Programming support for sharing resources across heterogeneous mobile devices

CLES: A Universal Wrench for Embedded Systems Communication and Coordination

“…The CLES service wrapper abstracts this interface, which opens opportunity for future development to extend this interface to include memory mapped I/O and common data distribution platforms such as Google ProtoBuf, ActiveMQ, and DDS. Embedding the network interface into the CLES SDK interface caters to the desires of the embedded systems community because unlike solutions such as the the RQL mobile device runtime [12], and data brokers like ActiveMQ, P2P CLES avoids passing of information between third-party runtimes or brokers. This helps reduce latency, but also supports deterministic real-time scheduling as defined by the parent application because this interface is called in-line directly by the parent with no additional non-deterministic processing constraints or data transmissions.…”

Towards Energy Efficiency in the Internet of Wearable Things: A Systematic Review

“…Zheng et al [107] outlined the software engineering support for appflication developers to utilize shared resources between mobile devices for optimal performance through seamless resource sharing to enhance programmer's productivity as well as reduce energy consumption and execution time of devices. Additionally, some studies propose using low power hardware for the development of future wearables by providing the overall energy consumption profile, and achieved energy savings [108], [112].…”

“…A, N, SW [46], [111] Energy awareness regarding neighboring nodes to select the optimal route [50] Multi parameter cost function for the next hop selection [60] Selective data routing based on the data priority Securityrelated aspects HW, DP, SW [110] Content agnostic privacy and encryption protocol eliminating the need for asymmetric encryption [180], [181] Integration of lightweight cryptography solutions including more appropriate elliptic curve types or algorithm implementations [186] More efficient utilization of manufacturer-provide SoCs accelerated for cryptographic primitives execution [187] Finding trade-offs between the primitive and required level of the provided security Processing limitations HW, DP, SW [54] The use of heterogeneous multicore processor gateway as compared to little cores gateway working as a router [64], [86], [104], [105] Task offloading to leverage high computing resources of nearby devices for improved performance [106] Edge/fog/cloud computing techniques for optimal performance [107] Seamless resource sharing between heterogeneous mobile devices Storage limitations HW [47], [55] Data compression to reduce the size of the dataset for efficient data processing and storage [106] Edge/Fog/Cloud computing techniques for better performance [173] Data summarization and aggregation Lack of hardware acceleration HW, SW [47], [55] Data compression to reduce the size of the data set for more efficient data processing and storage [64], [86], [104], [105] Task offloading to leverage high computing resources of the nearby devices for the improved performance [186] Identifying and use of present hardware acceleration, which may not be accessible by the default Inefficient use of energy consuming modules HW, SW [62] Configurable data acquisition modules [88] Replacing high power consumption modules with low power alternates, e.g., using two accelerometers instead of a gyroscope as...…”