Data access

About: Data access is a research topic. Over the lifetime, 13141 publications have been published within this topic receiving 172859 citations. The topic is also known as: Data access.

Characterizing and enhancing global memory data coalescing on GPUs

Abstract: Effective parallel programming for GPUs requires careful attention to several factors, including ensuring coalesced access of data from global memory. There is a need for tools that can provide feedback to users about statements in a GPU kernel where non-coalesced data access occurs, and assistance in fixing the problem. In this paper, we address both these needs. We develop a two-stage framework where dynamic analysis is first used to detect and characterize uncoalesced accesses in arbitrary PTX programs. Transformations to optimize global memory access by introducing coalesced access are then implemented, using feedback from the dynamic analysis or using a model-driven approach. Experimental results demonstrate the use of the tools on a number of benchmarks from the Rodinia and Polybench suites.

Distributed data management for large volume visualization

Abstract: We propose a distributed data management scheme for large data visualization that emphasizes efficient data sharing and access. To minimize data access time and support users with a variety of local computing capabilities, we introduce an adaptive data selection method based on an "enhanced time-space partitioning" (ETSP) tree that assists with effective visibility culling, as well as multiresolution data selection. By traversing the tree, our data management algorithm can quickly identify the visible regions of data, and, for each region, adaptively choose the lowest resolution satisfying user-specified error tolerances. Only necessary data elements are accessed and sent to the visualization pipeline. To further address the issue of sharing large-scale data among geographically distributed collaborative teams, we have designed an infrastructure for integrating our data management technique with a distributed data storage system provided by logistical networking (LoN). Data sets at different resolutions are generated and uploaded to LoN for wide-area access. We describe a parallel volume rendering system that verifies the effectiveness of our data storage, selection and access scheme.

Automatic folder access management

Abstract: Methods and systems are provided for decentralizing user data access rights control activities in networked organizations having diverse access control models and file server protocols. A folder management application enables end users of the file system to make requests for access to storage elements, either individually, or by becoming members of a user group having group access privileges. Responsibility for dealing with such requests is distributed to respective group owners and data owners, who may delegate responsibility to authorizers. The application may also consider automatically generated proposals for changes to access privileges. An automatic system continually monitors and analyzes access behavior by users who have been pre-classified into groups having common data access privileges. As the organizational structure changes, these groups are adaptively changed both in composition and in data access rights.

Comparison-based encryption for fine-grained access control in clouds

Abstract: Access control is one of the most important security mechanisms in cloud computing. However, there has been little work that explores various comparison-based constraints for regulating data access in clouds. In this paper, we present an innovative comparison-based encryption scheme to facilitate fine-grained access control in cloud computing. By means of forward/backward derivation functions, we introduce comparison relation into attribute-based encryption to implement various range constraints on integer attributes, such as temporal and level attributes. Then, we present a new cryptosystem with dual decryption to reduce computational overheads on cloud clients, where the majority of decryption operations are executed in cloud servers. We also prove the security strength of our proposed scheme, and our experiment results demonstrate the efficiency of our methodology.

Secure data outsourcing

Abstract: The networked and increasingly ubiquitous nature of today's data management services mandates assurances to detect and deter malicious or faulty behavior. This is particularly relevant for outsourced data frameworks in which clients place data management with specialized service providers. Clients are reluctant to place sensitive data under the control of a foreign party without assurances of confidentiality. Additionally, once outsourced, privacy and data access correctness (data integrity and query completeness) become paramount. Today's solutions are fundamentally insecure and vulnerable to illicit behavior, because they do not handle these dimensions. In this tutorial we will explore how to design and build robust, efficient, and scalable data outsourcing mechanisms providing strong security assurances of (1) correctness, (2) confidentiality, and (3) data access privacy. There exists a strong relationship between such assurances; for example, the lack of access pattern privacy usually allows for statistical attacks compromising data confidentiality. Confidentiality can be achieved by data encryption. However, to be practical, outsourced data services should allow expressive client queries (e.g., relational joins with arbitrary predicates) without compromising confidentiality. This is a hard problem because decryption keys cannot be directly provided to potentially untrusted servers. Moreover, if the remote server cannot be fully trusted, protocol correctness become essential. Therefore, solutions that do not address all three dimensions are incomplete and insecure.