Method and system for scheduling middle boxes in software defined network
A software-defined network and scheduling method technology, applied in the field of middlebox scheduling, can solve the problems of increasing network packet delay, not being able to cope with dynamically changing flows, increasing flow delay, etc., and achieving good delay performance
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0119] This embodiment proposes a Middlebox dynamic deployment method to solve the above problems, which includes the following steps:
[0120] 1) Obtain a series of parameters: the distribution TList of the given flow, the network G, and the list MBNumList of the number of each Middlebox required.
[0121] Stream distribution TList is a list of streams, and each stream f can be expressed as a list, such as (src,dst,(MB a ,MB b ,MB c )) indicates that the flow f starts from the source point src to the destination point dst, and must be sequentially processed by Middleboxes of types a, b, and c. This processing sequence (service chain) can be represented by f.chain. For network G, G.plist represents a list of resource pools, and each resource pool can run N Middleboxes.
[0122] 2) Maintain a score table corresponding to all Middlebox types of all resource pools, and determine the placement position of Middlebox by simulating K rounds of voting. i resource pool p i The x-t...
Embodiment 2
[0139] A Middlebox flow scheduling method
[0140] In Embodiment 1, the Middlebox dynamic placement problem is described and related solutions are given. On this basis, flow scheduling is performed, and each flow is assigned to a specific Middlebox according to the processing sequence. Given the specific traffic distribution TList, the placement scheme pl, and the maximum tolerated delay r, the flow scheduling problem is calculated through the modified Viterbi algorithm (Masked-Viterbi). For each flow f, we call the optimal scheduling path the minimum delay path (Minimum Delay Path, MDP), and the scheduling problem is to find the minimum delay path for each flow in the TList.
[0141] Stream Scheduling Algorithm
[0142]The flow scheduling algorithm is divided into two stages. In the first stage, the minimum possible delay for each stream is calculated regardless of the Middlebox load. At this stage, we consider the processing delay of each Middlebox to be the basic delay, ...
Embodiment 3
[0173] In this embodiment, the main part is implemented in the controller in the software-defined network, and the Middlebox software needs to be modified to allow the Middlebox software to accept the control of the controller. The realization of the whole scheme is composed of the algorithm module in the controller, the Middlebox control module and the controller adaptation module in the Middlebox, such as image 3 shown. The algorithm is implemented in the algorithm module, the controller controls the Middlebox through the Middlebox control module, and the interaction between the Middlebox and the controller is realized through the controller adaptation module. This process is as follows: Figure 10 shown. The client sends a packet to the switch, and the switch queries the flow table. If the switch cannot find the corresponding flow table, it sends the packet to the controller. The controller runs an algorithm, calculates the flow table corresponding to the packet, and send...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com