Paving the path from increased trust to more powerful cloud applications
For cloud computing to achieve its full market potential, the next logical step in its evolution would be a switch from storage to remote processing of data. This requires a high level of trust among stakeholders, which is what the TRESCCA project aims to achieve. Although the cloud is undoubtedly making our life easier, a world where you can use it with no fear of security issues still looks unlikely. Recent security breaches in some of the most popular storage services (remember the ‘Celebgate’?) have reinforced the idea among end users that sensitive data should stay far away from the cloud. And while cryptography is an efficient way to get around the problem for cloud storage, its use is incompatible with remote processing of the stored data. This poses a major problem: by 2016, over a quarter of all applications are predicted to be available on the cloud. Not only is the resort to cryptography slowing down this evolution, but service providers can also find it hard to trust and allow computation of data stored on their servers by their clients — thereby hindering innovation in the sector. To solve this issue, TRESCCA (TRustworthy Embedded systems for Secure Cloud Computing Applications) aims to propose and demonstrate innovative hardware security and virtualization techniques for the cloud. These would allow stakeholders to delegate the processing of sensitive data to a remote processing engine while avoiding a shift in paradigm, as they will complement existing legacy solutions with non-intrusive add-ons instead of replacing them. Andreas Herrolz, coordinator of the TRESSCA project, explains how the consortium’s work will help build the long-awaited trust between cloud users and service providers, thereby opening the door to a whole new field of services and applications. EU citizens are increasingly relying on cloud computing applications. Can they feel safe doing so? In general cloud services have become more reliable and secure over the last years. Nowadays they provide a high level of comfort and convenience and are particularly useful in the context of mobile and cross-device applications. Cloud service providers have learned from previous mistakes and generally improved the level of data protection and security of their services. However, there is still a significant risk involved when storing any kind of private and sensitive data in the cloud. Events like the Snowden revelations and recent hacking attacks on popular cloud services and companies have shown that even if service providers are assuring their users of a high level of security, end users cannot fully trust them. Would you say that the cloud market is currently hindered by this lack of trust between providers and users? Yes, definitely. The cloud has a very high potential to significantly change the way we use computers and mobile devices. It is also the backbone of future concepts like Smart Cities and the Internet of Things. However, as long as we cannot guarantee and verify a high level of security and protection for the data that is stored and processed on remote servers, these kinds of applications cannot be realized without making compromises. What kind of techniques does TRESCCA use to create a more secure and trusted cloud? TRESCCA is developing hardware and software components that help to create a safe and secure execution environment for any kind of computing applications. On the hardware level, TRESCCA is developing security components that are integrated into so-called system-on-chips. Such chips are already commonly used in computers, smartphones, tablets and set-top boxes. The TRESCCA security components, called Hardware Security Modules (HSM) protect the communication inside and outside of the chip, for example any attempt to modify or read data stored inside the RAM of a computer would be detected and prevented. On top of that, TRESCCA creates a secure software execution environment using virtualization technology to isolate applications in small light-weight virtual machines. These virtual machines can also be used to securely move trusted applications between devices. The combination of HW and SW technology provides a level of security that can be verified remotely by service providers and users. For demonstration purposes, TRESCCA is integrating these technologies into cloud-connected client devices, such as smart meters and set-top boxes. By providing trusted and verifiable execution environments, certain parts of the application and its corresponding data do not have to be moved to the cloud but can instead be executed locally. In the future, the same technology might also be integrated into cloud servers, enabling end users to assess and verify the security of the servers remotely before moving their data into the cloud. You claim that your system will enable new cloud services and applications. Can you explain? Currently, any sensitive data needs to be encrypted on client-side before it is stored in the cloud in order to be secure. However, this also prevents any kind of meaningful computation on that data in the cloud. On the other hand, if end user devices, like PCs or smartphones, are not trustworthy, any kind of local computation done by these devices will not be trusted by service providers. So we end up in kind of a dead-lock situation for any cloud application that requires service providers to access and process private data which itself is too sensitive to be stored in the cloud. With TRESCCA such applications are possible, with trustworthy end-user devices processing sensitive data locally. Then only the results need to be transmitted to the service provider. In general, this may introduce a paradigm shift on how cloud services are designed today, reduce data-monopolies of providers and lead to more decentralized architectures. TRESSCA solutions will also be made available for free. Why did you choose to do this? So far, most existing hardware security solutions are proprietary and either available under commercial license or not at all. However, we think that this limits adoption, and security solutions will only be accepted and deployed on a wide scale if they can be used, assessed and integrated into products by everyone without any restrictions. Also we would like to cooperate with other companies and researchers for further improvement of our solutions and we think that a free access model supports this vision. Nevertheless, there are still several options for TRESCCA partners to exploit its results commercially, for example by making end-user products based on TRESCCA technology or by offering development services to other companies. When can we expect your technology to reach the market? For the hardware solutions it could still take several years, since development of new system-on-chips is very costly and takes a lot of time, in particular if new components are being integrated for the first time. However, the software solutions delivered by TRESCCA are not dependent of these components and may already be used with existing hardware. So they may reach the market much earlier, within one or two years. The project is close to completion. What’s the industry feedback so far? Actually, there is a very high interest in free and open hardware security solutions from industry. This includes semiconductor companies, device manufacturers as well as cloud solution providers. With cloud applications extended towards mobile and industrial applications, having reliable security solutions in place becomes more critical or even essential for commercial success. The solutions offered by TRESCCA and its open access model are very attractive for companies. However I think we can still improve how we explain what exactly is available from TRESCCA and how it can be used to create secure HW/SW applications. In the end, TRESCCA is only one part of the solution and, as with all security concepts, it has to be combined with others and used in the right way to be fully effective. For further information, please visit: TRESCCA http://www.trescca.eu/
Countries
Germany