Riso-Tree: An Efficient and Scalable Index for Spatial Entities in Graph Database Management Systems

With the ubiquity of spatial data, vertexes or edges in graphs can possess spatial location attributes side by side with other non-spatial attributes. For instance, as of June 2018, the Wikidata knowledge graph contains 48,547,142 data items (i.e., vertexes) and 13% of them have spatial location attributes. The article proposes Riso-Tree, a generic efficient and scalable indexing framework for spatial entities in graph database management systems. Riso-Tree enables the fast execution of graph queries that involve different types of spatial predicates (GraSp queries). The proposed framework augments the classic R-Tree structure with pre-materialized sub-graph entries. The pruning power of R-Tree is enhanced with the sub-graph information. Riso-Tree partitions the graph into sub-graphs based on their connectivity to the spatial sub-regions. The proposed index allows for the fast execution of GraSp queries by efficiently pruning the traversed vertexes/edges based upon the materialized sub-graph information. The experiments show that the proposed Riso-Tree achieves up to two orders of magnitude faster execution time than its counterparts when executing GraSp queries on real graphs (e.g., Wikidata). The strategy of limiting the size of each sub-graph entry ( PN max ) is proposed to reduce the storage overhead of Riso-Tree. The strategy can save up to around 70% storage without harming the query performance according to the experiments. Another strategy is proposed to ensure the performance of the index maintenance (Irrelevant Vertexes Skipping). The experiments show that the strategy can improve performance, especially for slow updates. It proves that Riso-Tree is useful for applications that need to support frequent updates.

Location based Continuous Query Processing over Geo-streaming Data

In recent years, many data-intensive and location based applications have emerged that need to process stream data in applications such as network monitoring, telecommunications data management, and sensor networks. Unlike regular queries, a continuous query exists for certain period of time and need to be continuously processed during this time. The algorithms used for data processing for the traditional database systems are not suited to tackle complex and various continuous queries over dynamic streaming data. The indexing for finite queries is preferred to indexing on infinite data to avoid expensive operations of index maintenance. Previous related work focused on moving queries on static objects or static queries on moving object. But now-a-days queries as well as objects are dynamic. So, hybrid indexing for queries significantly reduces the space costs and scales well with the increasing data. To deal with the speed of unbounded data, it is necessary to use data parallelism in query processing. The data parallelism in query processing offers better performance, availability and scalability.

BLEH: Bit-Less Extendible Hashing for DBMS and Hard Disk Drives

Indexing techniques such as extendible hashing and B-trees are widely used to store, retrieve and search for data on files in most file systems. These techniques have been comprehensively explored to enhance the data structure for increasing the faster access to file contents. Extendible hashing is a dynamic hashing technique which handles dynamic files that keep changing in size. Traditional extendible hashing uses bit addresses to hash the data to buckets and restricts the directory size to be a power of 2 which has corresponding complications in implementation. Restriction on directory size also results in uneven distribution of data which increases the possibility of overflows. This in turn increases the cost of index maintenance. In this paper, an efficient and simpler to implement variation of Extendible hashing method named Bit-Less Extendible Hashing (BLEH) for dynamic files is proposed. The proposed method eliminates the need for binary representation of the hash address which reduces the complexity of implementation. Furthermore, it eliminates the need for the directory size to be a power of 2 providing flexibility in the choice of initial directory size. The experimental results show that the proposed method provides better performance in terms of split count upon insertion when compared to the traditional extendible hashing method with good space utilization.

Sensex and Nifty Indices: Are They the Right Benchmarks for Mutual Funds in India?

Recently two significant developments took place in the Indian capital markets: (a) SEBI’s decision making it mandatory for all mutual funds to disclose the scheme returns against a common benchmark index such as Nifty or Sensex and (b) Employee’ Provident Fund Organisation (EPFO) is permitted to invest a part of their funds into stock market through the exchange-traded fund (ETF) route, particularly SBI Sensex and SBI Nifty ETFs. Both the developments are tied by a common concept that stock market indices such as Nifty and Sensex are passive without any statistically significant alpha. In the fund management industry, alpha is a measure of the risk-adjusted excess returns from a portfolio that can be attributed to the stock-picking skills of a fund manager. In this article, an attempt is made to examine for the presence of significant alphas in the returns of both the indices. The results of the study indicate that both the indices have statistically significant excess returns, raising questions on their suitability to act as reference and/or benchmarks for evaluating performance of mutual funds in India. Further, the study examined the returns of Sensex and Nifty index ETFs and observed a statistically significant alpha. The results of the study have important implications not only for the index construction companies but also to the policymakers who are advocating investment of considerable amounts of provident fund money into stock market through ETFs linked to Sensex and Nifty. Index maintenance companies have to re-design the indices so that they remain passive and the EPFO Administration may rethink their decision to invest in the existing ETFs linked to the Sensex and Nifty indices, and should consider constructing a well-diversified stock portfolio that is truly passive so that their mandate to get exposure only to market risk is fulfilled.