LSM-tree has been widely used as a write-optimized storage engine in many key-value stores, such as LevelDB and RocksDB. However, conventional compaction operations on the LSM-tree need to read, merge, and write many SSTables, which we call Table Compaction in this paper. Table Compaction will cause two major problems, namely write amplification and block-cache invalidation. They will lower both write and read performance of the LSM-tree. To address these issues, we propose a novel compaction scheme named Block Compaction that adopts a block-grained merging policy to perform compaction operations on the LSM-tree. Block Compaction identifies the boundaries of data blocks and tries to avoid reusing data blocks, which not only reduces the write amplification but also alleviates the block-cache invalidation. We present cost analysis to theoretically demonstrate that Block Compaction is more efficient than the existing Table Compaction. Furthermore, we analyze the side-effects of Block Compaction and present three optimizations: (1) Selective Compaction is to reduce the space amplification of Block Compaction by integrating Table Compaction with Block Compaction. (2) Parallel Merging divides a compaction task into several sub-tasks and uses multiple workers to accomplish sub-tasks in parallel. (3) Lazy Deletion mitigates the overhead caused by traversing files at the tail of compaction operations. We implement a new key-value store named BlockDB based on Block Compaction and its optimizations. Then, we compare BlockDB with LevelDB, RocksDB, and L2SM using the YCSB benchmark. The results show that BlockDB can reduce write amplification up to 32% and running time by up to 43.6%, compared to its competitors. In addition, it can maintain the high performance for point lookups and range scans.

Playing Android games on Windows x86 PCs has gained enormous popularity in recent years, and the de facto solution is to use mobile emulators built with the AOVB (Android-x86 On VirtualBox) architecture. When playing heavy 3D Android games with AOVB, however, users often suffer unsatisfactory smoothness due to the considerable overhead of full virtualization. This paper presents DAOW, a game-oriented Android emulator implementing the idea of direct Android emulation, which eliminates the overhead of full virtualization by directly executing Android app binaries on top of x86-based Windows. Based on pragmatic, efficient instruction rewriting and syscall emulation, DAOW offers foreign Android binaries direct access to the domestic PC hardware through Windows kernel interfaces, achieving nearly native hardware performance. Moreover, it leverages graphics and security techniques to enhance user experiences and prevent cheating in gaming. As of late 2018, DAOW has been adopted by over 50 million PC users to run thousands of heavy 3D Android games. Compared with AOVB, DAOW improves the smoothness by 21% on average, decreases the game startup time by 48%, and reduces the memory usage by 22%.

Psoriasis is a chronic, recurrent, inflammatory skin disease, with varied incidence rates for various populations in different regions of the world. According to statistics, the prevalence of psoriasis in European countries is about 1% ∼ 3%, and 0.47% for China in 2008, which appears on the rise year by year. [1,2] Evidence suggests that genetic and environmental factors play important roles in the pathogenesis of psoriasis [1,2], but its exact etiology and pathogenesis are still unclear. Traditional Chinese medicine (TCM) treatment, based on syndrome differentiation, has a long history in treating this disease and achieved a significant effect in practice [3,4]. However, few studies on intelligent knowledge-based system (KBS), based on TCM treatment, for automatic reasoning and semantic retrieval have been reported, but much less for mature products supporting clinical practice. In order to inherit and develop the excellent tradition of TCM, with psoriasis as a use case, a knowledge base of psoriasis disease was tried to be constructed based on TCM Ontology combined with ontology, knowledge base and disease knowledge related to psoriasis in this paper, which can provide references for the construction of a shared and reusable diagnosis knowledge-base for the psoriasis diagnosis system of TCM experts.

Subcarrier allocation for multiuser MC-DS-CDMA systems is much more complicated when compared to OFDM systems due to the presence of multiple access interference (MAI) when multiple users transmit simultaneously on the same subcarrier. In this paper, we propose a novel suboptimal algorithm to adaptively allocate subcarriers for MC-DS-CDMA systems, with the effect of MAI taken into consideration. An earlier proposed algorithm minimizes the bit- error-rate (BER) of respective users rather than minimizes the overall system BER. Each user performs channel estimation, orders subcarrier channel gains and feedbacks an index vector to the base station (BS). BS uses these index vectors from all the users, together with subcarrier allocation coefficients of the previous updating stage, to adaptively generate subcarrier allocation coefficients of the current updating stage so that BER performance is optimized. However, since the assignment process depends on the earlier allocation coefficients of other users, an error can propagate and affect the decision to be made in the next stage. In the proposed approach, the problem is formulated using Quadratic Programming (QP), and a single objective function is used to optimize the overall system BER. The algorithm uses the actual values of channel gains and does not depend on any information from the previous updating stage. Thus, the discrepancy propagation is avoided. We use Kuhn-Tucker conditions to convert the formulated QP to a linear programming (LP) problem, and the solution can be obtained using standard LP solvers.

We propose a subcarrier allocation scheme for multiple access MC-DS-CDMA systems, where random signature sequences are used to identify the users. The optimal solution to such a subcarrier allocation problem has its complexity grows exponentially as the number of users and subcarriers grow. We proposed a sub-optimal method. In our proposal, each user performs channel measurement, orders the subcarrier channel gains and feedback the information in the order of the subcarriers channel gains to the base station. The base station uses these information from all users, together with the previous subcarrier assignment matrix, adaptively generates the next subcarrier assignment matrix for forward link transmission, with the objective to minimize the overall system BER performance. While selecting the subcarriers, we take into consideration the multiple access interference.