A computer-implemented method and system for operating at least one application in a trusted computing enabled platform and a corresponding application
The method and system address the challenge of managing TEEs across diverse platforms by enabling flexible deployment of microservices based on security and performance needs, optimizing their placement and management, and aligning attestation with deployment strategies, thus enhancing the operational efficiency of applications across different environments.
Patent Information
- Application Number
- PCT/EP2023/087914
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-05-12
- Filing Date
- 2023-12-28
- Publication Date
- 2024-11-21
AI Technical Summary
Current solutions lack a standard interface for developing and operating Trusted Execution Environments (TEEs) across different platforms, such as Intel SGX and RISC-V, leading to challenges in flexible and automated deployment, life cycle management, and attestation of applications using TEEs, particularly for cloud providers and 5G deployments.
A computer-implemented method and system that enables flexible deployment of microservices in either trusted or untrusted environments based on security, performance, and platform characteristics, using a Virtual Infrastructure Manager with a Secure Infrastructure Manager to optimize the placement and management of microservices across TEEs and Rich Execution Environments.
This approach allows for efficient operation of applications by enabling adaptable deployment and life-cycle management of microservices across various platforms, aligning attestation with optimized deployment strategies, and providing a single point of control for orchestrating inter-dependent software components in both trusted and untrusted environments.
Smart Images

Figure EP2023087914_21112024_PF_FP_ABST
Abstract
Description
[0001] A COMPUTER-IMPLEMENTED METHOD AND SYSTEM FOR OPERATING AT LEAST ONE APPLICATION IN A TRUSTED COMPUTING ENABLED PLATFORM AND A CORRESPONDING APPLICATION
[0002] The present invention relates to a computer-implemented method for operating at least one application in a trusted computing enabled platform,
[0003] Further, the present invention relates to a corresponding system and to a corresponding application
[0004] Corresponding prior art documents are listed as follows:
[0005] [1] https: / / www.etsi.org / deliver / etsi_gr / NFV-SEC / 001_099 / 011 / 01 .01 ,01_60 / gr_nfv- sec011v010101 p.pdf
[0006] [2] https: / / www.etsi.org / deliver / etsi_gs / nfv-sec / 001_099 / 012 / 03.01.01_60 / gs_nfv- sec012v030101 p.pdf
[0007] [3] https: / / www.etsi.org / deliver / etsi_gs / nfv-sec / 001_099 / 013 / 03.01.01_60 / gs_nfv- sec013v030101 p.pdf
[0008] [4] https: / / datatracker.ietf.org / wg / teep / about /
[0009] Further prior art documents:
[0010] “Secure Cloud Micro Services Using Intel SGX”, May 24, 2017, Stefan Brenner, Tobias Hundt, Giovanni Mazzeo, Rudiger Kapitza: This document discloses a method for integration of trusted execution based on Intel Software Guard Extensions, SGX, into micro service applications. The method involves Vert.x Vault, that supports SGX-based trusted execution in Eclipse Vert.x, a renowned tool-kit for writing reactive micro service applications, where secure micro services can run alongside regular ones and are inter-connected via the Vert.x event bus to build large Vert.x applications that can contain multiple trusted components, thus, providing the benefits of micro service architectures together with trusted execution to support privacy and data confidentiality for sensitive applications in the cloud at scale. “S-Blocks: Lightweight and Trusted Virtual Security Function with SGX”, January 2019, Juan Wang, Shirong Hao, Hongxin Hu, Bo Zhao, Hongda Li, Wenhui Zhang, Jun Xu, Peng Liu, Jing Ma: This document discloses S-Blocks, an architecture to modularize a virtual security function, VSF, and protect its key modules with SGX in an efficient manner. S-Blocks leverages a modular and micro-service-oriented architecture to design VSFs. It decomposes modules of VSFs to trusted elements and untrusted elements and provides proprietary APIs. Further, the prior art addresses state consistency and secure migration issues of security function scaling by designing a fine-grained state synchronization and migration mechanism to ensure lose-free, order-preserving, and state security for VSFs. Moreover, S-Blocks only imposes a manageable performance overhead, and low latency and resource consumption when protecting VSFs.
[0011] Trusted execution environments, TEEs, or enclaves, are sometimes defined as special secure areas of the main processor that guarantee certain properties to the code and data loaded on them. Such properties include for example integrity, i.e. the TEE prevents unauthorized entities from replacing or modifying protected code or data or confidentiality, i.e. unauthorized parties cannot read data inside an enclave. TEEs require special hardware and software to operate them, and applications need certain modifications to be able to run inside a TEE. There are various commercial and non-commercial solutions that implement them, e.g. Intel SGX, ARM Trustzone, AMD SEV, Sanctum or Keystone.
[0012] The root of trust, RoT, usually Hardware, is the foundation on which all the operations of a secure system depend. In the context of TEEs, the combination of hardware and software that provides the services to the TEEs according to some specification is known as the RoT. It ensures the integrity of the hardware, that the software of the applications running inside TEEs and its data has not been compromised and provides a mechanism to external verifiers to verify such state via attestation. As its name indicates, the RoT is inherently trusted and should be secure by design, i.e. it is the component of the platform where more efforts are spent for its formal verification. However, there is no standard interface to develop code for TEEs regardless of the platform where this code runs, e.g. Intel SGX or RISC-V based platforms, nor to operate them, i.e. each manufacturer establishes a different way to interact with the TEEs. This also poses a problem for cloud providers or 5G deployments, as currently there is no solution that allows flexible and automated deployment, life cycle management of applications or micro-services making use of TEEs. Additionally, no solution permits orchestrators of hypervisors to leverage features offered by the TEE platforms, such as Attestation, in order to - in turn - offer those to their own clients.
[0013] In the following it is provided a summary of the state of the art, some background, key scenario and problem statement:
[0014] ■ Today, physical and virtual resources in 5G or 6G scenarios are managed and orchestrated, MANO, per ETSI NFV guidelines and specifications o Open Source Software, OSS, exists and is being further developed, e.g. Open Source MANO, OSM, ONAP o Management of virtualized resources includes Virtual Machines, VM, on top of a Virtual Machine Monitor, VMM, also denoted as Hypervisor, as well as containers on top of a container runtime, such as docker and kubernetes
[0015] ■ As emerging technology to increase trust levels for sensitive software and associated data, Trusted Computing gets more traction and offers Trusted Execution Environments, TEE, where code and data can be stored and executed in separated runtime environments while being isolated in storage and compute from the rest of the environment, where traditional Operating Systems, OS, and applications are deployed and executed. o Installation of code in a TEE is performed by dedicated software frameworks such as keystone enclave for RISC-V platform and ENARX for, for example, Intel SGX and container runtime. Each platform exposes a different framework for deployment and attestation, which means a proof against a trusted source that the deployed code is original and not tampered. o Code in a TEE might be deployed altogether with a customized Runtime, RT, that performs some task instead of the OS, and only exposes an Application Programming Interface, API, to invoke its functionality. Similarly, TEE code can only access certain functions running outside the enclave that are specified at compile time. Therefore, TEE applications are consequently associated with peering code running in the untrusted environment of a computer platform, which is also denoted as Rich Execution Environment, REE.
[0016] ■ Today, MANO operations of functions and software on a compute platform and deployment / attestation of trusted applications and their runtime in a TEE are handled by different software frameworks and are not compatible.
[0017] ■ Trusted software is designed and developed in a customized way, taking the target platform, key performance indicators, KPI, and security / privacy related aspects, e.g. key management, treatment of sensitive data, mission critical code, etc., into account. Once an application has been developed for one particular target TEE platform, it cannot be migrated to another platform neither modified.
[0018] ■ There are attempts, such as OpenEnclave, of developing a common SDK for the development of TEE applications by creating an abstraction layer between the developer and the underlying platform.
[0019] ■ Open is the following:
[0020] 1. Design and development of an application as micro-services for flexible distribution and automated joint deployment in a TEE and REE.
[0021] 2. Joint deployment, life-cycle management and attestation of software and functions on - cloud - platforms and on trusted computing enabled environments. Such collaborating and / or dependent distributed software components / functions, comprising untrusted and trusted software components are denoted in the following as 'distributed trusted and untrusted micro-services ’ 3. Flexible and automated deployment of an application’s micro-services in a TEE or REE taking statics, e.g. platform KPIs, security / privacy aspects, and dynamics, e.g. capacity constraints, load, failure, etc., into account. Such procedure and technology is in the following denoted as ‘Trust Orchestration enablers’
[0022] These three open points refer to the current state of application development for TEEs, to the visibility that cloud operators or remote management systems have regarding these environments and finally to the way applications running inside TEEs or containing parts that run inside a TEE can be handled and managed. We provide more information about each of these items in the following.
[0023] 1. Development of trusted applications (design)
[0024] In the following, it is elaborated more in the different aspects of today’s planning and deployment of trusted applications for different TEEs.
[0025] Each TEE designer makes a choice regarding the techniques that are going to be used to protect the assets, regarding the threat model and the interface that can be used to access the features they offer. As an example, Intel SGX only considers the Hardware as trusted, the Operating System or the orchestrators are excluded. And it offers some services that can be called by the code running inside the TEEs via what they call platform enclaves, which in the end are just some special TEEs. Keystone on the other hand, might directly expose an interface to other parties, trusted or untrusted. Thus, when an application is designed for any of these environments, it has to be aware of the functions or interfaces that each platform exposes.
[0026] To better understand the development and deployment of applications for TEEs, it is worth to mention that these environments usually have some limitations regarding, for example, access to peripherals, the amount of memory they can use or the functions they can invoke. If an existing application needs to run inside a TEE, it has to be partitioned and it is up to the developer to specify which parts can be executed in the REE and which ones have to go to the TEE. The process is similar for applications designed from scratch. Splitting application code between trusted and rich application components is complex because of the compromise between security, performance and functionality requirements. As mentioned before, this decision is up to the developer, and after the decision is made, he / she usually can attach metadata regarding some of the requirements of the TEE, such as RAM memory or running threads to the final application package.
[0027] In a nutshell, developing trusted and rich applications for Trusted Execution Environments, TEEs, is challenging for several reasons including:
[0028] • Meeting security vs performance vs cost challenges
[0029] • Interoperability among different TEE systems
[0030] • It requires specialized skills and compatibility with different TEE platforms
[0031] • Testing TEE applications at runtime is not easy without partially compromising Security
[0032] The code that runs inside the TEE is sometimes referred as Trusted Application, TA, whereas the one that runs on the rich environment is referred as Rich Application, RA. The current approach for developing TAs and RAs includes the following steps:
[0033] • Analysis to decide which code should go to the TA and RA
[0034] • Design taking into account the dependencies between the components and the communication between the TA and RA
[0035] • Development taking in to account the Security performance and functional requirements
[0036] • Deployment to the Rich and Trusted Environment
[0037] • Attestation verify the deployment
[0038] In the following, it is elaborated a little bit more on the description of the aforementioned steps.
[0039] • Design the TA: o The first step in TA development is to define the security and performance requirements of the application. This includes identifying the data that needs to be protected, the operations that the application needs to perform, and the hardware and software resources that the Trusted application can access.
[0040] • Develop and test the TA: o The next step is to develop and test the TA for a TEE platform. This includes writing and debugging the code, testing for performance, and ensuring that the application meets the required security standards. o Once the TA has been developed and tested, i.e. it is ready to be deployed, it needs to be compiled, then the developer obtains a measurement, e.g. a hash, that identifies the TA binary and finally it is signed with a digital signature to ensure its authenticity and integrity. The signature can be verified later during attestation. Additionally, depending on the procedure followed to sign the binaries, it also ensures that the TA has not been tampered with.
[0041] • Deploy the TA: o The TA can be deployed in the TEE environment using the appropriate deployment tools. This may involve loading the TA into the TEE, installing it on a device, or uploading it to a cloud-based service. In general the OS or the orchestrator takes care of the deployment but such deployment is later verified by the Security Monitor, SM, in case of RISC-V processors or by the firmware of Intel SGX or by the TEE of choice.
[0042] • Attestation: Attestation is the process of verifying the integrity and authenticity of the TA. Attestation procedures may vary depending on the TEE platform, but generally involve the following steps: o The TA requests to the firmware or software that is part of the RoT a set of measurements that describe the initial state of the TA and the TEE environment, e.g. details about the platform or software version of the firmware / software supporting the TEE. o The measurements are signed by this privileged firmware with its digital signature to ensure their authenticity and integrity. Usually they also include a platform unique signature. o The signed measurements are sent to a remote verifier, which validates the signature and the measurements by comparing them with the ones the verifier obtained before. That is, the measurements should be exactly the same that were obtained when compiling the application. o If the measurements and signature are valid, the verifier has a proof that the TA is running in a specific trusted environment and that it has not been tampered with.
[0043] The attestation mechanism is similarly decided by the manufacturer. For example, in processors that support Intel SGX, a key burned in the processor at manufacturing time is used to sign the proofs, and for the developers it becomes necessary to use Intel services to validate these signatures, i.e. Intel keeps a register of the public keys corresponding to the private keys of the SGX enabled processors. Other technologies, such as TEEs on RISC-V processors controlled by Security Monitors, SMs, which are the most privileged software running in a RISC- V based processor, such as Keystone, usually sign the proofs for attestation with the key of the RoT accessible only to the trusted SM and may enhance it with some information regarding the SM and sign them again with the SM key.
[0044] Additionally, TEEs, e.g. Intel SGX, usually use the application binary measurement - hash - as part of the attestation proof. Similarly, they offer to developers the option to sign their applications so on the one hand the TEE refuses to execute the app if there is a mismatch between the developers signed application and the application the TEE is trying to launch and on the other hand, some information regarding the developer can be included in the attestation proof. Therefore, in the attestation process the developer validates its own signature and gets confirmation that the device signed proof is legit.
[0045] This explanation just provides background on how TAs are developed and the whole design process until they are deployed and are fully functional. This process has not considered an orchestrator managing the TEEs, applications that are composed of multiple micro-services with some of them running inside a TEE, need to be designed in a different way. In the following we sketch some of the reasons why managing the life cycle of an enclave or of applications comprised of multiple microservices is challenging: • Security: due to the various threats to the confidentiality, integrity, and availability of the data being processed within the enclave
[0046] • Performance and Compatibility: Security comes at the cost of Performance, technology variations for TEEs makes it difficult for compatibility
[0047] • Upgrades and maintenance: Is difficult because preserving the enclaves relationship with other systems and applications is challenging as their IDs are usually related with their code, e.g. the ID is sometimes derived from the hash of the application
[0048] • Debugging: is challenging because understanding what is going on in the enclave without violating the properties of the enclave or TEE is not easy
[0049] • Scalability and compliance: as data to be processed by enclave increases so does complexity, ensuring security compliance and scalability of an enclave is not trivial especially when we have more than one stakeholder
[0050] Fig. 1 depicts current state of the art for developing applications for Trusted Execution environments with remote attestation and depicts current state of the art for deploying an application in a trusted and rich environment. Further, Fig. 2 demonstrates the current state of the art for developing, packaging, signing and deploying Trusted Execution Environment applications. As shown in the figure it follows a static approach for splitting the application in to Trusted, Untrusted and Run-time part. Example pseudocode implementation for Fig. 3 could follow the following pattern:
[0051] # Define security and performance requirements for TA development def define_requirements(): protected_data = ... required_operations = ... hardware_resources = ... software_resources = ...
[0052] # proceed to TA development
[0053] # Develop and test the TA fora TEE platform def develop_and_test_ TA(): code = ... debug_ code(code) test_performance(code) ensure_ security_ standards(code)
[0054] # Compile and sign the TA with a digital signature for authenticity and integrity def compiie_and_sign_ TA(): compiled- TA = compiie(code) digitai_signature = sign(compiied_ TA) verify_ signature(digitai_ signature)
[0055] # Deploy the TA using the appropriate deployment tools def deploy_ TA(): deployment-tool = ... deploy_ TA_ with_ tooi(depioyment_ tool)
[0056] # Attestation - Verify the integrity and authenticity of the TA def verify_ TA_integrity():
[0057] TA_binary = get_ TA_binary()
[0058] TA_hash = hash(TA_binary)
[0059] TA_signature = sign(TA_binary)
[0060] TA_measurements = get_ TA_measurements()
[0061] TA_measurements_signature = sign(TA_measurements) verification_resuit=verify_signature(TA_signature) and compare_measurements(TA_measurements, TA_measurements_signature) return verification_resuit
[0062] 2. Distributed trusted and untrusted micro-services (visibility)
[0063] As for the current state of TEE applications development, one limitation for deployments is that once an application has been developed for a specific environment, the design is fixed and that application can only be launched in that particular type of TEE. In other words, if the application is comprised of multiple micro-services, e.g. 5G Network Function deployments, such micro-services can only run as a TA or as a RA regardless of the conditions of the platform where they are running. This hampers the ability of cloud providers or orchestrators from migrating the applications or the micro-services to different platforms of different environments, e.g. between the RA and the TA.
[0064] Another limitation, is that while current resource managers or orchestrators already are aware and can leverage features exposed by regular processors, i.e. processors that do not have hardware support for TEEs, to control the applications and microservices running on them, they do not have such visibility or control on the TEE hardware / software resources. In other words, some layer or interface that allows such awareness is missing.
[0065] 3. Trust Orchestration enablers (operation)
[0066] Joint Management and Orchestration, e.g. resource selection, software function deployment, monitoring, scaling, teardown, etc., of rich and trusted applications across different types of TEE platforms is still not considered in current solutions for cloud or 5G deployments. These solutions focus on Managing and Orchestrating either TAs or RAs for a specific TEE platform.
[0067] The ETSI NFV has been actively working on the standardization of TEEs for virtualized network functions, VNFs.
[0068] ETSI NFV has published specifications related to TEEs, including:
[0069] • ETSI GS NFV-SEC 011 : "Trusted Execution Environment (TEE) for NFV: Requirements and Architecture" [1]
[0070] • ETSI GS NFV-SEC 012: "Trusted Execution Environment (TEE) for NFV: Security Assessment" [2]
[0071] • ETSI GS NFV-SEC 013: "Trusted Execution Environment (TEE) for NFV: Security and Privacy Threats and Mitigation Techniques" [3]
[0072] These specifications provide a comprehensive overview of the requirements, architecture, assessment, and mitigation techniques for TEEs in the context of NFV. However, they do not explicitly target joint orchestration and LCM of TAs and RAs. Furthermore, the orchestration and LCM of trusted / rich applications across different types of platforms is not considered. Some of the limitations of current approaches include the following:
[0073] • Lack of common control and standardized APIs for integrating TEEs with ETSI NFV / MEC
[0074] • Limited coverage of hardware-based isolation o ETSI NFV mainly focuses on software-based isolation mechanisms, such as hypervisors for virtual machines and namespaces for containers
[0075] • Insufficient support for attestation of services made out of Trusted and Rich Applications
[0076] • While ETSI has standardized descriptors for VNFS, Services, Slices etc., applicable to various platforms, todays frameworks for TEE deployment are not standardized and they are platform dependent
[0077] Open Source Software, OSS, for MANO according to ETSI NFV specifications consider only orchestration and LCM of virtualized network functions, VNF, in a rich space where the host OS is operational.
[0078] Internet Engineering Task Force, IETF, for TEE systems is focused on the development of protocols and standards to support the secure deployment and management of TEEs.
[0079] The IETF has established a working group called the "Trusted Execution Environment Provisioning (TEEP) Working Group" [4] to focus on the development of the TEEP protocol. The goal of the TEEP protocol is to provide a secure and standardized way of remotely installing, updating, and deleting Trusted Applications, TAs, within a TEE.
[0080] The TEEP Working Group is currently working on the development of several standards and best practices for TEEs, including: • TEEP protocol: The TEEP protocol is a standardized way of remotely provisioning TAs within a TEE. The protocol defines a set of messages and procedures for securely installing, updating, and deleting TAs.
[0081] • TEE Attestation: The IETF is working on standards for attesting the security and integrity of TEEs, to ensure that TAs are executing in a secure environment.
[0082] • TEE Management: The IETF is developing standards for the management of TEEs, including the secure storage of TEE configuration data and the provisioning of TAs.
[0083] In summary, the current state of the art in the IETF for TEE systems is focused on the development of protocols and standards to support the secure deployment and management of TEEs. The TEEP Working Group is working to establish a common framework for TEEs that can be used across different platforms and devices.
[0084] However, Joint Management and Orchestration of rich / untrusted applications and trusted applications that are supported by different TEE technologies and platforms is still not considered. Current solutions focus on Managing and Orchestrating either TAs or RAs for a specific TEE platform.
[0085] Like ETSI ISG, IETF also has gaps in the following aspects:
[0086] • Lack of standards for integrating TEEs into existing protocols for security incl. Authentication, Authorization and Accounting protocols, AAA
[0087] • Insufficient guidance on secure TEE deployment o This can result in inconsistent deployment practices because different TEE platforms follow different application deployment and operation
[0088] • Limited support for solutions that provide hardware-based isolation o IETF focus on software based isolation mechanisms like ETSI
[0089] • Limited consideration for attestation o While the IETF has developed some attestation protocols, such as the Remote Attestation Procedure, RATS, there is still limited support for attestation in existing protocols. It is an object of the present invention to improve and further develop a computer- implemented method for operating at least one application in a trusted computing enabled platform and a corresponding system and application for providing an efficient operation of the at least one application by simple means.
[0090] In accordance with the invention, the aforementioned object is accomplished by a computer-implemented method for operating at least one application in a trusted computing enabled platform, wherein the platform comprises a trusted environment and an un-trusted environment, wherein the at least one application comprises multiple microservices, and wherein at least one of the multiple micro-services is designed as a flexible microservice, which can be deployed either in the trusted environment or in the un-trusted environment, comprising the steps:
[0091] - deciding on a deployment of the flexible micro-service in the trusted environment or in the un-trusted environment depending on at least one parameter of a requested security level of the at least one application and / or of a requested performance level of the at least one application and / or of a present or available platform characteristic or feature; and
[0092] - deploying the flexible micro-service in the trusted environment or in the un- trusted environment according to a result of the deciding step.
[0093] Further, the aforementioned object is accomplished by an application for operating in a trusted computing enabled platform, wherein the platform comprises a trusted environment and an un-trusted environment, wherein the at least one application comprises multiple micro-services, and wherein at least one of the multiple micro-services is designed as a flexible microservice, which can be deployed either in the trusted environment or in the un-trusted environment.
[0094] Further, the aforementioned object is accomplished by a system for performing a computer-implemented method for operating at least one application in a trusted computing enabled platform, comprising:
[0095] - a platform with a trusted environment and an un-trusted environment; - at least one application comprising multiple micro-services; wherein at least one of the multiple micro-services is designed as a flexible microservice, which can be deployed either in the trusted environment or in the un-trusted environment,
[0096] - deciding means for deciding on a deployment of the flexible micro-service in the trusted environment or in the un-trusted environment depending on at least one parameter of a requested security level of the at least one application and / or of a requested performance level of the at least one application and / or of a present or available platform characteristic or feature; and
[0097] - deploying means for deploying the flexible micro-service in the trusted environment or in the un-trusted environment according to a result of the deciding step.
[0098] According to the invention it has been recognized that it is possible to provide a very efficient operation of the at least one application by a suitable design of the at least one application. It has been further recognized that such a design is provided by an application which comprises at least one micro-service being designed as a flexible micro-service, which can be deployed either in the trusted environment or in the un- trusted environment. Such a flexible micro-service provides a high degree of adaptability to different platform types and environments. A deployment of such a flexible micro-service in the trusted environment or in the un-trusted environment can be decided depending on at least one parameter of a requested security level of the at least one application and / or of a requested performance level of the at least one application and / or of a present or available platform characteristic or feature.
[0099] Thus, on the basis of the invention an efficient operation of the at least one application by simple means as well as a corresponding system and application are provided.
[0100] According to an embodiment of the invention operating the at least one application can comprise control and / or orchestration and / or deployment and / or management of the at least one application, wherein control and / or orchestration and / or deployment and / or management of the at least one application can be performed easily at a single point. Thus, a single point of control and orchestration for the automated deployment of inter-dependent software components on a platform’s untrusted and trusted environments can be realized. An attestation procedure can be aligned with the selected and optimized deployment.
[0101] According to a further embodiment management of the at least one application can comprise life-cycle management of the at least one application, which simplifies the operation of the application.
[0102] In a further embodiment deployment of the at least one application in the platform can comprise deployment of its multiple micro-services, each in the trusted environment or in the un-trusted environment.
[0103] According to a further embodiment the deciding step can consider an automated and / or optimized deployment and / or management of the at least one application. Thus, a very efficient operation of the at least one application is possible.
[0104] According to a further embodiment the deciding step and the deploying step can provide a best match deployment of the flexible micro-service under consideration of at least two of the at least one parameter. As a result, and optimized deployment for an efficient operation of the at least one application can be provided.
[0105] In a further embodiment the platform characteristic or feature can comprise a static and / or a dynamic characteristic, wherein a static characteristic can include a type, a resource and / or a performance of the platform and wherein a dynamic characteristic can include a load and / or an availability of the platform. Thus, different characteristics or features can contribute to an optimized deployment decision for providing a very efficient operation of the at least one application.
[0106] According to a further embodiment the platform characteristic or feature can comprise a risk assessment, wherein the risk assessment can include a threat analysis and / or an attack potential. These kind of characteristics or features can contribute to a reliable and efficient operation of the at least one application. In a further embodiment the deciding step and the deploying step can be performed by means of a Virtual Infrastructure Manager, VIM, that includes a Secure Infrastructure Manager, SIM. As a result, a very reliable and efficient computation and enforcement of a best match deployment of one or more flexible micro-services of the at least one application is possible.
[0107] According to a further embodiment the deciding step and the deploying step can take into account a list of multiple descriptors that apply to a single application. State of the art NFV MANO can take into account only a single descriptor per state. This provides a very efficient operation of at least one application.
[0108] In a further embodiment the deciding step and the deploying step can be repeated for providing an updated deployment based on a change of at least one parameter of a requested security level of the at least one application and / or of a requested performance level of the at least one application and / or of a present or available platform characteristic or feature. This provides a suitable updated situation by a simple operation measure.
[0109] According to a further embodiment at least one micro-service can be designed under consideration of a code sensitivity, under consideration of a sensitivity of treated data or keys and an associated storage, under consideration of a type and variety of different platforms for later deployment and / or operation, and / or under consideration of costs of a deployment in a platform. This provides a high flexibility and adaptability of micro-services for providing an efficient operation of the at least one application.
[0110] In a further embodiment at least one micro-service can be classified as a microservice which has to be deployed in a trusted environment, which should be deployed in an un-trusted environment or which can be flexibly deployed in a trusted environment or in an un-trusted environment. According to such a classification an efficient operation of the at least one application is further simplified.
[0111] In a further embodiment at least one of the multiple micro-services can be designed for deployment only in the trusted environment or only in the un-trusted environment. According to a further embodiment the platform or system can comprise a common control and orchestration function comprising a Virtual Infrastructure Manager, VIM, and a Secure Infrastructure Manager, SIM.
[0112] In a further embodiment information and / or semantic can be added to an interface or interfaces of the control and orchestration function or of the platform for operation of the control and orchestration function or of the platform. The interface or interfaces can be a northbound interface toward service orchestration and / or a southbound interface toward a trusted computing platform.
[0113] According to a further embodiment the method can comprise one or more attestation mechanisms, wherein an attestation mechanism can comprise: attestation of micro-services that have been deployed in the trusted environment as a sub-package - group of micro-services that apply to the trusted environment - or individually per micro-service that applies to the trusted environment, or an option function for the attestation of micro services that have been deployed in the untrusted environment as a sub-package - group of micro-services that apply to the untrusted environment - or individually per micro-service that applies to the untrusted environment, or enabling the attestation of at least those micro-services that apply to the trusted environment regardless of the underlying hardware.
[0114] Advantages and aspects of embodiments of the present invention are summarized as follows:
[0115] Embodiments propose an efficient method for automated and optimized deployment and life-cycle management of an application made out of multiple micro-services, that include at least one flexible micro-service that can be deployed either in a trusted environment or an untrusted environment of a trusted computing enabled platform. Such embodiments can comprise the steps of
[0116] 1) computing and enforcing the best match orchestration decision for the deployment of each of the one or multiple flexible micro-service(s) in the trusted or untrusted environment 2) the best match characterized by an optimized tradeoff between the application’s level of security, its performance, as well as available platform characteristics, features and / or resources
[0117] 3) computation of the best match can include the trusted compute platform’s static characteristics, e.g. type, resources, performance, etc., and dynamic characteristics, e.g. load, availability, etc., as well as a current risk assessment, e.g. threat analysis, attack potential, etc.
[0118] Further advantages and aspects of embodiments of the present invention are summarized as follows:
[0119] 1) Design and development of an application based on a micro-service design, where one or multiple micro-services are flexible in their deployment in either the trusted environment or in the untrusted environment of a trusted computing enabled platform,
[0120] 2) Computation and enforcement of a best match deployment of an application’s one or more flexible micro-service(s) based on increased perception of the infrastructure, including a compute platform’s trusted environment by means of an extended Virtual Infrastructure Manager, VIM, that includes a Secure Infrastructure Manager, SIM,
[0121] 3) Computation and enforcement of an optimized deployment in terms of taking security, platform and environmental situation into account as well as the list of multiple descriptors - instead of single descriptor per state of the art NFV MANO - that apply to a single application. Common control, monitoring and orchestration for deployment and life-cycle management of the trusted service’s micro-services in both, the trusted and untrusted environments
[0122] Embodiments of the present invention can provide a single point of control and orchestration for the automated deployment of inter-dependent software components on a platform’s untrusted and trusted environments. An Attestation procedure can be aligned with the selected and optimized deployment.
[0123] A tailoring / customization of a deployment can be based on a variety of static and / or dynamic parameters. Further, options for closed-loop control operations and re-computation of an updated deployment based on changes parameters can be provided.
[0124] Further embodiments comprise a method to automate the deployment and life cycle management of applications made out of multiple micro-services, where some of the application’s micro-services are deployed in a trusted execution environment of a trusted computing-enabled platform, while others are deployed in an untrusted - rich - execution environment. Placement and instantiation decisions for the application’s micro-services are made based on various static and dynamic platform- and environmental criteria, aiming at a suitable match between service application security, performance and costs factors.
[0125] Within a further embodiment the platform characteristic or feature can comprise a cost estimation or cost factor for deployment and / or operation of the at least one application.
[0126] Further embodiments comprise an automated and optimized control, orchestration and life-cycle management of micro-service based applications in a trusted computing enabled platform.
[0127] There are several ways how to design and further develop the teaching of the present invention in an advantageous way. To this end it is to be referred to the following explanation of examples of embodiments of the invention, illustrated by the drawing. In the drawing
[0128] Fig. 1 shows in a diagram the current state of the art for developing and deploying an application in a trusted and rich environment,
[0129] Fig. 2 shows in a diagram a method according to an embodiment of the present invention, to classify applications into trusted - that must run in the trusted environment -, flexible - that can run both in the trusted and un-trusted environments - and rich / un-trusted - that should run in the un-trusted environments - micro-services, Fig. 3 shows in a diagram an image factory for various TEE platforms leveraging classified micro-services into trusted, un-trusted and flexible microservices and other requirements,
[0130] Fig. 4 shows in a diagram an example of initial deployment of trusted, un-trusted and flexible micro-services and operating flexible micro-services both in the trusted and un-trusted environments,
[0131] Fig. 5 shows in a diagram a method for attestation of the trusted applications according to an embodiment of the present invention,
[0132] Fig. 6 shows in a diagram a secure lifecycle management of trusted and un- trusted applications according to an embodiment of the inventive method for designing, developing, testing, packaging, deploying, remote attestation, common control and orchestration of the applications,
[0133] Fig. 7 shows in a diagram a trust orchestration engine for joint management and orchestration of trusted and un-trusted applications with remote attestation according to an embodiment of the present invention,
[0134] Fig. 8 shows in a diagram a closed loop operation for joint MANO operation according to an embodiment of the present invention,
[0135] Fig. 9 shows in a diagram a closed loop operation for joint cloud operation according to an embodiment of the present invention, and
[0136] Fig. 10 shows in a diagram an exemplary implementation and integration of the trust orchestration engine with VIM and SIM capabilities for joint orchestration of an application’s trusted and un-trusted micro-services.
[0137] In a nutshell, embodiments of this invention address the limitations of any system that tries to manage and control applications that make use of TEEs while trying to maintain certain Security guarantees. To this end, it is proposed a new way of categorizing micro-services as trusted, untrusted and flexible. From those, what is called in this invention flexible micro-services, can be deployed in both Trusted and Rich environments, depending on the conditions of the host, e.g. load, availability of resources in the TEE, and the requirements of the developers. Embodiments of this invention provide Cloud and infrastructure managers with the ability to choose in which TEE platform they are going to launch the applications or micro-services and at the same time they make them aware of the TEE platform and its properties, i.e. embodiments of this invention make it possible to jointly manage and orchestrate RAs and TAs. Additionally, they propose a solution to leverage and extend features offered by TEEs such as attestation in order to provide proofs about the deployments and the decisions made to the application developers.
[0138] In the following are summarized the key technical items that embodiments of the present invention comprise:
[0139] ■ Design, develop and package services / applications as a set of microservices, allowing a mixed deployment in the trusted environments and untrusted environments, see Fig. 2, and assign each of them a label or categorize them: o Consider micro-services that must be deployed in a trusted environment pS? o Consider micro-services that should be deployed in an untrusted environment pSu o Consider micro-services that can be flexibly deployed in a trusted environment or in an untrusted environment pSp. A preference indicator p can be assigned to each pSp that indicates preference for deployment in either environment, for example based on the service’s functionality in treating sensitive data, such as keys.
[0140] ■ Design of a micro-service and its classification as pS? , pSu or pSp can follow well-known software design principles and can include additional reasoning, o code sensitivity o sensitivity of treated data (keys) and associated storage / CDS o type and variety of different platforms for later deployment and operation o costs of deployment in a TEE
[0141] ■ Opt for automated design of micro-services and their classification per the previous item o apply design criteria per the previous item to Design Centers for code design and generation o apply AI / ML to support the automated computation of the design and code generation
[0142] Fig. 3 illustrates a method according to an embodiment, used to generate the trusted micro-services based on KPI and resource requirements, cost and type of TEE. The Smart TEE SW engine will use the inputs to generate an array of M trusted and N Rich Space applications. These images could be signed as a whole or separately according to the expected security requirements, remote attestation and role of the developer vs operator. The proposed methods for design, development, testing, packaging, deployment, and remote attestation of both trusted and untrusted applications are illustrated in Fig. 6. A proposed approach involves utilizing key performance indicators, KPIs, available resources, resource requests, costs, Trusted Execution Environment, TEE, types, and other information to create images automatically through the Smart TEE software engine. After the necessary image has been constructed, packaged, and signed, a deployment request is submitted to the orchestrator to initiate deployment. Following deployment, the application is subjected to remote attestation procedures outlined in Fig. 5, at which point it becomes fully operational.
[0143] ■ Prepare individual or group of micro-services for an attestation procedure by signing individual or groups of micro-services
[0144] ■ Package individual or groups of micro-services for automated deployment, e.g. image files, and provide one or multiple suitable package descriptors
[0145] ■ Package descriptors comprise information about: o inter-working between u-services based on their final deployment in trusted / untrusted environment o policies applying to the attestation procedure o platform and resource expectations ■ Utilize a common controller or orchestration function, that handles deployment and life-cycle management of micro-services in an untrusted as well as a trusted environment
[0146] ■ The common controller or orchestration function deploys an application’s micro-services, IST, pSu and pSp, per their classification and optimizes pSp deployment per the assigned preference as well as additional reasoning, incl.: o Platform capabilities in supporting the pSp, e.g. access to peripherals o Platform resources o Environmental risks, e.g. hardware resources in a protected or open environment o KPIs associated with communication between u-services, incl. between trusted and untrusted environment
[0147] ■ Micro-services that apply to the trusted environment can be deployed in a single enclave or in multiple enclaves o If they need frequent communication and there is no shared memory between enclaves, and the platform capabilities allow it, e.g. there is enough free memory, this reduces the communication overhead. o Deployment of single micro-service per enclave with communication between two or more micro-service enclaves allows the common controller and orchestration function to scale on a per micro-service basis
[0148] ■ Perform attestation of micro-services that have been deployed in the trusted environment o as a sub-package - group of micro-services that apply to the trusted environment - o individually per micro-service that applies to the trusted environment
[0149] ■ Opt for the attestation of micro services that have been deployed in the untrusted environment o as a sub-package - group of micro-services that apply to the untrusted environment - o individually per micro-service that applies to the untrusted environment ■ In a closed-loop operation, the common controller or orchestration function may re.-configure the deployment by changing the environment for the deployment of a single of multiple micro-services - from untrusted environment to trusted environment, or vice versa - o Practically that may be accomplished by applying a new deployment scheme and associated strategy o New deployments represent the re-configuration and may follow a make-before-break - new deployment applied and operational before the previous deployment is removed - or break-before-make - previous deployment is removed before the new deployment is applied and operational -
[0150] ■ Get a proof / confirmation that the micro-services have been stopped and that its assigned memory has been freed securely, e.g. by zeroing the values on memory
[0151] ■ Computing and enforcing the best match in a mixed deployment, where some of the application’s micro-services apply to the platform’s trusted environment, whereas some other micro-services apply to the untrusted environment
[0152] ■ Enabling the attestation of at least those micro-services that apply to the trusted environment regardless of the underlying hardware
[0153] ■ Including the trusted compute platform’s static - type, resources, performance - and dynamic - load, availability - characteristics, as well as a current risk assessment - threat analysis, attack potential - into the computation of the best match deployment
[0154] ■ Periodic or event-based re-computation, enforcement and attestation of an updated deployment
[0155] While orchestrators already leverage features exposed by regular processors, i.e. processors that do not have hardware support for TEEs, to control the applications and micro-services running on them, Virtual Infrastructure Managers need to be extended with an adaptation layer, the proposed Secured Virtual Manager, SIM, to be able to deal with TEEs. This extension needs to consider the particular characteristics of each platform. For example, Keystone supports three ways of writing enclave applications, eapps, as:
[0156] 1. standalone Keystone-native enclave applications,
[0157] 2. un-modified RISCV binaries with RT support, or
[0158] 3. partitioned applications running selected parts in the enclave
[0159] A practical way to implement the proposed method is the 3rdapproach. It could be implemented as follows:
[0160] ■ Introducing Adaptation layer for managing both TAs by means of Secured Virtual / lnfrastructure Manager, SVIM / SIM, and RAs by means on VI Ms.
[0161] ■ Introducing service / network / application descriptor for both TAs and RAs comprising both requirements
[0162] ■ Registering Nodes with TEE features
[0163] ■ Leverage node descriptor information for TEE application placement
[0164] ■ Remote attestation for both TA and RA
[0165] The adaptation layer may comprise a TEE image factory based on the underlying TEE infrastructure, Security requirement, available resource, or any other useful input.
[0166] Example use case
[0167] When writing an application as a set of micro-services, the developer specifies which of the micro-services need the protection of the TEE, which ones do not, and which ones he / she allows to run on both TEEs or REEs, i.e. flexible micro-services, depending on, for example, the benefits that the cloud provider is offering, e.g. cost savings, or on the state of the hosts, e.g. load, KPIs. In general, this can be specified by means of a descriptor file. Considering only the requirements referring to security affecting the management of TEEs, this file might look as follows:
[0168] Application:
[0169] • Micro-service 1 : Trusted
[0170] • Micro-service 2: Flexible (Conditions) Micro-service 3: Untrusted
[0171] • Micro-service N: Flexible (Conditions)
[0172] That is, the developer can specify which micro-services can be deployed in both the trusted and untrusted environments and under which conditions. For example, if a Machine Learning, ML, service of inference is split into different micro services, e.g. each layer of a Neural Network is considered as a micro-service. If the developer considers that only the first and the last layers need to be secured against any kind of inference because they process clear texts whereas the intermediate layers process encrypted data and an attacker could not infer anything from them, he would mark the initial and last layers as Trusted, i.e. they would be deployed in a TEE, and the rest of the layers as Flexible, i.e. they could run in both the TEE or the REE depending on certain constraints.
[0173] Flexible micro-services also allow controllers to dynamically move them between the trusted and the untrusted environments. As an example, Fig. 4 considers an application composed of four micro services pSui , pS-n , PSFI and SF2. For simplicity all of them are depicted in the same host, although they could be running on different hosts. Similarly, the rest of the applications that might be running on the host are not depicted. For example, if the host has enough space in the TEE it might run both micro-services inside it (a), it might need to balance the load between both environments and move one of the micro-services to the untrusted environment (b) or it might move both to the untrusted environment (c). Note that these three options are also possible when the secure manager makes the allocation decision.
[0174] The procedure described in Fig. 5 for attestation has the centralized Control and Orchestration layer or manager as key actor. It is considered as a trusted entity. As such, it is in charge of deciding whether a deployment was successful or not. To this end, it attests each of the micro-services that has to run in the secure area of the processor, i.e. within the TEE, and generates a proof after completion. Note that the orchestration manager might not have direct access to the hardware, e.g. it is not integrated as part of the SM in RISC-V based processors, and as a result, it cannot get a proof signed by the embedded hardware key to attest the micro-services deployed in the untrusted environment. However, since it is considered trusted, it can generate proofs signed by itself certifying that certain micro-service was launched on the untrusted environment.
[0175] In a nutshell the process is as follows:
[0176] 1- The Control & Orchestration, C&O, layer receives a request to deploy a set of micro services and it also gets a file with the descriptors of each of them including the security requirements, i.e. which micro services need to run inside a TEE, which on the REE and which ones are flexible
[0177] 2- Considering the KPIs, available resources, especially machines with TEE capabilities and their architectures, the request file, costs, etc., the C&O layer decides the allocation of the resources. a. It obtains or generates an identifier, e.g. hash, for each of the microservices or images that should run inside a TEE, and for the ones that it has decided to allocate inside a TEE b. It composes a file with the summary of those services with their assigned identifier - some kind of attestation template - and signs it Proof of the decision
[0178] 3- Services are deployed according to the decision made at stage 2, and whenever each host launches the application, it starts the RA procedure with the central manager
[0179] 4- Once the general manager has received a fresh attestation statement from each host, it compares the initial attestation template - stage 2. a - with a new generated file that is composed out of the proofs that each machine sends. If the two files are equal, attestation is considered successful, and the manager can then send the signed file as a proof of attestation to the deployment requester
[0180] 5- Optionally, the manager can include information about the micro-services that it decided to launch on the untrusted side on the proof of attestation, and sign it as responsible for the deployment and the decision.
[0181] Note that the signature at stage 2.b is not needed for the system to work, it just provides extra security guarantees as host can enforce some policy and only launch applications that come directly from the manager, i.e. signed by it. Fig. 6 depicts our proposed method for designing, developing and packaging microservices comprising parts running in the trusted and untrusted environments and flexible micro-services that can be able to run in both trusted and untrusted environments. The classification of the types of micro-services is done according to Fig. 6 in an automated manner. Fig. 6 shows the full steps considered in the proposed method, an example pseudo code implementation for its operation is detailed below:
[0182] # Step 1 : Design, develop and package micro-service / micro-services
[0183] # Define categories of micro-services micro-services = [
[0184] {"name": "servicer, "category": "pST"}, # Microservice to be deployed in trusted environment
[0185] {"name": "service2", "category": "pSU"}, # Microservice to be deployed in untrusted environment
[0186] {"name": "services", "category": "pSF", "preference": "trusted"} # Microservice can be deployed in either environment, with preference for trusted environment with levels for example from 0 to 8
[0187] ]
[0188] # Step 2: Design and classify micro-services
[0189] # design and classification of micro-services def automate_design(micro-services): for service in micro-services:
[0190] # Apply design criteria code_sensitivity = determine_code_sensitivity(service) data_sensitivity = determine_data_sensitivity(service) deployment_platforms = determine_deployment_platforms(service) deployment_costs = determine_deployment_costs(service)
[0191] # Classify micro-services based on criteria if code_sensitivity == "high" or data_sensitivity == "keys": service["category"] = "pST" elif deployment_costs == "low": service["category"] = "pSU" return micro-services
[0192] # Step 3: Attestation and packaging of micro-services def prepare_micro-services(micro-services):
[0193] # Sign individual or groups of micro-services for attestation sign_micro-services(micro-services)
[0194] # Package micro-services for automated deployment package_micro-services(micro-services)
[0195] # Package descriptors for inter-working between u-services and attestation policies package_descriptors(micro-services) return packaged_micro-services
[0196] # Step 4: Deployment and life-cycle management def orchestrate_micro-services(packaged_micro-services):
[0197] # Deploy micro-services in trusted and untrusted environments per their classification for service in packaged_micro-services: if service["category"] == "pST": deploy_microservice_in_enclave(service) elif service["category"] == "pSU": deploy_microservice_in_untrusted_environment(service) else: deploy_microservice_flexibly(service)
[0198] # Optimize pSF deployment per preference and additional reasoning optimize_flexible_services(packaged_micro-services)
[0199] # Scale deployment on a per microservice basis scale_micro-services(packaged_micro-services)
[0200] # Perform attestation of micro-services in trusted and untrusted environments perform_attestation(packaged_micro-services)
[0201] # Reconfigure deployment by changing environment for single or multiple microservices reconfigure_deployment(packaged_micro-services)
[0202] # Ensure memory has been freed securely secure_memory_deallocation(packaged_micro-services)
[0203] # Compute and enforce the best match in a mixed deployment compute_best_match_deployment(packaged_micro-services)
[0204] # Enable attestation of micro-services in trusted environment regardless of underlying hardware enable_trusted_attestation(packaged_micro-services)
[0205] # Include static and dynamic characteristics, as well as current risk assessment, in computation of best match deployment incorporate_characteristics_into_computation(packaged_micro-services)
[0206] # Periodic or event-based re-computation, enforcement, and attestation of updated deployment recompute_updated_deployment(packaged_micro-services)
[0207] Fig. 7 shows an example deployment based on the RISC-V architecture and other architectures may follow the same approach. The closed loop operation for the coordinated management of both trusted and untrusted / rich applications is illustrated in Fig. 8 for the case of ETSI NFV MANO and in Fig. 9 for a general cloud service. The monitoring module gathers telemetry and other valuable data and sends it to an AI / ML data processing engine, which generates policies for lifecycle management. The orchestration module then enforces the recommended policies.
[0208] The proposed method is depicted in Fig. 10, which demonstrates a common MANO layer with northbound API interface directed towards the orchestration and control layer, and southbound interfaces directed towards the trusted computing enabled platform. The common MANO module functions as an abstraction layer for both the virtual infrastructure manager, VIM, and the secure infrastructure manager, SIM, and it is responsible for converting updated API parameters via the SBI and NBI to the TEE and the Orchestration and control layer.
[0209] Many modifications and other embodiments of the invention set forth herein will come to mind to the one skilled in the art to which the invention pertains having the benefit of the teachings presented in the foregoing description and the associated drawings. Therefore, it is to be understood that the invention is not to be limited to the specific embodiments disclosed and that modifications and other embodiments are intended to be included within the scope of the appended claims. Although specific terms are employed herein, they are used in a generic and descriptive sense only and not for purposes of limitation.
Claims
C l a i m s1. A computer-implemented method for operating at least one application in a trusted computing enabled platform, wherein the platform comprises a trusted environment and an un-trusted environment, wherein the at least one application comprises multiple microservices, and wherein at least one of the multiple micro-services is designed as a flexible microservice, which can be deployed either in the trusted environment or in the un-trusted environment, comprising the steps:- deciding on a deployment of the flexible micro-service in the trusted environment or in the un-trusted environment depending on at least one parameter of a requested security level of the at least one application and / or of a requested performance level of the at least one application and / or of a present or available platform characteristic or feature; and- deploying the flexible micro-service in the trusted environment or in the un- trusted environment according to a result of the deciding step.
2. The method according to claim 1 , wherein operating the at least one application comprises control and / or orchestration and / or deployment and / or management of the at least one application.
3. The method according to claim 2, wherein management of the at least one application comprises life-cycle management of the at least one application.
4. The method according to any one of claims 1 to 3, wherein deployment of the at least one application in the platform comprises deployment of its multiple microservices, each in the trusted environment or in the un-trusted environment.
5. The method according to any one of claims 1 to 4, wherein the deciding step considers an automated and / or optimized deployment and / or management of the at least one application,and / or wherein the deciding step and the deploying step provide a best match deployment of the flexible micro-service under consideration of at least two of the at least one parameter.
6. The method according to any one of claims 1 to 5, wherein the platform characteristic or feature comprises a static and / or a dynamic characteristic, wherein a static characteristic can include a type, a resource and / or a performance of the platform and wherein a dynamic characteristic can include a load and / or an availability of the platform, and / or wherein the platform characteristic or feature comprises a risk assessment, wherein the risk assessment can include a threat analysis and / or an attack potential.
7. The method according to any one of claims 1 to 6, wherein the deciding step and the deploying step are performed by means of a Virtual Infrastructure Manager, VIM, that includes a Secure Infrastructure Manager, SIM, and / or wherein the deciding step and the deploying step take into account a list of multiple descriptors that apply to a single application.
8. The method according to any one of claims 1 to 7, wherein the deciding step and the deploying step are repeated for providing an updated deployment based on a change of at least one parameter of a requested security level of the at least one application and / or of a requested performance level of the at least one application and / or of a present or available platform characteristic or feature.
9. The method according to any one of claims 1 to 8, wherein at least one microservice is designed under consideration of a code sensitivity, under consideration of a sensitivity of treated data or keys and an associated storage, under consideration of a type and variety of different platforms for later deployment and / or operation, and / or under consideration of costs of a deployment in a platform.
10. The method according to any one of claims 1 to 9, wherein at least one microservice is classified as a micro-service which has to be deployed in a trusted environment, which should be deployed in an un-trusted environment or which can be flexibly deployed in a trusted environment or in an un-trusted environment,and / or wherein at least one of the multiple micro-services is designed for deployment only in the trusted environment or only in the un-trusted environment.
11. The method according to any one of claims 1 to 10, wherein the platform comprises a common control and orchestration function comprising a Virtual Infrastructure Manager, VIM, and a Secure Infrastructure Manager, SIM.
12. The method according to claim 11 , wherein information and / or semantic is added to an interface or interfaces of the control and orchestration function for operation of the control and orchestration function, wherein the interface or interfaces can be a northbound interface toward service orchestration and / or a southbound interface toward a trusted computing platform.
13. The method according to any one of claims 1 to 12, wherein the method comprises one or more attestation mechanisms, wherein an attestation mechanism can comprise: attestation of micro-services that have been deployed in the trusted environment as a sub-package - group of micro-services that apply to the trusted environment - or individually per micro-service that applies to the trusted environment, or an option function for the attestation of micro services that have been deployed in the untrusted environment as a sub-package - group of micro-services that apply to the untrusted environment - or individually per micro-service that applies to the untrusted environment, or enabling the attestation of at least those micro-services that apply to the trusted environment regardless of the underlying hardware.
14. An application for operating in a trusted computing enabled platform, preferably for use in the computer-implemented method according to any one of claims 1 to 13, wherein the platform comprises a trusted environment and an untrusted environment, wherein the at least one application comprises multiple microservices, and wherein at least one of the multiple micro-services is designed as a flexible microservice, which can be deployed either in the trusted environment or in the un-trusted environment.
15. A system for performing a computer-implemented method for operating at least one application in a trusted computing enabled platform, preferably for carrying out the computer-implemented method according to any one of claims 1 to 13 and / or preferably with an application according to claim 14, comprising:- a platform with a trusted environment and an un-trusted environment;- at least one application comprising multiple micro-services; wherein at least one of the multiple micro-services is designed as a flexible microservice, which can be deployed either in the trusted environment or in the un-trusted environment,- deciding means for deciding on a deployment of the flexible micro-service in the trusted environment or in the un-trusted environment depending on at least one parameter of a requested security level of the at least one application and / or of a requested performance level of the at least one application and / or of a present or available platform characteristic or feature; and- deploying means for deploying the flexible micro-service in the trusted environment or in the un-trusted environment according to a result of the deciding step.
Citation Information
Patent Citations
Method for loading an application consisting of a plurality of components into a device consisting of a plurality of components
US20150331698A1
Optimizing redeployment of functions and services across multiple container platforms and installations
US20190102157A1
Optimizing deployment and security of microservices
US20220121470A1
Cited By
Trusted cloud native security level treatment system and method
CN121585481A
