Jul 2026· International Symposium on Service Oriented Software Engineering· pp. 90-97· 0 citations· 31 references
Abstract
Edge AI is evolving from isolated inference toward long-running services that coordinate model pipelines, data streams, state, and accelerators near users and physical environments. Cloud-native and edge-native platforms offer useful foundations, but their primary control objects–containers, nodes, links, and enrolled sites–generally do not expose the model, data, state, quality, and participation semantics required by these services. This paper presents eAI+, a vision for an edgeAI-native service platform built around three first-class control objects: AI service graphs, dynamic edge resource fabrics, and participant contracts. eAI+ aims to preserve service quality under latency, privacy, reliability, cost, and participation constraints through three coordinated mechanisms. Runtime would select safe execution adaptations based on current workload, environment, and contract signals. Deployment would map service-graph components and prepared fallbacks to heterogeneous resources. PolyLink is the participant-contract module for plug-and-play resource onboarding; it would register contributors, verified resource offers, capabilities, and participation terms. Once a resource is onboarded, it would become available to Deployment for placing eligible service-graph components under the registered contract, while PolyLink would maintain metering, reputation, rewards, and exit events. Migration would transfer only continuity-critical state or control when mobility, overload, policy changes, or contributor lifecycle events invalidate the current placement. This framing treats edge AI as a coordinated service-platform problem across models, data, state, resources, and contracts.
Physical AI policies require inference throughout their lifecycle, including model evaluation, cloud reinforcement learning rollout, edge GPU serving, and onboard deployment. Although these settings share the same checkpoint and action semantics, they often rely on separate inference programs. To unify them, we build PhyAI, a Physical AI inference engine with a single runtime that keeps architecture-specific conditioning, solver, cache, and output logic in model adapters while sharing graph execution, kernels, memory management, and parallel services. The same codebase runs vision-language-action (VLA) models and world-action models (WAMs) on single or multiple GPUs across onboard, edge, and cloud deployments. We used the adapter interface to add MiniCPM-Robot on the day of its release. PhyAI achieves 1.40x-4.65x speedups over the official implementations of pi0, pi0.5, GR00T N1.7, and MiniCPM-Robot. On Cosmos3-Nano-Policy-DROID it reduces latency from 2.46 to 1.18 s on eight H20 GPUs (CFG=2, TP=4), a 2.08x speedup. Specialized runtimes remain faster in several configurations, so our goal is one runtime with competitive latency rather than the fastest result in every case. Detailed profiles reveal why different models need different execution policies: on a Hopper-series GPU at batch size one, the pi0.5 action expert accounts for 8.8% of FLOPs but 57.2% of latency; at batch size 32 its share drops to 13.5% and throughput reaches about 100 samples/s. Cosmos3 remains generation-dominated and gains only 14.3% throughput as batch size increases from 1 to 16. We further introduce the control-time Roofline, which distinguishes inference-bound from environment-bound control; the measured pi0.5 points on four LIBERO suites are environment-bound while Cosmos3 stays inference-bound. Code and benchmarks: https://github.com/mingti-org/phyai.
Chenghua Wang, Daliang Xu, Dongqi Cai et al.· 0 citations
Frontier open-weight models are increasingly available, but serving them still largely assumes datacenter infrastructure. We present FreeToken, an edge-native MoE serving system that treats a personal machine not as a small GPU, but as a unified, elastic inference platform. FreeToken co-designs the full serving stack, including model layout and loading, expert residency, CPU--GPU execution, agentic state reuse, and runtime memory management, around two realities of local AI: agent workloads continuously change their execution pattern, and edge hardware exposes heterogeneous resources whose balance differs from machine to machine. Rather than committing to a fixed offloading strategy, FreeToken continuously maps computation and model state onto the resources actually available. FreeToken supports more than 20 MoE models and real coding and tool-using agents across hardware ranging from an 8GB laptop GPU to a single workstation GPU. More importantly, it changes what these machines can practically serve, from a 35B model on a laptop to a 284B model on a gaming desktop and the 753B GLM-5.2 on a single workstation GPU. FreeToken turns open weights into deployable local software, making the machines users already own a practical platform for frontier-scale intelligence. We release the system at flashml.ai.
Shuo Yang, Xiao-yun Fan, Melissa Z. Pan et al.· 0 citations
Cloud Continuum applications require experimental environments capable of combining heterogeneous Edge, Fog, Cloud, and high-performance computing resources while preserving reproducibility, observability, and control over distributed deployments. This paper presents a two-level reference architecture for Cloud Continuum experimentation built on top of the SLICES Cloud Continuum Blueprint. The proposed approach separates the research-infrastructure layer, which exposes and manages distributed resources, from the application layer, where Cyber-Physical workflows are organized according to an Edge-Fog-Cloud pattern in which placement, timing, and data provenance are treated as first-class experimental concerns. The architecture is designed to support multiple continuum applications rather than a single domain-specific prototype. At the Edge, applications interact with physical devices and perform low-latency sensing or safety actions; at the Fog, they execute near-source coordination, mediation, and stream-processing logic; at the Cloud, they consolidate global knowledge through analytics, optimization, and visualization. This partitioning enables researchers to deploy, customize, and compare alternative control and monitoring strategies over the same programmable infrastructure substrate. The approach is validated through two representative use cases: Renewable Energy Community management, where distributed Digital Twin coordination and time-window-based energy control are requested, and AirWatch, a monitoring pipeline focused on anomaly detection, low-latency alerting, and cloud-side aggregation. Both workloads are evaluated through a systematic campaign of 40 runs comparing virtualized and physical edge deployments over a geographically distributed infrastructure.
Fabio Orazio Mirto, Giuseppe Tricomi, L. D’Agati et al.· 0 citations
Modern distributed applications operate across heterogeneous cloud, edge, and cyber-physical environments where maintaining object identity, synchronizing state, and enabling real-time interaction remain challenging. Existing communication frameworks are mainly message-centric and treat application state as serialized payloads, which increases overhead when objects are continuously observed, incrementally updated, or shared across multiple nodes. This paper presents Esiur, a resource-centric distributed runtime that provides reference-based interaction through network-transparent resources exposing properties, events, and functions. Esiur preserves object identity across nodes and performs incremental state synchronization without application-level coordination. It defines a unified binary protocol based on the Type Representation Unit (TRU) and Transfer Data Unit (TDU), enabling runtime type definition (TypeDef) discovery and efficient transmission of strongly typed and heterogeneous data without external schemas. The framework is evaluated through serialization benchmarks, protocol-overhead measurements, ordered-notification analysis, annotation-guided invocation, subscriber and attachment scalability tests, and cross-language reattachment recovery between C# and TypeScript implementations. Results show a 36.4%–68.7% reduction in serialization size relative to JavaScript Object Notation (JSON), statistically bounded notification latency, successful runtime TypeDef-based invocation, final-state convergence after transient wide-area network (WAN) disconnection, and stable behavior under concurrent subscriber and attachment loads.
Ahmed Kh. Zamil, Ammar D. Jasim· IEEE Access· 0 citations
Industrial and autonomous systems increasingly depend on AI, automation, and real-time coordination to act on operational data as it is generated. Yet conventional architectures often require that data to pass through centralized platforms before decisions can be made. Cloud systems remain valuable for training, reporting, and long-term analytics, but they add latency and external dependencies to the critical decision path and become harder to scale as each new site adds additional edge devices and data. As intelligence spreads across machines, sites, facilities, and vehicles, continued dependence on centralization will constrain response time, resilience, scalability, and autonomous operation. This paper presents the AnyLog Edge Data Fabric, an agent- and edge-based platform that manages operational data at its source while presenting distributed data, assets, compute resources, and services as one logical system. Through its Distributed Metadata Layer, Virtual Data Lake, Unified Namespace, Single System Image, and Model Context Protocol, authorized users, applications, automation services, and AI agents can discover, query, process, and act on distributed resources without knowing where they are hosted. Queries and computation execute at the agents holding the relevant data, so only requests and results traverse the network. This preserves local ownership, reduces data movement, supports continued operation during connectivity disruptions, and enables repeatable deployment from validated digital-twin configurations. AnyLog provides a cloud-like operating model for distributed SQL, real-time automation, Edge AI, federated learning, and resilient decision-making without a single point of failure or any dependence on centralized infrastructure.
Roy Shadmon, Mark Davidson, Eric Aquaronne et al.· 0 citations
Semantic mapping is a core service that enables grounded interactions in emerging Extended Reality (XR) applications such as AI assistants. Deploying this capability on mobile XR devices requires a system that is open-vocabulary, real-time, and low-power. Existing approaches are compute-intensive and assume server-class resources. Cloud offloading offers a practical path, but no existing system splits semantic mapping between the device and the cloud, and current approaches do not address how to manage communication, execution, and memory footprint across the device-cloud boundary. We present SemanticXR, the first device-cloud system for real-time, open-vocabulary semantic mapping and querying under XR power, bandwidth, and memory constraints. Our key insight is to elevate semantically identifiable objects to first-class units of system design, governing how the system communicates, executes, and manages memory across the device and the server. Evaluation against a new, aggressive device-cloud baseline shows that object-level system organization improves server-side mapping latency by 2.2x at equivalent semantic quality. Object-level depth-mapping co-design maintains upstream bandwidth under 2.5 Mbps. On the device, an object-level sparse local map with incremental updates and update prioritization enables sub-100 ms query latency for up to 10,000 objects even under network drops, supports tens of thousands of objects within 500 MB memory footprint, and scales downstream bandwidth with map changes rather than total scene size. The system adds only about 2% to idle device power.
Rahul Singh, Devdeep Ray, Connor Smith et al.· 0 citations