Mobile synchronization framework to enhance large object management in mobile cloud storage service

Supporting large file upload and retrieval is crucial for the mobile cloud storage services, as file sizes are trending larger and mobile users’ access or share files in large size. These files range from large PDFs to media files in various formats such as MPEG videos. These files need to be used offline, sometimes updated and shared among users. The mobile environment with frequent disconnections and limited bandwidth, impact the data and transaction management as well as the data consistency guarantees. Most of the mobile frameworks use a data-sync paradigm in order to handle disconnected operations, in which data is stored locally on the device and replicated to the cloud asynchronously. Many applications are storing large amounts of structured and unstructured data. However, it is challenging for mobile synchronization frameworks to manage data consistency in such environment. This is due to the fact that handling large data and maintaining consistency are very challenging in cases of local storage and updates. This research proposes an enhanced cloud-based mobile synchronization framework to address the two main problems that is data management for large objects and end-to-end support for data consistency for large objects. This research work proposes techniques to improve large file object access and synchronization in mobile cloud environment based on segmentation and object chunking. This framework mainly focuses on large file handling and providing support for both table and objects data models that can be tuned for three consistency semantics: resembling strong, causal and eventual consistency. Experimental results conducted using representative workloads showed that the proposed enhanced Mobile Synchronization framework can handle large files with the size ranging from 100MiB up to 1GiB and is able to reduce synchronization time with object chunking (2, 4, 8 and 16 MiB) in the experiment settings by 65.4% for upload and 93.7% for download on average when compared to other frameworks.

Text FSKTM 2022 1 IR.pdf Download (1MB)

Actions (login required)

View Item