Real-time object detection on edge devices operates under tight computational, memory, and power budgets. Prior work typically treats model compression and hardware acceleration independently, yielding suboptimal trade-offs among accuracy, latency, and energy. We present AHCO-YOLO, an algorithm–hardware co-optimization framework (AHCO) that unifies model design, quantization, design space exploration (DSE), and hardware implementation. This approach overcomes the limitations of isolated methods and delivers synergistic gains. We introduce a hardware-friendly, lightweight You Only Look Once (YOLO) model with batch normalization (BN)-preserving quantization method that reduces the model size while maintaining accuracy at low precision. In addition, we propose a layer-specific resource–latency-aware DSE (LSRLA-DSE) method that selects the optimal dataflow based on layer-wise features and searches hardware design parameters under latency and resource constraints. Furthermore, we propose a FIFO-based streaming architecture with layer-wise dynamic dataflows that maintains high processing element (PE) utilization while minimizing off-chip traffic. Moreover, we introduce a semantic partition and regrouping strategy (SPRG) that improves resource efficiency and throughput. Implemented on a Xilinx ZCU104 FPGA, AHCO-YOLO-T achieves 79.8 FPS at 64.8% mAP, delivering 41.9 FPS/W and 80.5 GOPS/W. Across comparisons with existing YOLO accelerators, AHCO-YOLO achieves state-of-the-art efficiency, demonstrating suitability for real-time, energy-efficient object detection on edge platforms.

For latency-critical transactional applications, durability is often what limits performance. That is, executing transactions is fast, but guaranteeing that they are durable is slow. As a result, most of each transaction's latency is attributable to durability. To address this problem, some database systems allow applications to sacrifice durability guarantees in exchange for lower transaction latencies. These ad hoc techniques are effective, but they can make it difficult for applications to understand and manage the risks associated with failures. In this paper, our goal is to offer a more principled foundation for these kinds of performance/durability tradeoffs. The major obstacle to doing this is the transaction model itself, because it couples transaction durability with transaction commit. That is, the model defines a single point at which a transaction becomes visible and durable. This forces all transaction guarantees to wait for the slowest one, which is often durability. The primary contribution of this work is a new eventually durable transaction model, which decouples commit from durability. Transactions commit first, and become durable later. We argue for making this model the basis of the contract between transactional data systems and applications. We describe what it means to correctly implement eventually durable transactions, and consider how they can be exposed to applications. We also describe a prototype implementation of eventual durability in PostgreSQL, and show that it enables applications to reduce transaction latencies while managing the durability risks.

In remote sensing environments, image compression is essential to efficiently transmit and store high-resolution images due to the limited bandwidth and storage capacity. However, compression often leads to image quality degradation, requiring compression artifact removal technology in the postprocessing stage. Although deep neural networks have shown remarkable performance in image restoration, most existing methods have not adequately considered the compression conditions specific to remote sensing environments and have been evaluated primarily on synthetic datasets. To solve these issues, we propose a multidomain adaptive spatial–spectral variable compression artifact removal network (MASCAR) that effectively restores the earth surface details of compressed images in remote sensing environments. We introduce a multidomain local-patch collaborative learning strategy that extracts diverse features by decomposing the input local patch into different domains. In addition, we propose a detail focusing approach to direct the network’s focus toward fine-texture detail restoration and ensure stable training of remote sensing images with significant deviations in pixel distribution of local patches. Furthermore, a detail enhancement approach is presented to enhance the details of the restored images. Moreover, we propose an incorporated compressed image quality adaptation mechanism to respond flexibly to unknown compression ratios in remote sensing environments. The performance of MASCAR applied with the proposed method is evaluated on synthetic and real-world remote sensing datasets. Experimental results demonstrate that the proposed method has better quantitative performance and visual quality than existing methods.

The demand for multimedia data is rapidly increasing in many applications running on edge devices. The data compression method is essential for efficient communication in limited network bandwidth. However, the compressed data contains blocking artifacts that cause perceptual quality degradation and visual recognition problems. Recently, deep learning-based works have shown excellent progress in deblocking tasks. However, most previous works have used complex and deep network architectures that require high computational cost and large amounts of memory to improve performance. Therefore, deploying these networks as edge device applications is highly challenging work. In this paper, we propose an FPGA-based lightweight deblocking convolutional neural network (CNN) for edge devices. We present a lightweight CNN architecture to efficiently reduce blocking artifacts in the color domain. In addition, two optimization methods were introduced to further decrease the computation and memory requirement. Compared with CNNs proposed in other works, the number of MAC operations and the model size were reduced by approximately$\times 145$and$\times 2,300$, respectively. Finally, the proposed deblocking CNN has been implemented on a ZCU104 FPGA board. As a result of measuring the frame rate, it achieved 30.73 FPS.

Hot standby techniques are widely used to implement highly available database systems. These techniques make use of two separate copies of the database, an active copy and a backup that is managed by the standby. The two database copies are stored independently and synchronized by the database systems that manage them. However, database systems deployed in computing clouds often have access to reliable persistent storage that can be shared by multiple servers. In this paper we consider how hot standby techniques can be improved in such settings.