Google services continuously generate vast amounts of application data. This data provides valuable insights to business users. We need to store and serve these planet-scale data sets under the extremely demanding requirements of scalability, sub-second query response times, availability, and strong consistency; all this while ingesting a massive stream of updates from applications used around the globe. We have developed and deployed in production an analytical data management system, Napa, to meet these requirements. Napa is the backend for numerous clients in Google. These clients have a strong expectation of variance-free, robust query performance. At its core, Napa's principal technologies for robust query performance include the aggressive use of materialized views, which are maintained consistently as new data is ingested across multiple data centers. Our clients also demand flexibility in being able to adjust their query performance, data freshness, and costs to suit their unique needs. Robust query processing and flexible configuration of client databases are the hallmark of Napa design. Most of the related work in this area takes advantage of full flexibility to design the whole system without the need to support a diverse set of preexisting use cases. In comparison, a particular challenge we faced is that Napa needs to deal with hard constraints from existing applications and infrastructure, so we could not do a "green field" system, but rather had to satisfy existing constraints. These constraints led us to make particular design decisions and also devise new techniques to meet the challenges. In this paper, we share our experiences in designing, implementing, deploying, and running Napa in production with some of Google's most demanding applications.

The proliferation of social networks in digital media has proved to be fruitful, but this rise in popularity is accompanied by user privacy concerns. Social network data has been published in various ways and preserving the privacy of individuals in the published data has become an important concern. Several algorithms have been developed for privacy preservation in relational data, but these algorithms cannot be applied directly to social networks as the nodes here have structural properties along with labels. In this paper, we propose an algorithm to achieve k-anonymity and l-diversity in social network data which provides structural anonymity along with sensitive attribute protection. The proposed algorithm uses novel edge addition techniques which are also presented in this paper. We also propose a concept of partial anonymity to reduce anonymization cost for d>1. The empirical study shows that our algorithm requires significantly less number of edge additions for anonymization of social network data and has a substantially lower running time than the other algorithms previously proposed in the field.

Fuzzy logic deals with partial truth. A fuzzy based approach to blog analysis, on the basis of various feature words, allows us to determine the degree to which a blogger's style belongs to a particular age or gender group. Each blog was represented by a set of normalized word frequencies of selected feature words in it. Using membership values obtained from applying Fuzzy C-Means (FCM) algorithm to these blog representations, we can call the blogger's style to belong weakly, fairly, strongly or very strongly to a particular class. The advantage of using fuzzy logic for this problem is that a weak belonging to a particular class means that there is a decent belonging to the other class (es). Hence when a search or query is carried out, no useful blog will be left out of the results for that other class (es).