When: Friday, March 7, at 14.00

Where: PCRI building, room 445

Who: Benoit Groz

Title: Filtering with the Crowd in the CrowdScreen model

Abstract:
Filtering a set of items -to select photographs presenting certain properties for instance- is a common application of Crowd-Sourcing.
When the workers are error-prone, each item is presented to multiple users, to limit the probability of misclassification. Since the Crowd is a relatively expensive resource, minimizing the number of questions per item may naturally result in big savings. Several algorithms to address this minimization problem have been presented in the CrowdScreen framework by Parameswaran et al. However, those algorithms do not scale well and therefore cannot be used in scenarios where high accuracy is required in spite of high user error rates.
We propose optimizations of those as well as new theoretical insights that lead to scalable filtering algorithms.

When: Tuesday, March 4, at 14.00

Where: PCRI building, room 455

Who: Benjamin Nguyen

Title: Privacy-Preserving Query Execution using a Decentralized Architecture and Tamper Resistant Hardware

Abstract:
Current applications, from complex sensor systems (e.g. quantified self) to online e-markets acquire vast quantities of personal information which usually ends-up on central servers. Decentralized architectures, devised to help individuals keep full control of their data, hinder global treatments and queries, impeding the development of services of great interest. This paper promotes the idea of pushing the security to the edges of applications, through the use of secure hardware devices controlling the data at the place of their acquisition. To solve this problem, we propose secure distributed querying protocols based on the use of a tangible physical element of trust, reestablishing the capacity to perform global computations without revealing any sensitive information to central servers. There are two main problems when trying to support SQL in this context: perform joins and perform aggregations. In this paper, we study the subset of SQL queries without joins and show how to secure their execution in the presence of honest-but-curious attackers. Cost models and experiments demonstrate that this approach can scale to nationwide infrastructures.

Joint work with Quoc-Cuong To and Philippe Pucheral.

When: Thursday, February 6, at 15.00

Where: PCRI building, room 445

Who: Laurent Daynès, Oracle Labs

Title: Implementing language in Java with Alphabet Soup: how it may impact the Oracle Database

When: Friday, January 10, at 14.00

Where: PCRI building, room 445

Who: Luca Cabibbo, Università Roma Tre

Title: NoSQL Database Design for Next-Generation Web Applications

Abstract:
NoSQL database systems especially support the development of next-generation web applications, for which: (i) data of interest have a flexible structure, (ii) data access is based on simple read-write operations, (iii) foremost quality requirements are good horizontal scalability, response time, and availability, as well as a certain degree of consistency.

Although NoSQL databases are claimed to be “schemaless,” the design of data organization requires important decisions, to map application data to the modeling elements (collections, tables, documents, key-value pairs) available in the target datastore.
These decisions are significant, because of their impact on the above major quality requirements.
Given the high heterogeneity in the NoSQL world, this design activity is usually based on best practices and guidelines that are strictly related to the selected system, but with no systematic methodology.

In this talk we present NoAM (NoSQL Abstract Model), a logical approach to NoSQL database design, with initial activities that are independent of the specific target system.
The approach aims at exploiting the commonalities of the various NoSQL datastores.
It is based on an intermediate, abstract data model, to represent application data as collections of complex-value objects, called aggregates.
Aggregates are units of distribution (to support scalability) and consistency (to the extent it is needed); aggregates can be partitioned in smaller data elements (for the sake of performance).
The intermediate representation can be then implemented in target NoSQL systems, taking into account their specific features.

We then present ONDM (Object-NoSQL Datastore Mapper), a framework that supports the management of persistent objects in NoSQL database systems.
It provides application developers with a uniform programming interface, as well as the ability to map application data to different data representations, in a flexible way, towards a variety of NoSQL systems.
In particular, ONDM can be used to support developers in performing experiments that are usually needed during the design of a NoSQL database, to assess the most suitable design solution among a set of valid candidate (but alternative) representations.

Slides
Luca Cabibbo: NoSQL Database Design for Next-Generation Web Applications