Network Device Data Replication via Deterministic Transmission
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data center solutions for mission-critical data replication face challenges in reducing latency and ensuring reliable replication across geographically separate locations, leading to potential data loss and prolonged recovery times during failure scenarios.
Innovation Solution
Implementing a network device that uses deterministic and lossless transmission media, such as optical networks, to replicate data across remote locations, allowing for immediate acknowledgement of data storage completion independent of confirmation from remote devices, and employing mini-block transmission with error correction mechanisms to ensure reliable replication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data replication waits for confirmation from remote network devices before acknowledging storage completion, then data replication reliability is improved, but latency and recovery time are increased
Solution Approach 1:
The system performs preliminary data storage and transmission actions before confirmation is received. The network device writes data to local storage, transmits it to remote devices, and sends acknowledgments to clients before receiving confirmation of successful remote replication, based on predetermined latency characteristics of the transmission medium
Solution Approach 2:
The patent introduces an intermediary mechanism - predetermined latency characteristics of the transmission medium - that allows the system to calculate and account for transmission time without waiting for actual confirmation. This intermediary enables the acknowledgment to be sent based on expected rather than actual completion time
2Reliability
If data is transmitted to geographically separate remote network devices, then data resiliency is improved, but transmission latency and recovery time are increased
Solution Approach 1:
The system performs preliminary data transmission to geographically separate remote network devices before acknowledgment is required. By calculating predetermined latency characteristics of the transmission medium, the system can send acknowledgments immediately after transmission initiation, reducing perceived recovery time while maintaining geographic distribution for resiliency
3Reliability
If acknowledgement is sent only after confirming successful remote replication, then data loss risk is reduced, but Recovery Point Objective duration is increased
Solution Approach 1:
The system performs preliminary storage and transmission actions, then sends acknowledgments based on predetermined latency calculations rather than waiting for actual confirmation. This allows the system to maintain data loss prevention through careful latency accounting while significantly reducing RPO duration
Solution Approach 2:
The patent changes the parameter for acknowledgment timing from actual confirmation receipt to calculated expected completion time based on predetermined latency characteristics. This parameter change enables faster acknowledgments while maintaining reliability through the deterministic nature of the transmission medium
Data Source
AI summary
A network device includes at least one processor, a storage device and at least one memory including computer program code. The at least one memory and the computer program code are configured to, with the at least one processor, cause the network device to: write data to the storage device; replicate the data at one or more remote network devices via at least one deterministic transmission medium; and output an acknowledgement in response to determining that the data has been written to the storage device prior to receiving confirmation of successful replication of the data at the one or more remote network devices. The methods, systems or computer readable mediums leverage the deterministic and measurable nature of the transmission media to reduce the Recover Point Objective durations.