Parallel Data Update Synchronization Without Transfer Locks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for updating data across multiple computing devices often require serial synchronization, leading to inefficiencies and potential errors due to the need for data transfer locks, which can cause devices to remain idle and increase the time required for updates.
Innovation Solution
A system and method that allow multiple computing devices to receive data updates in parallel without establishing a data transfer lock, using a synchronization manager to send and receive updates from a server site, thereby reducing the demand on the lock and enabling simultaneous updates across devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If serial synchronization with data transfer lock is used, then data consistency is maintained, but update time and device idle time increase
Solution Approach 1:
The patent segments the synchronization process into two distinct phases: a locking phase for writing update data to the server (ensuring data consistency) and an unlocked parallel reading phase for distributing data to multiple devices (reducing time loss). This segmentation allows different devices to operate in parallel during the reading phase while maintaining consistency guarantees during the writing phase.
Solution Approach 2:
The patent implements preliminary action by establishing the data transfer lock only for the critical writing operation to the server, then releasing it before the parallel distribution phase. This preliminary locking ensures data consistency is established before parallel operations begin, allowing subsequent devices to read without waiting for locks.
2Reliability
If data transfer lock is established for each update, then data conflicts are prevented, but device productivity decreases
Solution Approach 1:
The patent segments the synchronization operations into a locked writing phase (single device uploading to server) and an unlocked reading phase (multiple devices downloading in parallel). This segmentation prevents conflicts during writing while enabling high productivity during distribution, as multiple devices can simultaneously receive updates without acquiring locks.
Solution Approach 2:
The server acts as an intermediary that receives update data from one device under lock protection, then distributes this data to multiple other devices without requiring them to acquire locks. This intermediary approach isolates the locking requirement to only the critical update submission point, while enabling parallel productivity during distribution.
3Productivity
If parallel updates are implemented, then update speed increases, but data transfer lock demand management becomes complex
Solution Approach 1:
The patent extracts the data transfer lock requirement from the parallel distribution phase and concentrates it only in the update submission phase to the server. By taking out the locking mechanism from the parallel operations, the system achieves high update speed through parallelism while simplifying lock management to a single critical section rather than coordinating locks across multiple parallel operations.
Data Source
AI summary
A system includes a server site that includes a memory for storing update data sets that correspond to data sets stored on multiple computing devices of a user. The system also includes a synchronization manager for determining that one computing device associated with the user and another computing device associated with the user are absent one or more data updates stored in the memory at the server site. The synchronization manager is configured to send in parallel, absent establishing a data transfer lock, the one or more data updates to the both computing devices of the user for updating the corresponding data stored on each computing device.