The dual-arm space robot usually forms a closed-chain constraint system with a target through collaborative operations when performing tasks. Most previous related research has focused on the performance of open-chain robots themselves. Studying the manipulation performance of a closed-chain robot system is of great significance. This article proposes a reconfigurable space robot (RSR) system for on-orbit servicing. A graph theory based framework for the automatic generation of reconfigurable robot models is proposed to address the characteristics of variable topology structures. Meanwhile, a new index – the closed-chain inertia matching index is proposed to evaluate its configuration effectively. Compared with traditional dynamic manipulability ellipsoid (DME) and manipulating force ellipsoid (MFE), the effectiveness of the proposed closed-chain inertia matching ellipsoid (IME) is verified. Compared with the DME and MFE, the IME effectively considers the influence of load and can effectively express the dynamic torque force/torque transmission efficiency from the joint actuator to the load in a closed-chain system. The IME can not only be used to determine the optimal joint configuration under a specific closed-chain configuration, but also to determine the optimal nonisomorphic configuration of a reconfigurable robot. Finally, the results of configuration optimization of the closed-chain dual-arm reconfigurable space robot are given.Note to Practitioners—This paper aims to study the determination of the optimal configuration for the RSR during task execution. Due to its various configurations, the RSR has a complex modeling process. This article proposes a modeling framework for automatically generating RSR models that can effectively achieve automatic modeling of various configurations. In addition, most of the previous related research focused on the performance of open-chain robots themselves, ignoring the impact of load. This article proposes a new closed-chain system performance indicator to evaluate the robot configuration. The proposed evaluation indicator effectively considers the influence of load, which is of practical significance to improve the efficiency of task execution. The effectiveness of the proposed index in determining the optimal configuration is demonstrated by simulation verification. In the future, we will focus on multi-arm collaborative operations under a specific configuration.

With the rapid growth of cloud computing, efficient management of multi-tenant databases has become a vital challenge for cloud service providers. It is particularly important for Alibaba, which hosts a distributed multi-tenant database supporting one of the world's largest e-commerce platforms. It serves tens of millions of sellers as tenants, and supports transactions from hundreds of millions of buyers. The inherent imbalance of shopping preferences from the buyers essentially generates a drastically skewed workload on the database, which could create unpredictable hotspots and consequently large throughput decline and latency increase. In this paper, we present the architecture and implementation of ESDB (ElasticSearch Database), a cloud-native document-oriented database which has been running on Alibaba Cloud for 5 years as the main transaction database behind Alibaba's e-commerce platform. ESDB provides strong full-text search and retrieval capability, and proposes dynamic secondary hashing as the solution for processing extremely skewed workloads. We evaluate ESDB with both simulated workloads and real-world workloads, and demonstrate that ESDB significantly enhances write throughput and reduces the completion time of writes without sacrificing query throughput.

Cloud backup is becoming the preferred way for users to support disaster recovery. In addition to its convenience, users are deeply concerned about reducing storage costs in the face of large-scale backup data. Data deduplication is an effective method for backup storage. However, current deduplicate methods lack the utilization of cloud resources to provide scalable backup service for cloud backup users, and cannot meet the biased preference for different backup versions. For new backup versions, users want higher deduplicate and restore speed to reduce the waiting time. Conversely, reducing storage costs is more necessary for old backup versions.In this paper, we present SLIMSTORE, with a cloud-based deduplication architecture that disassembles the system into a storage layer and a computing layer to support elastic utilization of cloud resources. We propose two types of processing nodes with different design focuses to meet the needs of cloud-based backup. The L-node exploits locality and similarity, and adopts a history-aware strategy to provide fast online deduplication service. L-node also optimizes online restoration to realize high restore efficiency. Meanwhile, the G-node provides exact deduplication offline for the old versions, and helps the restore performance of the new versions by optimizing their physical storage. We compare SLIMSTORE with some state-of-art deduplicate and restore methods. Experimental results show that SLIMSTORE can achieve fast deduplication, efficient restoration, and effective space reduction. Furthermore, SLIMSTORE attains scalable deduplication and restoration.