Global Pool Control Unit Optimizing I/O Latency in HCI Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In Hyper-Converged Infrastructure (HCI) systems, the performance of I/O processing is degraded due to increased communication between nodes when the host Virtual Machine (VM), logical area owner SDS, and physical area owner SDS are configured on different nodes, leading to higher latency and reduced efficiency in data transfer.
Innovation Solution
A computer system architecture that includes a host unit, local control units managing local logical areas, and a global control unit managing global logical areas, where the global control unit optimizes data transmission based on commonality relationships among nodes, reducing the number of inter-node communications and improving processing efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the host VM, logical area owner SDS, and physical area owner SDS are arranged on different nodes to distribute load and expand storage capacity, then system scalability and load distribution are improved, but the performance of I/O processing is degraded due to increased inter-node communication
Solution Approach 1:
The patent introduces a data transfer unit as an intermediary component that is selectively placed either in the host node or in the storage node. This intermediary optimizes data transfer by determining the most efficient location for data movement operations, thereby reducing unnecessary inter-node communications while maintaining load distribution benefits. The data transfer unit acts as a mediator that intelligently routes data transfers to minimize latency and maximize I/O performance.
Solution Approach 2:
The system dynamically adjusts the placement of the data transfer unit between host node and storage node based on operational conditions. This dynamic configuration allows the system to adapt to varying workloads and access patterns, optimizing performance for different scenarios while preserving the scalability and load distribution capabilities of the distributed architecture.
2Quantity of substance
If the host VM, logical area owner SDS, and physical area owner SDS are arranged on different nodes, then storage capacity expansion is achieved, but the number of communication times between nodes increases leading to higher latency
Solution Approach 1:
The data transfer unit serves as an intermediary that consolidates and optimizes data transfer operations. By positioning this unit strategically in either the host or storage node, the system reduces the number of separate communication interactions between nodes, thereby decreasing cumulative latency while maintaining expanded storage capacity across multiple nodes.
Solution Approach 2:
The patent merges multiple data transfer operations into a single coordinated process managed by the data transfer unit. This consolidation reduces the number of individual communication transactions between nodes, minimizing the cumulative time loss from multiple inter-node communications while preserving the distributed storage capacity.
3Adaptability or versatility
If multiple inter-node communications are used for data transfer, then flexibility in data management is improved, but the performance of I/O processing is degraded
Solution Approach 1:
The data transfer unit acts as an intermediary that intelligently manages data transfer operations, maintaining data management flexibility through configurable placement and operation modes while optimizing throughput by reducing unnecessary communication steps. This intermediary coordinates transfers to minimize performance degradation.
Solution Approach 2:
The system performs preliminary determination of the optimal placement for the data transfer unit based on anticipated data access patterns and workload characteristics. This preliminary configuration optimizes I/O processing throughput in advance, allowing the system to maintain flexibility in data management strategies while avoiding performance degradation from suboptimal communication patterns.
Data Source
AI summary
A computer system includes a host unit that issues a request of an I/O processing to a volume VOL, a local pool control unit that is in charge of management of a local pool based on a storage area of a drive of one node, and a global pool control unit that is in charge of management of a global pool based on a plurality of local pools, wherein the global pool control unit controls transmission of target data of the I/O processing performed by the host unit based on a commonality relationship among a first node that is formed with the host unit performing the I/O processing, a second node that is formed with the global pool control unit, and a third node that is formed with the local pool control unit managing the local pool.