In LSM trees, background compaction tasks contend with foreground queries for CPU cycles, cache space, and also SAN bandwidth, if deployed in a disaggregated storage architecture. This study proposesNear-Data Compaction(NDC), which executes compaction on the storage node to utilize its underutilized computing resources. However, enabling NDC introduces several challenges. First, it has to support concurrent file access from both compute and storage nodes. In addition, it has to decide which compaction tasks to be executed on the storage node since the computing resources of a storage node are not unlimited. This study presentsTetherDB, an LSM tree for disaggregated storage architecture that addresses these challenges through lightweight dual-node coordination and selective NDC admission policies. Our evaluation demonstrates that TetherDB improves throughput by up to 2.1× compared to RocksDB in write-heavy workloads.

Self-supervised monocular depth estimation has been widely studied recently. Most of the work has focused on improving performance on benchmark datasets, such as KITTI, but has offered a few experiments on generalization performance. In this paper, we investigate the backbone networks (e.g., CNNs, Transformers, and CNN-Transformer hybrid models) toward the generalization of monocular depth estimation. We first evaluate state-of-the-art models on diverse public datasets, which have never been seen during the network training. Next, we investigate the effects of texture-biased and shape-biased representations using the various texture-shifted datasets that we generated. We observe that Transformers exhibit a strong shape bias and CNNs do a strong texture-bias. We also find that shape-biased models show better generalization performance for monocular depth estimation compared to texture-biased models. Based on these observations, we newly design a CNN-Transformer hybrid network with a multi-level adaptive feature fusion module, called MonoFormer. The design intuition behind MonoFormer is to increase shape bias by employing Transformers while compensating for the weak locality bias of Transformers by adaptively fusing multi-level representations. Extensive experiments show that the proposed method achieves state-of-the-art performance with various public datasets. Our method also shows the best generalization ability among the competitive methods.

Field trial measurements of small cell sleep mode with three different wake-up solutions are presented in this paper. A small heterogeneous LTE test network consisting of one macro base station and four small cells was deployed in Bundang, South Korea. Two UEs, one stationary and one moving, were used in the field trial. The studied solutions were (1) uplink interference based stand-alone small cell activation; (2) macro- assisted load-based small cell activation; and (3) macro-assisted load-based and timing advance-based small cell activation. Solution 1 was evaluated in a scenario where the same frequency band is used in the macro and small cell layers and the results show that Solution 1 provides an average reduction in energy consumption of 10% as well as a user throughput increase of 24% compared to a reference case without small cell sleep mode. For solutions 2 and 3 different frequency bands were used for the small cells and the macro layer and here the energy reduction gains were 16% and 23% respectively while the increase in user throughput was 28% and 51% respectively.