Method and system for executing workload (APP) in execution environment
By using the allowability list and integrity verification module in the execution environment to evaluate IT security risks and dynamically manage the execution of workloads, the IT security risks problem on the untrusted computing platform is solved, and the secure workload on the open computing platform is realized implement.
Patent Information
- Application Number
- CN202380031536.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2022-03-31
- Filing Date
- 2023-03-13
- Publication Date
- 2024-11-15
AI Technical Summary
In an execution environment, it is impossible to effectively prevent critical workloads from being executed on untrusted computing platforms, resulting in IT security risks.
Determine and manage workloads executed in execution environments through allowance lists, leverage software-based integrity verification modules to evaluate IT security risks, and dynamically adjust the allowance lists based on risk information to ensure they are executed only in trusted environments Workloads, adopt the 'zero trust' paradigm to adapt to risk and reliability.
Implement dynamic management of workload execution without assuming the credibility of the execution environment, improves network security, and ensures that workloads only operate in environments with low IT security risks. It is suitable for open computing platforms, especially in Effectively maintain safety in high-risk environments.
Smart Images

Figure CN118974726A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a method for executing a workload in an execution environment and a system having an execution environment configured to execute a workload. Background Art
[0002] In an execution environment, for example, on a computing platform such as an industrial PC or an edge cloud computing server, different and partially critical software components are executed. In particular, critical functionality, for example in the form of applications (Apps) and / or containers or native software components, can be implemented on an open computing platform. In general, such applications, containers and other software components can also be referred to as "workloads".
[0003] However, critical workloads cannot be prevented from being executed on computing platforms that do not guarantee the required level of trustworthiness. Summary of the invention
[0004] The object of the present invention is therefore to provide an improved method for executing a workload in an execution environment and an improved system having an execution environment configured for executing a workload. In particular, in the case of the method and in the case of the system, it should be possible to execute a workload even on an execution environment that is not to be classified as trusted from the outset.
[0005] This object of the invention is achieved with a method having the features specified in claim 1 and with a system having the features specified in claim 10. Preferred developments of the invention are described in the associated dependent claims, the subsequent description and the drawings.
[0006] In the case of the method according to the invention for executing a workload in an execution environment, the workload is permitted for execution by means of a permissibility list of permitted workloads, wherein in the case of the method risk information characterizing an IT security risk of the execution environment is determined and the permissibility list of permitted workloads is changed as a function of the determined risk information. The method is particularly preferably a computer-implemented method.
[0007] It goes without saying that an admissibility list of allowed workloads means an admissibility list of allowed workloads, ie an admissibility list having allowed workloads listed in the list.
[0008] The IT security risk of an execution environment may preferably be understood in particular as an IT security risk caused by the execution environment for executing the workload and / or may suitably characterize a host trust state.
[0009] In the case of the method according to the invention, the IT security risk of the execution environment is preferably determined by means of a software-based integrity verification module. That is, the integrity of the execution environment is preferably verified by using an integrity verification module based on software, and risk information is determined based on software according to the verified integrity. Suitably, high integrity means low IT security risk, and vice versa.
[0010] According to the present invention, it is therefore not necessary to believe that the execution environment is trustworthy from the outset, but according to the present invention, only when the trustworthiness of the execution environment has been determined and verified, if then the permissibility list for the allowed workload has been selected according to the verified trustworthiness, and the workload is subsequently included in the permissibility list, the workload is executed in the execution environment. Therefore, in the case of the method according to the present invention, the "zero trust" paradigm (Paradigma) of network security is advantageously transferred to the execution of the workload in the execution environment. Therefore, a permissibility list adapted to the risk of the execution environment and therefore to the reliability of the execution environment can be used by means of the method according to the present invention. Suitably, the workload is a software component. The execution environment is preferably a computing platform, in particular an open computing platform. Exactly in the case of an execution environment that is constituted by means of an open computing platform and therefore contains a higher risk for manipulation or attack, it is advantageously possible to continuously adapt the permissibility list to the risk information of the computing platform by means of the method according to the present invention. In the case of the method according to the present invention, the workload is preferably executed using a changed permissibility list. In particular, the workload comprises a control functionality of a manufacturing or maintenance process, so that the control functionality can be executed by means of the method according to the invention and the manufacturing or maintenance process can be controlled and executed.
[0011] Preferably, in the case of the method according to the invention, the workload comprises one or more of the elements of the following list: a virtual machine, an operating system container, in particular a Linux LXC or Docker container or a container-based application, a native operating system application, a service, in particular a Unix daemon Or a Windows service, an operating system module, preferably an operating system module dynamically loadable by the Linux kernel.
[0012] The aforementioned software components can be stored, preferably installed, on the execution environment, or can be loaded from a connected network storage into the execution environment for execution. According to the invention, the software components are only started, i.e. executed and / or loaded into the working memory, if they are defined as permitted by the permissibility list.
[0013] In the case of the method according to the invention, the risk information simultaneously represents the host trust status of the execution environment, because the higher the risk indicated by the risk information for the execution, the less trustworthy the execution environment is.
[0014] In the case of the method according to the invention, preferably in the ongoing operation of the execution environment, the corresponding risk information of the execution environment is determined, and the permissibility list is changed, in particular selected or constructed accordingly. In the case of selecting a permissibility list, there are multiple permissibility lists, one of which is selected according to the risk information of the execution environment. In the case of constructing a permissibility list, a permissibility list that depends on the determined risk information of the execution environment is determined so that only such workloads that are compatible with the determined risk information of the execution environment enter the permissibility list. Thus, a database can be maintained, which indicates for each workload, that is, for each software component: in the case of which risk information of the execution environment the workload can enter the permissibility list, and in the case of which risk information it cannot enter. The risk information suitably includes a numerical value, and for each workload, it is noted in the database: from which numerical value or until which numerical value the workload can be recorded in the permissibility list. In this way, the permissibility list can be generated simply and automatically.
[0015] In the case of the method according to the invention, the already started workload or those software components are suitably stopped or terminated when the execution of that already started workload or those software components is no longer allowed according to the updated permissibility list. Thus, the already started software components can be stopped, terminated or migrated to another host depending on whether the permissibility list is maintained or has been changed.
[0016] The permissibility list preferably contains authorization information, i.e. information about which runtime authorizations are granted to the workload, in particular the corresponding software components, such as Linux capabilities, mandatory access control policies or access authorizations to specific hardware components of the computing server, such as in particular trust anchors for storing and providing cryptographic keys and functions or to graphics accelerators or to AI accelerators.
[0017] In a preferred development of the method according to the invention, the execution environment comprises a runtime environment, in particular for workloads in the form of software components, preferably containers and / or applications.
[0018] In the case of a method according to a further development of the invention, the execution environment preferably comprises an operating system and / or at least one input / output module and / or a signal connection and / or a switch and / or a router.
[0019] In the case of the method according to the invention, the risk information is expediently determined with the aid of the execution environment or with the aid of a device in which the execution environment is located or with the aid of software running in the execution environment.
[0020] The risk information of the execution environment can be determined on the execution environment itself, in particular by means of a dedicated software component, which is in particular executed in a "trusted execution environment" (abbreviated as "TEE") of the execution environment, suitably Intel SGX or Arm-TrustZone; or by means of a hardware extension module, such as a plug-in card, such as in particular a PCIe card; or by means of an integrated hardware component of the execution environment; or by means of a backend / edge cloud service. Alternatively or additionally, the risk information of the execution environment can be determined by means of a device or platform that is communicatively connected to the execution environment.
[0021] In the context of the method, in a preferred development of the invention, the permissibility list is a positive list, ie the software components that are permitted for execution are entered in the permissibility list.
[0022] Suitably, in the case of the method according to the invention, the permissibility list is cryptographically protected. In the case of this improved scheme, the permissibility list is advantageously not easily changed accidentally or due to manipulation attempts by an attacker. Therefore, the method according to the invention is particularly safe in this improved scheme. The cryptographic protection of the permissibility list can be carried out in particular by means of a digital signature of the permissibility list, for example on the side of a permissibility list adaptation module that may exist as described below. In this way, an attacker who wants to manipulate the permissibility list must not only manipulate the permissibility list, but also the digital signature or the permissibility list adaptation module that may exist. Alternatively or additionally, the permissibility list can be maintained in an encrypted manner by means of a cryptographic key, so that the permissibility list can only be implemented when the cryptographic key or the private key belonging to the cryptographic key is manipulated or damaged at the same time. In particular, the permissibility list can be protected by means of a cryptographic check value, such as a cryptographic checksum or by means of authenticated encryption. The permissibility list is preferably in the form of a so-called "verifiable credential" or "verifiable expression". Alternatively or additionally, the permissibility list may be stored in a distributed database, in particular a distributed ledger or a blockchain, or the check value of the permissibility list may be stored in a distributed ledger or a blockchain.
[0023] Alternatively or additionally, the permissibility list is protected by means of a cryptographically protected communication channel, preferably by means of TLS and / or HTTPS.
[0024] In the case of the method according to the invention, the first version of the permissibility list can preferably be determined when the device of the execution environment starts up, in particular when booting. For this purpose, the permissibility list can be loaded from a non-volatile memory or from a server.
[0025] During the ongoing operation of the execution environment, the permissibility list is preferably repeatedly updated, i.e. the method according to the invention is preferably repeatedly executed. This can be done in particular in a time-controlled manner, preferably at constant time intervals or alternatively and also preferably at specific events, in particular at configuration changes or changes to the installed software, when installing patches or when a user, in particular an administrator or a privileged user, for example a user with root privileges, logs in or out. If an update does not take place within a specified time period or a validation of the validity of the permissibility list does not take place, a locally stored backup permissibility list can optionally and advantageously be provided.
[0026] Preferably, in the case of the method according to the invention, in addition to the determined risk information of the execution environment, risk information of other components that are communicatively connected to the execution environment can also be determined and taken into account when forming the permissibility list of the execution environment. The communicatively connected components can be, in particular, input / output modules via which control applications executable in the execution environment can access physical sensors and / or actuators directly or by means of a data network, preferably by means of Ethernet TSN, by means of 5G mobile radio, HART, Profinet, OPC UA.
[0027] The system according to the invention having an execution environment is configured to execute workloads in the execution environment. In the case of the system according to the invention, there is a permissibility list of permitted workloads, and the system is configured to execute the workloads according to the permissibility list. Additionally, the system according to the invention has a risk determination module for determining risk information characterizing an IT security risk of the execution environment and a permissibility list adaptation module, the permissibility list adaptation module being configured to change the permissibility list of permitted workloads according to the risk information.
[0028] The system according to the invention is expediently designed as a one-piece or integral device or as a one-piece or integrally operable device.
[0029] The system according to the invention is particularly preferably designed to carry out the method according to the invention as described above, in particular repeatedly or continuously. The system according to the invention has the corresponding advantages as already described and explained for the method according to the invention.
[0030] The method according to the invention as described above is preferably performed using the system according to the invention as previously described. BRIEF DESCRIPTION OF THE DRAWINGS
[0031] The present invention will be explained in more detail below based on the embodiments shown in the accompanying drawings.
[0032] Figure 1 An embodiment of a system for executing a workload in an execution environment according to the present invention is schematically shown in a simplified principle diagram, wherein the system is used to execute a method for executing a workload in the execution environment according to the present invention, and
[0033] Figure 2 The sequence of the method according to the invention is schematically shown in a flow chart. DETAILED DESCRIPTION
[0034] exist Figure 1 In the case of the automation system according to the invention shown in FIG, the open computing platform CS has a runtime environment APPRTE, in which a plurality of applications APP are executed, in the embodiment shown, Linux containers or virtual machines.
[0035] The open computing platform CS also has an operating system OS and some hardware components, in the illustrated embodiment a processor CPU, a working memory RAM, a hard disk storage SSD, a network interface NWIF, a hardware trust anchor TPM, a graphics accelerator GPU and an AI accelerator IFE.
[0036] One of the applications APP implements control device functionality, for example in the form of a virtual storable programmable component, which accesses sensors S and actuators A by means of a control network CN and by means of an input-output module IOM. The application APP can only be executed, i.e. started and loaded into the working memory RAM, if it is allowed according to a software component whitelist APPWL stored in the execution environment. The execution environment is implemented here with the aid of a runtime environment APPRTE. The application whitelist APPWL is repeatedly updated during ongoing operation based on risk information that describes the IT security risk of the runtime environment APPRTE. The risk information simultaneously describes a high degree of trustworthiness of the runtime environment APPRTE in the case of low IT security risks. Conversely, a high IT security risk describes a low degree of trustworthiness of the runtime environment APPRTE. In this regard, the risk information is equivalent to the trust status information of the runtime environment APPRTE.
[0037] In addition to the risk information of the runtime environment APPRTE, in the case of the automation system according to the invention, the risk information of the IO modules IOM and the risk information of the sensors S as well as the risk information of the actuators A and / or the control network CN or individual network components of the control network CN, such as switches and / or routers and / or cabling and / or WLAN access points and / or 5G mobile radio base stations, can be taken into account. This is because in addition to the runtime environment APPRTE, the above-mentioned other elements also belong to the execution environment of the open computing platform CS.
[0038] In the case of the automation system according to the invention, the aforementioned risk information is continuously determined and the software component whitelist APPWL is respectively adapted as a function of the risk information.
[0039] This takes place at the computing platform CS as follows: With the aid of a software-based integrity check module HIC, the integrity of the computing platform CS is checked and risk information is determined. High integrity means low IT security risk and vice versa. Based on the determined risk information, the software component whitelist APPWL is adapted at the computing platform CS with the aid of a whitelist adaptation module AWA. The application launcher AL, which intends to start the application APP as described above, first checks with the aid of the query OK? to the check module AWC: Is the application APP included in the software component whitelist APPWL as permitted for execution in the runtime environment APPRTE. The check module AWC transmits the answer Y / N back to the application launcher AL. If the application APP is contained in the software component whitelist APPWL, the application APP is started by the application launcher AL.
[0040] Other critical applications APP that are executed only based on the trust status of the host, i.e. permitted through a respectively correspondingly adapted software component whitelist, are, for example, encryption applications for encrypting or decrypting data or network transmissions, applications for onboarding or provisioning other devices, or applications for training AI models.
[0041] exist Figure 2 In the example, the method is again succinctly illustrated in a flow chart:
[0042] The method is triggered in an initiation step STAR. This initiation step STAR can be performed regularly, for example at constant time intervals, or in a manner adapted to requirements, for example in the event of a change in the execution environment or in the device in which the execution environment is implemented.
[0043] In a determination step DETTRU, risk information of the execution environment is determined, which quantifies the IT security risk of the execution environment, in the described exemplary embodiment by means of an integrity checking module HIC.
[0044] According to the risk information, a software component whitelist APPWL belonging to the risk information is selected for the allowed workload, the allowed application APP that can be executed in the runtime environment APPRTE in the embodiment shown. To this end, in the embodiment shown, the risk information constitutes a risk metric value, which indicates a large risk when the risk metric value is high, and indicates a low risk when the risk metric value is small. The range of values that the risk metric value can take is divided into value intervals. A specific software component whitelist APPWL is assigned to each value interval, so the software component whitelist APPWL can be easily determined according to the risk metric value. All software component whitelists APPWL in different value intervals are different from each other. The higher the risk metric value of the value interval, the fewer entries the software component whitelist has. As the risk metric value increases, the entries that still exist are taken out from the software component whitelist APPWL of the value interval that is respectively in the front in the increasing direction. With the help of this allocation criterion, the allocated software component whitelist APPWL is now determined according to the risk metric value in the determination step DETWL.
[0045] In the exemplary embodiment shown, all software component whitelists APPWL are cryptographically protected and are additionally signed by the user of the computing platform CS by means of a digital signature.
[0046] With the aid of the software component whitelist APPWL thus determined, it is now possible to respond appropriately to incoming queries RECREQ for executing workloads: at each query RECREQ, the workload in question is respectively compared with the software component whitelist APPWL determined with the aid of the determination step DETWL for executing the workload in question, and in a verification step ADMIS it is checked whether the workload in question is included in the software component whitelist APPWL as permitted with an entry. If this is confirmed, the workload in question is executed in an execution step EXEC with the aid of the application launcher AL. If the workload in question is not included in the software component whitelist APPWL as permitted with an entry, the verification step ADMIS ends with a denial of execution, and the method ends with an end step END without the execution step EXEC.
[0047] In the exemplary embodiment shown, the application launcher AL additionally checks whether the software component whitelist APPWL has the required cryptographic protection and whether the digital signature of the software component whitelist is valid.
[0048] It goes without saying that in other embodiments not specifically shown, the software component whitelist APPWL does not have to be provided in the form of a positive list of allowed workloads, but instead an exceptional blacklist of allowed workloads can be provided which is adapted according to the risk information.
Claims
1. A method for executing a workload (APP) in an execution environment (APPRTE, IOM), wherein the workload is permitted for execution with the aid of a permissibility list (APPWL) of permitted workloads, wherein in the method risk information characterizing an IT security risk of the execution environment is determined (DETTRU) and, as a function of the determined risk information, the permissibility list of permitted workloads is changed (DETWL).
2. A method according to the preceding claim, wherein the workload can include one or more elements from the following list: a virtual machine, an operating system container, in particular a Linux (LXC) or Docker container or a container-based application, a native operating system application, a service, in particular a Unix Demon or Windows service, an operating system module, preferably an operating system module that can be dynamically loaded through the Linux kernel.
3. A method according to any of the preceding claims, wherein risk information of the execution environment (APPRTE, IOM) is determined during ongoing operation of the execution environment (APPRTE, IOM), and the permissibility list is changed, in particular selected or constructed, based on the determined risk information.
4. The method according to any of the preceding claims, wherein the execution environment (APPRTE, IOM) comprises a runtime environment (APPRTE), in particular for workloads in the form of software components, preferably containers and / or applications (APP).
5. The method according to any of the preceding claims, wherein the execution environment (APPRTE, IOM) comprises an operating system (OS) and / or at least one input-output module (IOM) and / or at least one signal connection and / or at least one switch and / or at least one router.
6. A method according to any one of the preceding claims, wherein the risk information is determined with the aid of the execution environment (APP, IOM) or with the aid of the device (CS) in which the execution environment is located or with the aid of software running in the execution environment, or wherein the risk information is determined with the aid of a component that is communicatively connected to the execution environment.
7. Method according to any of the preceding claims, wherein the permissibility list (APPWL) is a positive list.
8. The method according to any of the preceding claims, wherein the permissibility list (APPWL) has at least two or more entries of allowed workloads.
9. The method according to any of the preceding claims, wherein the permissibility list (APPWL) is cryptographically protected.
10. A method according to any of the preceding claims, wherein the risk information is determined (DETTRU) repeatedly, in particular at time-constant intervals or in an event-controlled manner, and the permissibility list of permitted workloads is changed (DETWL) in each case as a function of the determined risk information.
11. A system having an execution environment (APPRTE, IOM) for executing a workload (APP), wherein there is a permissibility list (APPWL) of allowed workloads (APP), and wherein the system is constructed to execute the workload according to the permissibility list (APPWL), the system comprising a risk determination module (HIC) for determining risk information characterizing IT security risks of the execution environment (APPRTE, IOM) and a permissibility list adaptation module (AWA), the permissibility list adaptation module being constructed to change the permissibility list (APPWL) of allowed workloads (APP) according to the risk information.
12. System according to any of the preceding claims, which constitutes a device (CS) which is or can be operated in one piece or in one piece. 13 . The system according to claim 11 , which is designed to carry out the method according to claim 1 .
