or second directory, (iii) establishing communication with an external server, (iv) protecting a file or directory with a password, or (v) encrypting a third file or a third directory.

26. The computerized method of claim 24, wherein the information associated with the triggering event is provided from the first virtual machine to the security virtual machine via the hypervisor communicatively coupled to the first virtual machine and the security virtual machine.

27. The computerized method of claim 24, wherein a first set of the plurality of virtual machines is configured for utilization by a first entity and a second set of the plurality of virtual machines is configured for utilization by a second entity, and wherein the first entity and the second entity are each one of an enterprise or an entrepreneur.

28. The computerized method of claim 24, wherein the plurality of virtual machines are contained within a virtual machine host.