Microprocessor Optimizations for the Internet of Things: A Survey.

The Internet of Things (IoT) refers to a pervasive presence of interconnectedand uniquely identifiable physical devices. These devices' goal is to gatherdata and drive actions in order to improve productivity, and ultimately reduceor eliminate reliance on human intervention for data acquisition,interpretation, and use. The proliferation of these connected low-power deviceswill result in a data explosion that will significantly increase datatransmission costs with respect to energy consumption and latency. Edgecomputing reduces these costs by performing computations at the edge nodes,prior to data transmission, to interpret and/or utilize the data. While muchresearch has focused on the IoT's connected nature and communicationchallenges, the challenges of IoT embedded computing with respect to devicemicroprocessors has received much less attention. This paper explores IoTapplications' execution characteristics from a microarchitectural perspectiveand the microarchitectural characteristics that will enable efficient andeffective edge computing. To tractably represent a wide variety ofnext-generation IoT applications, we present a broad IoT applicationclassification methodology based on application functions, to enable quickerworkload characterizations for IoT microprocessors. We then survey and discusspotential microarchitectural optimizations and computing paradigms that willenable the design of right-provisioned microprocessors that are efficient,configurable, extensible, and scalable. This paper provides a foundation forthe analysis and design of a diverse set of microprocessor architectures fornext-generation IoT devices.