Cluster Node Session Context Synchronization to Eliminate Traffic Redirection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In scalable clusters, data traffic redirection is inefficient due to the need for significant computing and network resources, especially during resizing and failovers, as traffic is often sent to nodes without context information, leading to increased redirection and resource utilization.
Innovation Solution
A system and method where nodes in a cluster acquire and store session contexts, allowing them to process data packets without redirection by synchronizing contexts through a control plane, prioritizing based on latency requirements, and reducing the need for external router interactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traffic distribution is carried out by hashing mechanism based on source IP address, then traffic can be distributed among cluster nodes, but data packets may be sent to nodes without context information requiring redirection
Solution Approach 1:
The patent applies preliminary action by pre-distributing session contexts to cluster nodes before traffic arrives. The control plane proactively replicates context information from the active node to standby nodes based on predicted traffic patterns, so when data packets arrive, nodes already have the necessary context information ready, eliminating the need for redirection and context acquisition delays.
Solution Approach 2:
The control plane acts as an intermediary that manages context distribution among cluster nodes. It receives context information from the active node and intelligently distributes it to appropriate standby nodes, mediating between the active node and standby nodes to ensure proper context placement without requiring direct communication or redirection between data nodes.
2Loss of time
If cluster nodes advertise correct paths to upstream routers, then redirection can be eliminated, but significant computing and network resources are consumed during advertising and resizing
Solution Approach 1:
The patent extracts the context information from the active node and separates it from the data traffic flow. By taking out the session context and distributing it independently to standby nodes via the control plane, the system eliminates the need for upstream router advertising and redirection mechanisms, removing the harmful resource consumption associated with path advertising during resizing and failover events.
Solution Approach 2:
The system creates copies of session context information from the active node and distributes these copies to standby nodes. Instead of relying on upstream routers to advertise paths and redirect traffic, the context is copied and placed in advance on standby nodes, allowing them to immediately handle traffic without redirection while consuming minimal computing and network resources.
3Ease of operation
If session context is acquired from further node, then data packet can be processed without redirection, but acquisition time may exceed packet delivery time
Solution Approach 1:
The control plane performs preliminary action by proactively distributing session contexts to standby nodes before traffic is forwarded to them. This advance preparation ensures that when data packets arrive at standby nodes, the context information is already available in local memory, eliminating the need for real-time context acquisition and ensuring processing speed exceeds packet delivery time.
4Adaptability or versatility
If cluster resizing or failover occurs, then cluster adaptability is improved, but redirection amount significantly increases
Solution Approach 1:
During cluster resizing or failover events, the control plane performs preliminary action by proactively redistributing session contexts to appropriate nodes before traffic patterns change. This ensures that when failover occurs or nodes are added/removed, the new active and standby nodes already have the necessary context information, preventing traffic redirection and maintaining seamless adaptability without increasing redirection volume.
Solution Approach 2:
The system maintains continuity of useful action by ensuring session contexts remain continuously available on standby nodes through proactive distribution. During resizing and failover, this continuous availability of context information allows seamless transition of traffic without interruption or redirection, maintaining adaptability while preventing increases in redirection volume.
Data Source
AI summary
Provided are methods and systems for eliminating a redirection of data traffic in a cluster. An example method may include receiving, by one or more nodes of the cluster, a data packet associated with a service session. The method may include determining, by the node, that the data packet is directed to a further node in the cluster. The method may further include, in response to the determination, acquiring, by the node, a session context associated with the service session. Acquiring the session context may include sending, by the node, a request for the session context to the further node and receiving the session context from the further node. The method may further include processing, by the one or more nodes, the data packet based on the session context.