P2P Database Node Selection for SSD HDD Performance Cost Trade-off
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In Peer to Peer (P2P) distributed database systems, mixing nodes with SSDs and HDDs can lead to deteriorated responsiveness when an HDD is selected to create the first replica, as it does not leverage the high performance of SSDs effectively, making it difficult to achieve high availability and low cost simultaneously.
Innovation Solution
A distributed database system is configured with management information storage units that store response count information, a storage destination determining unit that selects nodes with high input/output performance for write requests, and a storage control unit that ensures data is stored on these nodes, optimizing the use of SSDs and HDDs based on redundancy and match levels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If nodes with HDDs are selected to create the first replica in P2P distributed database, then cost is reduced, but responsiveness is deteriorated
Solution Approach 1:
The patent applies local quality by assigning different storage device types (SSD vs HDD) to different nodes based on their specific roles and requirements. High-performance SSD nodes are designated as leader candidates or read leaders, while HDD nodes serve as follower or read follower nodes. This allows each node to have optimized storage characteristics suited to its functional requirements, resolving the contradiction between cost reduction and responsiveness.
Solution Approach 2:
The system dynamically selects and assigns node roles (leader, read leader, follower, read follower) based on storage device performance characteristics. The cluster manager automatically identifies SSD-equipped nodes and assigns them performance-critical roles, while HDD nodes are assigned cost-optimized roles. This dynamic assignment ensures the system adapts to hardware variations without manual configuration, maintaining both cost-effectiveness and responsiveness.
2Speed
If all nodes use SSDs in distributed database, then responsiveness is improved, but cost increases
Solution Approach 1:
Instead of uniformly equipping all nodes with expensive SSDs, the patent applies local quality by assigning SSDs only to nodes where high performance is critical (leaders and read leaders), while using cheaper HDDs for follower nodes that primarily perform data storage and replication. This selective deployment maintains system responsiveness while significantly reducing overall cost.
Solution Approach 2:
The patent implements partial action by providing high-performance SSD storage only to the extent necessary for maintaining system responsiveness - specifically to leader nodes that handle write operations and coordination. Follower nodes use HDDs since they primarily perform data replication and storage, where performance is less critical. This partial optimization achieves acceptable system performance at reduced cost.
3Quantity of substance
If P2P distributed database uses mixed SSD and HDD nodes, then cost is reduced, but responsiveness deteriorates when HDD creates first replica
Solution Approach 1:
The system dynamically determines node roles based on storage device characteristics rather than using static assignments. The cluster manager identifies SSD-equipped nodes and automatically assigns them as leader or read leader candidates, ensuring that high-performance nodes handle time-critical operations. HDD nodes are dynamically assigned as follower or read follower roles. This dynamic role assignment prevents the responsiveness deterioration that would occur if HDD nodes were selected to create the first replica.
Solution Approach 2:
The system incorporates feedback mechanisms where the cluster manager continuously monitors node performance characteristics and storage device types. Based on this feedback, the system adjusts node role assignments to ensure SSD nodes perform leader functions that require high responsiveness. This feedback-driven optimization resolves the contradiction by preventing HDD nodes from creating the first replica, thereby maintaining responsiveness while still utilizing cost-effective HDD nodes in the cluster.
Data Source
AI summary
The present invention ensures improving responsiveness in a P2P type distributed database system that includes a plurality of node computers capable of storing management target data in storage devices. The storage devices include a first type storage device having a high input/output performance and a second type storage device having a low input/output performance. There is provided the storage devices that store KeySpace management tables including response count information indicative of a count of the node computers from which the identical management target data are to be read. The identical management target data are required to respond to a read request of the management target data. The node computers by a count indicated by the response count information and capable of storing the management target data in the first type storage devices among the plurality of node computers are determined as storage destinations when a CPU receives a write request of new management target data. The determined node computers are caused to store the management target data.