15.30, room 445, PCRI

Abstract
We introduce the database fragment of RDF, which extends the popular Description Logic fragment, in particular with support for incomplete information. We then provide novel sound and complete saturation- and reformulation-based techniques for answering the Basic Graph Pattern queries of SPARQL in this fragment. Notably, we extend the state of the art on pushing RDF query processing within robust / efficient relational database management systems. Finally, we experimentally compare our query answering techniques using well-established datasets.

14.00, room 455, PCRI

Abstract
An increasing part of the world’s data is either shared through the Web or directly produced through and for Web platforms, in particular using structured formats like XML. Cloud platforms are interesting candidates to handle large data (and document) repositories, due to their elastic scaling properties. We present NubeX, a scalable store for managing a large corpora of XML documents built on top of off-the-shelf cloud infrastructure. NubeX implements smart indexes to both speed up query evaluation and reduce the warehouse operating cost, by helping direct queries to the documents which may have answers to them. We present the architecture, indexing strategies, and extensive experiments demonstrating their performance.

14.00, room 435, PCRI

Abstract
Cloud computing has been massively adopted recently in many applications for its elastic scaling and fault-tolerance. At the same time, given that the amount of available RDF data sources on the Web increases rapidly, there is a constant need for scalable RDF data management tools. In this paper we propose a novel architecture for the distributed management of RDF data, exploiting an existing commercial cloud infrastructure, namely Amazon Web Services (AWS). We study the problem of indexing RDF data stored within AWS, by using SimpleDB, a key-value store provided by AWS for small data items. The goal of the index is to efficiently identify the RDF datasets which may have answers for a given query, and route the query only to those. We devised and experimented with several indexing strategies; we discuss experimental results and avenues for future work.

14.00, room 455, PCRI

Abstract
Stratosphere is an open-source platform for information management in the cloud, co-developed by TU Berlin. In this talk, I will first introduce the Stratosphere architecture, programming model, and design decisions. Then, I will present some ongoing research on optimizing data flows that consist of black-box operators with unknown semantics, and efficiently executing iterative algorithms in a data flow system.

14.00, room 435, PCRI

Abstract
The term ‘cloud computing’ is nowadays synonymous to computing services offered by large-scale infrastructures. The key to the success of cloud computing is to provide seamless and efficient management of large dynamic disseminated data collections, such as scientific data, in order to maximize their availability while minimizing capital expenditure. This talk leverages lessons learned from financial management to solve the problem of both cost- and time-efficient management on clouds offering online data services. We propose a novel economy model for a cloud where users pay on-the-go for the data services they receive and user payments can be used for service provision, infrastructure maintenance and profit. The economy employs a cost model that takes into account all the available resources in a cloud, such as disk space and I/O operations, CPU time and network bandwidth. In order to ensure the economic viability of the cloud, the cost of offering new services has to be amortized to prospective users that will use them. We propose a novel cost amortization model that predicts the extent of amortization in time and number of users. The economy is completed with a dynamic pricing scheme that achieves optimal cloud profit while ensuring user satisfaction with service prices. The talk concludes with future research directions on the provision of online data services.