The field of real-time data analytics and streaming is experiencing significant growth, driven by the increasing availability of satellite networks and the need for low-latency data processing. Recent developments have focused on optimizing data transfer and processing over wide area networks (WANs) and satellite constellations, enabling more efficient and reliable data analytics and streaming applications. Notably, researchers have proposed innovative frameworks and algorithms for dynamic WAN bandwidth prediction, real-time analytics over Earth observation data, and adaptive bitrate selection for live streaming over Low Earth Orbit (LEO) satellite networks. These advancements have the potential to revolutionize various applications, including disaster response, environmental monitoring, and immersive media systems.

Noteworthy papers include: WANify, which introduces a machine learning-based framework for dynamic WAN bandwidth prediction, enabling more efficient data transfer and processing. OrbitChain, which presents a collaborative analytics framework for real-time Earth observation data processing over satellite constellations. INDS, which proposes an adaptive streaming framework for real-time point cloud video streaming over Information-Centric Networking (ICN) architectures. Databelt, which introduces a state management framework for serverless workflows in dynamic environments, such as the 3D Compute Continuum. CausalMesh, which presents a formally verified causal cache for stateful serverless computing, enabling coordination-free and abort-free read/write operations.