Hyper-Converged Infrastructure Node Upgrade Sequencing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional hyper-converged infrastructure (HCI) maintenance processes are inefficient due to sequential node upgrades, which do not account for data synchronization and resource distribution, leading to prolonged downtime and increased risk of faults during software updates.
Innovation Solution
An analytics-based approach utilizing predictive analytics and a node sequencing algorithm to optimize the upgrade sequence of HCI nodes, considering data gravity, resource availability, and network bandwidth, allowing for simultaneous upgrades of compatible nodes to minimize downtime and ensure high availability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If sequential upgrade schedule is used for HCI nodes, then data availability is maintained, but maintenance duration is excessively long
Solution Approach 1:
The patent segments the upgrade process by dividing nodes into groups based on data overlap analysis. Nodes with no data overlap are grouped together and can be upgraded simultaneously, while nodes with overlapping data are upgraded sequentially. This segmentation allows parallel processing of independent nodes, reducing total maintenance duration while maintaining data availability through proper grouping strategies.
Solution Approach 2:
The patent performs preliminary analysis of data placement and overlaps before generating the upgrade schedule. By pre-identifying which nodes have overlapping data and which don't, the system can plan parallel upgrade groups in advance. This preliminary action enables optimization of the upgrade sequence without compromising data availability, as the schedule is generated with full knowledge of data relationships.
2Stability of the object's composition
If conventional sequential maintenance process is used, then system stability is maintained, but productivity during maintenance is reduced
Solution Approach 1:
The patent introduces dynamic scheduling that adapts to the specific data placement configuration of each cluster. Rather than a fixed sequential process, the system dynamically generates upgrade schedules based on real-time analysis of data overlaps and node relationships. This dynamic approach maintains system stability by respecting data dependencies while maximizing productivity through parallel execution of independent node upgrades.
Solution Approach 2:
The patent changes the parameter of upgrade execution from strictly sequential to conditional parallel. By analyzing data overlap parameters and using this information to determine which nodes can be upgraded simultaneously, the system transforms the maintenance process from a single-threaded sequential operation to a multi-threaded parallel operation where possible, thereby improving productivity without sacrificing stability.
3Loss of time
If nodes with overlapping data are upgraded simultaneously, then maintenance time is reduced, but data unavailability and faults increase
Solution Approach 1:
The patent incorporates feedback mechanisms that analyze data placement information and use this feedback to determine safe parallel upgrade groups. The system continuously evaluates data overlap relationships and adjusts the upgrade schedule accordingly, only allowing parallel upgrades of nodes that have been feedback-confirmed to have no data dependencies on each other. This feedback-driven approach enables time reduction through parallelism while preventing data unavailability.
4Device complexity
If sequential node upgrade is performed, then data synchronization is simplified, but total maintenance time increases significantly
Solution Approach 1:
The patent segments the cluster into independent upgrade groups based on data synchronization relationships. By identifying nodes that can be upgraded in parallel without affecting each other's data synchronization, the system reduces the overall upgrade time while maintaining manageable synchronization complexity within each segment. Nodes within the same group are synchronized independently, preventing complex cross-group synchronization issues.
Data Source
AI summary
Analytics-based optimized maintenance operations for a hyper-converged infrastructure are described. An example includes instructions to establish an order for a sequence of maintenance operations including collecting data points that relate to high availability of multiple nodes in a storage infrastructure; performing analysis of the collected data, including discovery of groups of nodes that don't have mutual relations with other paired nodes; receiving a request to perform a maintenance operation for the plurality of nodes; generating an ordered sequence of groups of nodes for the maintenance operation based at least in part on the analysis of the collected data, each group including one or more nodes; and performing the maintenance operation for the plurality of nodes according to the sequence of groups of nodes, wherein the maintenance operation includes a power cycle for each of the plurality of nodes.