Du Xiaoyong

Parallel Techniques for Large Data Analysis in a Futures Trading Evaluation Service System

Futures trading evaluation system is used to analyze trading history of individuals, to find out the root cause of profit and loss, so that investors can learn from their past and make better decisions in the future. To analyze trading history of investors, the system processes a large volume of transaction data, to calculate key performance indicators, as well as time series behavior patterns, finally concludes recommendations with the help of an expert knowledge base. The paper firstly presents the working logic of the evaluation system, then it focuses on parallel data processing techniques that the system is based on. Parallel processing architecture, data distribution scheme, key performance indicators calculating algorithms and distributed time series analysis algorithms are elaborated in details. The system is highly scalable, and by exploiting the power of parallel processing, the generation time of an evaluation report is cut down from 1 to 3 minute, to 30 to 45 seconds.

Cluster Based Partition for Multi-dimensional Range Query in DAS Model

Database-as-a-Service (DAS) is an emerging database management paradigm wherein querying on encrypted data directly is a performance critical problem, to which partition based index is an effective solution. For multi-dimensional range query, generating partition on each dimension would increase information leakage to a large extent, while previous multi-dimensional partition would cause large efficiency loss, especially when the data distribution is sparse. To achieve guaranteed security with much less efficiency loss, in this paper we propose cluster based multi-dimensional partition (CBMP). First, CBMP decompose the whole space into clusters, which only cover non-empty area. To get better clustering effect, a new cluster criteria based on full neighbor is proposed. Second, since optimal secure partition is NP-hard, several heuristic based algorithms including distance based and Hilbert based are proposed. Experiments on real dataset and synthetic dataset show that distance based algorithm could achieve approximately least efficiency loss.