Content-aware compression for big textual data analysis

A substantial amount of information on the Internet is present in the form of text. The value of this semi-structured and unstructured data has been widely acknowledged, with consequent scientific and commercial exploitation. The ever-increasing data production, however, pushes data analytic platforms to their limit. This thesis proposes techniques for more efficient textual big data analysis suitable for the Hadoop analytic platform. This research explores the direct processing of compressed textual data. The focus is on developing novel compression methods with a number of desirable properties to support text-based big data analysis in distributed environments. The novel contributions of this work include the following. Firstly, a Content-aware Partial Compression (CaPC) scheme is developed. CaPC makes a distinction between informational and functional content in which only the informational content is compressed. Thus, the compressed data is made transparent to existing software libraries which often rely on functional content to work. Secondly, a context-free bit-oriented compression scheme (Approximated Huffman Compression) based on the Huffman algorithm is developed. This uses a hybrid data structure that allows pattern searching in compressed data in linear time. Thirdly, several modern compression schemes have been extended so that the compressed data can be safely split with respect to logical data records in distributed file systems. Furthermore, an innovative two layer compression architecture is used, in which each compression layer is appropriate for the corresponding stage of data processing. Peripheral libraries are developed that seamlessly link the proposed compression schemes to existing analytic platforms and computational frameworks, and also make the use of the compressed data transparent to developers. The compression schemes have been evaluated for a number of standard MapReduce analysis tasks using a collection of real-world datasets. In comparison with existing solutions, they have shown substantial improvement in performance and significant reduction in system resource requirements.

Kinetics and coverage dependent reaction mechanisms of the copper atomic layer deposition from copper dimethylamino-2-propoxide and diethylzinc

Atomic layer deposition (ALD) has been recognized as a promising method to deposit conformal and uniform thin film of copper for future electronic devices. However, many aspects of the reaction mechanism and the surface chemistry of copper ALD remain unclear. In this paper, we employ plane wave density functional theory (DFT) to study the transmetalation ALD reaction of copper dimethylamino-2-propoxide [Cu(dmap)2] and diethylzinc [Et2Zn] that was realized experimentally by Lee et al. [ Angew. Chem., Int. Ed. 2009, 48, 4536−4539]. We find that the Cu(dmap)2 molecule adsorbs and dissociates through the scission of one or two Cu–O bonds into surface-bound dmap and Cu(dmap) fragments during the copper pulse. As Et2Zn adsorbs on the surface covered with Cu(dmap) and dmap fragments, butane formation and desorption was found to be facilitated by the surrounding ligands, which leads to one reaction mechanism, while the migration of ethyl groups to the surface leads to another reaction mechanism. During both reaction mechanisms, ligand diffusion and reordering are generally endothermic processes, which may result in residual ligands blocking the surface sites at the end of the Et2Zn pulse, and in residual Zn being reduced and incorporated as an impurity. We also find that the nearby ligands play a cooperative role in lowering the activation energy for formation and desorption of byproducts, which explains the advantage of using organometallic precursors and reducing agents in Cu ALD. The ALD growth rate estimated for the mechanism is consistent with the experimental value of 0.2 Å/cycle. The proposed reaction mechanisms provide insight into ALD processes for copper and other transition metals.