In order to solve the index calculation problems of high complexity, strong completeness and low efficiency in the filed of financial foreign exchange market monitoring, a novel distributed computing framework for foreign exchange market monitoring based on Spark big data structure was proposed. Firstly, the business characteristics and existing technology framework for foreign exchange market monitoring were analyzed and summarized. Secondly, the foreign exchange business features of single-market multi-indicator and multi-market multi-indicator were considered. Finally, based on Spark's Directed Acyclic Graph (DAG) job scheduling mechanism and resource scheduling pool isolation mechanism of YARN (Yet Another Recourse Negotiator), the Market-level DAG (M-DAG) model and the market-level resource allocation strategy named M-YARN (Market-level YARN) model were proposed, respectively. The experimental results show that, the performance of the proposed distributed computing framework for foreign exchange market monitoring improves the performance by more than 80% compared to the traditional technology framework, and can effectively guarantee the completeness, accuracy and timeliness of foreign exchange market monitoring indicator calculation under the background of big data.

Aiming at the problem of slow speed of the existing model-based Multivariate Time Series (MTS) clustering algorithm when dealing with MTS wtih unequal length, an improved clustering algorithm named MUltivariate Time Series Clustering Algorithm based on Lift Ratio (LR) Component Extraction (MUTSCA〈LRCE〉) was proposed. Firstly, the equal frequency discretization method was used to symbolize MTS. Then, the LR vector was calculated to express the temporal pattern between the dimensions of time series of MTS samples. Each LR vector was sorted and a fixed number of different key components were extracted from both ends. All the extracted key components were spliced to form a model vector for representing the MTS samples. The MTS sample set with unequal length was transformed into a model vector set with equal length. Finally, the k-means algorithm was used for the clustering analysis of generated model vector set with equal length. The experimental results on multiple common data sets show that, compared with the model-based MTS clustering algorithm named MUltivariate Time Series Clustering Algorithm〈LR〉(MUTSCA〈LR〉), the proposed algorithm can significantly improve the clustering speed of MTS data sets with unequal length under the premise of guaranteeing clustering effect.