Control Plane ARP Queueing for Lower NIC CPU Overhead
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network interface card (NIC) communications in computing environments face high CPU cycle costs and inefficiencies due to the need for address resolution protocol (ARP) processing, which can be exacerbated by excessive ARP and Neighbor Discovery traffic, affecting data plane performance.
Innovation Solution
Implementing an enhanced queued direct input/output (EQDIO) facility with control plane queues and an adapter module to offload and filter ARP processing, allowing control programs to maintain their own ARP caches and reducing data plane traffic by handling ARP flows at the control plane.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If ARP processing is performed at the data plane level using traditional NIC communications, then address resolution can be achieved, but CPU cycle costs increase and system performance decreases
Solution Approach 1:
The patent extracts ARP processing functionality from the data plane and relocates it to a dedicated control plane. The control plane includes an ARP processing module that handles address resolution requests separately from data transmission operations, thereby reducing CPU cycle consumption in the data plane while maintaining ARP resolution capability.
Solution Approach 2:
The patent segments the network communication plane into two distinct planes: a control plane for handling ARP processing and a data plane for handling actual data transmission. This segmentation allows independent optimization of each plane, with the control plane dedicated to management tasks and the data plane focused on high-speed data transfer, improving overall ARP processing efficiency.
2Ease of operation
If ARP and Neighbor Discovery traffic is allowed to flow freely in the data plane, then address resolution requests can be processed, but data plane performance is degraded due to excessive traffic
Solution Approach 1:
The patent introduces a control plane as an intermediary layer between the data plane and the network interface card. The control plane intercepts and processes ARP and Neighbor Discovery traffic before it reaches the data plane, acting as a mediator that filters out management traffic from the data transmission path, thereby preserving data plane throughput while maintaining address resolution capability.
3Reliability
If continuous ARP querying is performed to maintain ARP cache updates, then address resolution accuracy is maintained, but system latency increases and processing efficiency decreases
Solution Approach 1:
The patent implements periodic ARP cache updates through the control plane, which refreshes ARP cache entries at scheduled intervals rather than continuously querying. This periodic action maintains ARP cache accuracy while significantly reducing the time loss associated with continuous querying, as the control plane can perform updates without interfering with data plane operations.
Data Source
AI summary
A control program executing on a computing device of a computing environment transmits control request information to an adapter module executing within the computing environment. The transmitting uses a control plane transmit queue of a set of queues. The control plane transmit queue identifies a location to store the control request information to be retrieved by the adapter module. The control program receives from the adapter module an indication of another location storing control reply information replying to the control request information. The indication is obtained from a control plane receive queue of the set of queues. The control program retrieves from the another location the control reply information and performs processing based on the control reply information.