Data Corruption

About: Data Corruption is a research topic. Over the lifetime, 435 publications have been published within this topic receiving 6784 citations.

Preventing data corruption due to pre-existing split brain

Abstract: Various methods, systems, and processes to prevent data corruption caused by a pre-existing split brain condition in a cluster are presented. In response to determining that a node is no longer part of a sub-cluster, another node in the sub-cluster is designated as a racer node. The racer node accesses a matrix, and the matrix includes information to determine whether a majority of coordination points in a cluster are accessible by nodes in the sub-cluster. Based on the accessing, a determination is made that the information indicates that the majority of coordination points are accessible by the nodes. The information is then broadcasted.

Auditing data integrity

Abstract: Various approaches are described for auditing integrity of stored data. In one approach, a data set is provided from a client to a storage provider, and the data set is stored at a first storage arrangement by the storage provider. An auditor determines whether the data set stored at the first storage arrangement is corrupt without reliance on any part of the data set and any derivative of any part of the data set stored by the client. While the auditor is determining whether the data set stored at the first storage arrangement is corrupt, the auditor is prevented from being exposed to information specified by the data set. The auditor outputs data indicative of data corruption in response to determining that the data set stored at the first storage arrangement is corrupt.

Public integrity auditing for dynamic regenerating code based cloud storage

Abstract: To protect cloud data against corruption, enabling data integrity checking along with error recovery becomes quite critical. Recently, regenerating code has caught researchers' attention due to low repair traffic while preserving fault tolerance. Existing integrity checking schemes for regenerating code storage only support static data and have high auditing cost, which are not suitable for practical application. In this work, We propose a public auditing scheme for regenerating code based cloud storage supporting data dynamics. Our scheme allows the auditor to check the integrity of data stored across multiple servers and identify the location of corrupted data when data corruption is detected within one round. To support data dynamics, we design a novel update method based on matrix and index mechanism. The update method support multiple update requests performed in batch way without downloading and re-coding the data file and re-computing the data tags, such that the communication and computation cost can be greatly reduced. Theoretic analysis and experimental results show the security and efficiency of our proposed scheme.

Tvarak: software-managed hardware offload for redundancy in direct-access NVM storage

Abstract: Production storage systems complement device-level ECC (which covers media errors) with system-checksums and cross-device parity. This system-level redundancy enables systems to detect and recover from data corruption due to device firmware bugs (e.g., reading data from the wrong physical location). Direct access to NVM penalizes software-only implementations of system-level redundancy, forcing a choice between lack of data protection or significant performance penalties. We propose to offload the update and verification of system-level redundancy to Tvarak, a new hardware controller co-located with the last-level cache. Tvarak enables efficient protection of data from such bugs in memory controller and NVM DIMM firmware. Simulation-based evaluation with seven data-intensive applications shows that Tvarak is efficient. For example, Tvarak reduces Redis set-only performance by only 3%, compared to 50% reduction for a state-of-the-art software-only approach.