The cloud-edge-end network architecture is an emerging model that integrates the advantages of both cloud and edge computing to effectively enhance the system's computational capacity. However, a key challenge within this architecture is how to efficiently offload tasks to fully utilize distributed computing resources while minimizing task execution latency. Most existing approaches assume that complete system information can be obtained in real time to make task offloading decisions. However, this assumption often leads to high communication overhead in practical applications, and the system's dynamic nature can make obtaining complete information impractical. To address this challenge, we propose a partially observable task offloading algorithm (POTO) that makes efficient offloading decisions based on partial system information. First, we model the task offloading problem as a partially observable Markov decision process (POMDP) and design an encoder using a Long Short-Term Memory (LSTM) network to extract system state information from past offloading decisions. Finally, we improve the actor-critic algorithm with a clipping mechanism and action masking to achieve an efficient offloading strategy. Experimental results show that POTO achieves a 95% task success offloading rate while reducing latency by at least 14% compared to other algorithms. Furthermore, even with partial system information, POTO performs at 96% of the effectiveness of algorithms utilizing complete system information.

In the trend towards hardware specialization, FPGAs play a dual role as accelerators for offloading, e.g., network virtualization, and as a vehicle for prototyping and exploring hardware designs. While FPGAs offer versatility and performance, integrating them in larger systems remains challenging. Thus, recent efforts have focused on raising the level of abstraction through better interfaces and high-level programming languages. Yet, there is still quite some room for improvement. In this paper, we present Coyote v2, an open-source FPGA shell built with a novel, three-layer hierarchical design supporting dynamic partial reconfiguration of services and user logic, with a unified logic interface, and high-level software abstractions which facilitate application deployment, multi-tenancy and transparent workload pipelining. Experimental results indicate Coyote v2 reduces synthesis times between 15% and 20% and run-time reconfiguration times by an order of magnitude, when compared to existing systems. We also demonstrate the advantages of Coyote v2 by deploying several realistic applications, including HyperLogLog cardinality estimation, AES encryption, and neural network inference. Finally, Coyote v2 places a great deal of emphasis on integration with real systems through reusable and reconfigurable services, including a fully RoCE v2-compliant networking stack, a shared virtual memory model with the host, and a DMA engine between FPGAs and GPUs. We demonstrate these features by, e.g., seamlessly deploying an FPGA-accelerated neural network from Python.