Network Pipeline Abstraction Layer Emulation for Hardware-Accelerated Steering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current SFC architectures lack support for flexible steering rules, configurable and dynamic interface mappings, and network acceleration in a single accelerated data plane, which hinders scalability and efficiency in modern, cloud-centric networks.
Innovation Solution
Implementing a network pipeline abstraction layer (NPAL) on a DPU that supports hardware-accelerated flexible steering rules, configurable interfaces, and fast link recovery, enabling a unified interface for multiple network protocols and functions, and allowing user-defined logic for enhanced network management and acceleration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional middlebox hardware devices are used for network security and performance functions, then network security and performance are improved, but capital investment, space occupation, and operational complexity increase significantly
Solution Approach 1:
The patent combines multiple traditional middlebox functions (firewall, load balancer, IDS, DLP, VPN, WAN optimization) into a single integrated DPU device. This consolidation maintains comprehensive network security and performance capabilities while reducing the number of separate hardware devices, lowering operational complexity, and freeing up data center space.
Solution Approach 2:
The DPU is designed as a universal platform that can perform multiple network functions simultaneously - security functions (firewall, IDS), performance functions (load balancing, WAN optimization), and data management functions. This multi-functionality eliminates the need for separate specialized hardware for each function, reducing both capital investment and operational overhead.
2Adaptability or versatility
If more physical middlebox devices are acquired to scale network functions, then network functionality is enhanced, but infrastructure cost and integration complexity increase
Solution Approach 1:
By consolidating multiple network functions into a single DPU device, the system enables scaling of network functionality through software configuration rather than physical device integration. This approach maintains enhanced adaptability and versatility while eliminating the integration complexity associated with adding multiple separate physical middleboxes.
Solution Approach 2:
The DPU employs dynamic, software-defined network functions that can be configured, modified, and scaled without physical hardware changes. This dynamic approach allows network functionality to adapt to changing requirements through software updates and reconfiguration, rather than requiring integration of new physical devices.
3Ease of operation
If specialized personnel are deployed for middlebox operation and maintenance, then device management is improved, but operational costs and human resource requirements increase
Solution Approach 1:
The DPU incorporates automated self-management capabilities including self-provisioning, self-configuration, and self-diagnosis. The device can automatically detect its environment, configure appropriate network functions, and monitor its own health status, reducing the need for specialized personnel while maintaining ease of operation through intelligent automation.
Solution Approach 2:
The DPU implements comprehensive monitoring and feedback mechanisms that automatically track device performance, security events, and operational status. This feedback system enables automated response to events and provides visibility into device management without requiring constant human intervention, reducing personnel requirements while improving ease of operation.
4Productivity
If hardware-accelerated network pipeline is implemented, then network processing performance is improved, but hardware complexity and cost increase
Solution Approach 1:
The DPU acts as an intermediary device between the host CPU and the network, providing hardware-accelerated network processing functions. By positioning the DPU as a mediator that handles network acceleration tasks, the system achieves high network processing performance while keeping the host hardware relatively simple and reducing overall system complexity.
Solution Approach 2:
The DPU replaces software-based network processing with hardware-accelerated processing, substituting mechanical/software operations with optimized hardware circuits. This substitution delivers significant network processing performance improvements while consolidating hardware complexity into a single specialized device rather than requiring complex configurations across multiple devices.
Data Source
AI summary
Technologies for creating an optimized and accelerated network pipeline using an emulated network pipeline abstraction layer (NPAL) of an emulated data processing unit (DPU), including an emulated processing device and an emulated acceleration hardware engine, are described. The emulated NPAL supports multiple network protocols and network functions in an emulated network pipeline. The emulated network pipeline includes a set of tables and logic organized in a specific order to be accelerated by the emulated acceleration hardware engine. The emulated acceleration hardware engine can process network traffic data using the emulated network pipeline.