Objective 2: Provide platform experts with efficient means to capture their knowledge
Summary: This objective is to enable efficient documentation and formalization of the knowledge of platform experts (e.g., cloud expert mastering advanced frameworks, or embedded system developers able to optimize memory or power consumption on resource-constrained platforms) into reusable forms. The capabilities and peculiarities of each platform should be well-documented and should be leveraged at their maximum instead of being "harmonized" according to their least common features denominator.
Challenges: This objective is a daunting challenge due to the tremendous diversity of the future computing continuum. Despite an apparent homogeneity (x86 architecture, IP protocol and middleware) the cloud providers' offerings are typically incompatible with each other (vendor lockin). Moreover, the extreme diversity of smart device architectures (ARM, AVR, PIC, etc.) and communication protocols (WiFi, and a large offer of non-IP protocols such as X-Bee, BlueTooth, RF, etc.) cannot be handled by use of software frameworks or middleware due to the stringent resource constraints of most devices (typically 8-bit, 8-32MHz, 1-16Kb RAM). Moreover, most of the legacy devices currently on the market are based on proprietary protocols and platforms making any classical top-down engineering process impractical.
Approach: The HEADS approach will employ generative techniques in order to map models to different platforms. When targeting cloud platform, HEADS will employ model-to-model transformation in order to map to existing MDE-based approach for cloud, such as the CloudML language developed in the FP7 MODAClouds and PaaSage projects. When targeting resource-constrained devices, HEADS will employ model-to-text generation to generate specific code (typically C/C++) optimized for a minimal runtime overhead and a full access to the device capabilities
- An open-source code generation framework with a plugin mechanism to integrate other platforms, in particular legacy platforms managed by the HEADS case study and technology providers. This open-source framework will not impose any particular license on the generated drivers, which could be open-source or proprietary.
Final HEADS project results on the future computing continuum!Jun 28
After 42 months of research on heterogeneous and distributed services for future computing continuum, HEADS project concluded in
Tellu participating at Vitalis 2017May 02
Tellu participated at Vitalis 2017 in Gothenburg, Sweden, on 25th-27th of April 2017.
TelluCloud innovation developed in the HEADS project at EC's stand in CEBIT 2017Mar 20
We are very glad to announce that Tellu AS has officially been selected as one of the top inn