Client-Enabled Routing for Data Center Caching Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional cloud-computing systems face challenges in maintaining up-to-date routing connections between clients and data center nodes due to infrastructure limitations and internal topology changes, leading to increased latency and diminished throughput in caching services.
Innovation Solution
A hybrid routing service that intelligently selects and maintains routing connections based on usage patterns, allowing clients to directly link to primary nodes through a load balancer, reducing the number of hops and enabling efficient data access across the data center by using a client-side routing table and connection table.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If clients create routing tables locally to access data center nodes, then routing flexibility is improved, but the routing connections become quickly out-of-date due to internal topology changes
Solution Approach 1:
The system implements a feedback mechanism where the data center actively pushes routing table updates to clients whenever internal topology changes occur. This ensures clients receive real-time information about node availability and routing paths, preventing their local routing tables from becoming out-of-date while maintaining routing flexibility.
Solution Approach 2:
The data center proactively maintains and updates routing tables before clients need to access data. By pre-establishing and pushing updated routing information to clients in advance, the system ensures routing connections remain valid without requiring clients to continuously manage their own routing tables.
2Adaptability or versatility
If the number of routing connections increases to support more nodes, then data access capability is improved, but infrastructure limitations prevent creating and maintaining all connections
Solution Approach 1:
Instead of requiring clients to establish connections to all data center nodes, the system uses a partial action approach where clients maintain routing connections to a manageable subset of nodes. The data center's routing table provides mappings that allow clients to access any node through these limited connections, reducing complexity while maintaining access capability.
Solution Approach 2:
The data center acts as an intermediary that manages the complexity of routing connections. It maintains the comprehensive routing table and uses this information to facilitate client access to data nodes, allowing clients to operate with simplified connection management while still accessing the full data center topology.
3Productivity
If routing connections are maintained to all nodes, then direct data access is improved, but infrastructure limitations and topology changes make this unsustainable
Solution Approach 1:
The system dynamically adapts routing connections based on changing conditions. When internal topology changes occur, the data center pushes updated routing information to clients, allowing the routing connections to dynamically reflect current network state. This maintains data access efficiency while adapting to topology changes without requiring static connections to all nodes.
Data Source
AI summary
Methods, systems, and computer-storage media for intelligently creating and using routing connections between on-premises servers within a client and nodes of a data center are provided. When routing connections being requested by the client are not presently established, a caching service within the data center is programmed to direct a client's cache request over available routing connections, thereby scaling out storage capacity across the data center. When routing connections being requested by the client are established, a load balancer within the data center is programmed to direct the cache request over a requested routing connection, thereby bypassing intervening hops within the data center, such as gateway nodes that host instances of the caching service. To bypass intervening hops, the client network is provided with access to a client-side routing table for identifying an appropriate network key and a connection table for identifying a routing connection mapped to the network key.