Page 45 - Kaleidoscope Academic Conference Proceedings 2024
P. 45

RESEARCH ON SCENARIO-BASED DYNAMIC INSPECTION METHODOLOGY USING
                                                EXPRESSION ENGINE




                                                                                     2*
                                          1
                                                                          1
                                                   1
                                   Jian, Wu ; Jun, Yao ; Guohui, Li ; Jianhong, Gu ; Lijun, Quan
                                                             1
                                      1 China Mobile Communications Group Co., Ltd, China
                              2 School of Computer Science and Technology, Soochow University, China
                                           *Corresponding author: ljquan@suda.edu.cn


                              ABSTRACT                        There are three main traditional inspection methods: manual
                                                              verification  inspection[3],  inspection  combined  with
           With the increasing scale of centralized management, data   blockchain  technology[4][5],  and  inspection  through  big
           quality  issues  are  becoming  more  prominent,  so  an   data analytics[6][7]. Manual verification, while constrained
           automated method is needed to check the logical compliance   by a standardized process, still carries hidden risks in terms
           of configuration item content and field-to-field. In this paper,   of  efficiency  and  accuracy.  The  application  scenarios  for
           we  use  an  expression  matching  engine  to  implement  an   inspection  combined  with  blockchain  technology  are
           automated inspection method, organizing business scenarios   relatively  limited,  as  the  premise  of  data  "immutability"
           into  inspection  rules  and  using  the  expression  matching   cannot be universally applied to all fields. Inspection through
           engine  in  conjunction  with  abstracted  logic  extensions  to   big  data  analysis  is  more  challenging  to  implement  in
           implement  universal  inspection  formulas.  The  inspection   scenarios with moderate and lightweight data requirements.
           formulas  are  applied  to  various  inspection  scenarios  and   Consequently,  in  the  actual  application  process,  it  is
           results  are  visualized  to  assist  with  decision-making.  The   necessary to choose the most suitable inspection technology
           proposed method has been applied to the CMDB system, and   based  on  specific  requirements  and  combine  it  with
           experimental analysis shows that this inspection mechanism   customized inspection solutions for compliance checks.
           meets  the  performance  and  accuracy  requirements  of
           production  environments.  Additionally,  it  significantly   Specifically,  in  the  realm  of  automated  operations  and
           reduces  code  modifications  and  cutover  costs,  ultimately   maintenance  (O&M),  the  Configuration  Management
           achieving  real-time  compliance  with  customer  auditing   Database  (CMDB)  holds  a  crucial  position[8].  Given  that
           needs.                                             CMDB contains all configuration information in the system,
                                                              inspecting CMDB data is an indispensable step in automated
              Keywords – Dynamic inspection, Expression engine,   O&M. Through inspection of CMDB data, system stability
               Configuration management database, Data quality.   and security can be ensured, preventing issues arising from
                                                              data errors. For instance, in the Cloud Start platform of China
                         1.  INTRODUCTION                     Mobile's IT cloud, the relationships between configuration
                                                              items  are  relatively  complex,  involving  physical
           In  the  era  of  big  data,  the  interconnection  among  cloud   relationships,  real-time  communication  relationships,  non-
           computing, the internet, and big data is closely linked, giving   real-time  communication  relationships,  and  dependency
           rise  to  complex  operational  environments[1].  Automated   relationships. As of now, CMDB has managed 80 resource
           operations  and  maintenance  (O&M)  refer  to  the  use  of   pools,  836  clusters,  over  10,000  application  systems,  and
           automated tools and technologies to manage and maintain IT   more than 550,000 configuration items. With the continuous
           systems, thereby enhancing the efficiency and stability of IT   expansion  of  the  managed  scale,  CMDB  gradually  faces
           systems[2]. Simultaneously, automated O&M monitors and   operational challenges such as abundant data, inconsistent
           manages  the  resources  of  IT  systems  through  automated   standards,  and  rapid  changes,  leading  to  numerous  data
           tools, ensuring optimal utilization of resources. However, in   quality issues[9]. Addressing the CMDB data issues in the
           the  process  of  automated  O&M,  various  data  issues  are   Cloud  Start  platform,  the  specific  implementation  and
           frequently  encountered,  necessitating  continuous  data   maintenance  process  face  three  challenges:  (1)Diverse
           monitoring and governance. Consequently, the demand for   CMDB  Models:  CMDB  comprises  numerous  models,
           inspection  technology  is  increasing  day  by  day,  and  the   ranging from the physical perspective of routers, switches,
           mature  inspection  models  are  bound  to  be  a  perfect   firewalls,  host  machines,  storage  servers,  management
           integration of the new generation of information technology.   servers, to the business perspective of application systems,
                                                              business  systems,  tenants,  and  more.  Currently,  there  are
                                                              over 40 models in CMDB, and these models are subject to





            978-92-61-39091-4/CFP2268P @ITU 2024           – 1 –                                    Kaleidoscope
   40   41   42   43   44   45   46   47   48   49   50