| With the booming technological revolution and industrial change,a new generation of information technology is being applied and permeated in all walks of life,and the Industrial Internet has become a strategic choice for countries to improve their competitiveness in the industry.The industrial internet identification and resolution system is an important infrastructure for the industrial internet to achieve interconnection and interoperability of equipment,coordination and dispatch of resources,and coordination and cooperation in production.However,with the rapid expansion of the number of users and the continuous access of a large number of devices,the existing identification resolution system faces problems such as high concurrency and overload,resulting in the system being prone to performance degradation,single point of failure and long query time,etc.Therefore,it is necessary to design an identification resolution system solution with load scheduling capability to realize the triage of resolution requests through load balancing and improve the hit rate of requests through caching algorithm,and the two can work together to effectively reduce the query delay of users,provide reliable services and improve user experience.Firstly,this thesis proposes a recursive node caching scheme with differentiated cache granularity to address the problem of high user query latency in the industrial internet identity resolution system.The paper first establishes a content popularity prediction model based on a multiple linear regression model to mine the hotness trend and request distribution of requests.Then a granularity caching scheme is designed based on the identity data characteristics,so that recursive nodes can decide the granularity of their stored information when caching content,thus effectively utilising the cache space.This paper then proposes a caching decision method based on an improved genetic algorithm to allocate cache space to make caching more efficient.The experimental results show that the proposed algorithm performs well in terms of user query latency and request hit rate,and can solve the problem of high user query latency in existing identity resolution systems to a certain extent.Secondly,this paper designs a dynamic load balancing algorithm for the recursive nodes of identity resolution to address the high concurrency problem in the industrial Internet identity resolution system,and proposes a load rebalancing algorithm based on interest matching to address the load migration needs arising from server downtime or system scale expansion that may occur during the operation of server clusters.By considering the remaining capacity of the recursive node server and the cache situation,this paper forwards the user’s parsing requests to the cache server with a higher interest match,which not only achieves load balancing of the requests,but also effectively utilizes the cache information of the nodes,enhancing the stability and scalability of the system.The problem is then formulated as a multi-objective optimization problem,and an improved artificial bee colony algorithm is proposed for solving it,which avoids prematurely falling into local optimal solutions while retaining the original local search capability,and makes the algorithm more capable of global search.The experimental results show that the proposed algorithm can effectively reduce the processing delay and improve the overall performance of the server cluster while ensuring the load balance,which has certain feasibility and reference significance. |
