assessing the risks of it infrastructure – a personal network perspective
DESCRIPTION
Assessing the Risks of IT Infrastructure – A Personal Network Perspective. Daniel J. Hinz Jochen Malinowski. Summary. Information technology (IT) management for end-user computing systems like desktop computers is of high importance, especially the availability of these systems - PowerPoint PPT PresentationTRANSCRIPT
Assessing the Risks of IT Infrastructure – A Personal Network Perspective
Daniel J. HinzJochen Malinowski
HICSS 39 20. April 2023
Summary
Information technology (IT) management for end-user computing systems like desktop computers is of high importance, especially the availability of these systems
Users are a vital part of these systems, and any user incident renders the system unavailable until the incident is resolved
Classical approaches consider only measurable user incidents, eg. number of helpdesk calls
Integrating personal network characteristics may give a more realistic view of user incident behaviour
HICSS 39 20. April 2023
IT infrastructure management is a hot topic for more than 15 years now
1990 "IT infrastructure" appears for the first time in the top 10 issues identified by the survey
1995 "Building a responsive infrastructure" was ranked number one challenge
2000
2001
"IT infrastructure management" ranked third in these two informal surveys
2003 "Infrastructure developments" ranked second in the category of top application and technology developments
Survey to identify the most critical issues in IS management
Conducted regularly by the Society for Information Management (SIM), supported by the MIS Research Center (MISRC)
Among their members consisting of top executives as well as IS researchers
SIM survey
Source: [Niederman, Brancheau, Wetherbe 1991], Brancheau, Janz, Wetherbe 1996], [Luftmann, McLean 2004]
HICSS 39 20. April 2023
End-user computers are a central component of IT infrastructure
End user computers are a central component, as they are influenced by all other infrastructure components
Availability is a major aspect [Niessink and Van Vliet, 1998]
Potential causal model for desktop availability
Findings from the literature
Users may have major effect on desktop availabilityIf the user is lost, the system is inoperational, although all hardware and software is working
HICSS 39 20. April 2023
And users are a vital part
HICSS 39 20. April 2023
What would you do, if ….
… You want to order a Canon MP730 printer in the online purchasing platform of your company, but cannot find it in the printers section?
Keep on searching? (perhaps it's listed as "input device")
Call the helpdesk?
Ask the colleague next door, who just got his new Canon last week?
Order a different model?
Give up?
?
HICSS 39 20. April 2023
Classical approaches focus on "official" incidents, but they are only one part
SituationClassical approaches measure mostly official incidents (e.g., helpdesk calls) [Niessink and Van Vliet, 2000]
Complication Not all incidents are reported
but instead solved by asking co-workers or known experts
How can those aspects be considered to get a more realistic number of incidents?
HICSS 39 20. April 2023
Social network characteristics may have an influence on problem solving behavior
Social network characteristics may influence problem solving behavior…
… and Social Network Analysis (SNA) may offer a promising solution to address them[Wassermann and Faust 1994; Scott 2000]
The problem solving capability of the social network of co-workers influences the number of ‘unofficial’ user incidents
The problem solving capability of the helpdesk influences the number of ‘official’ user incidents
Focus on the relations between individuals
Body of qualitative measures of network structures
Widely applied in sociology and organizational studies
HICSS 39 20. April 2023
The density of the social network has an influence on problem solving
Based on expert interviews, two SNA measures to influence problem solving were chosen
Socio-Centric Density (SCD)of the network of co-workers [Barnes 1974]
)1(
nn
lSCD
n
lECD
Ego-Centric Density (ECD)of the helpdesk [Scott 2000]
HICSS 39 20. April 2023
SCD and ECD can be used to get a more realistic number of user incidents
ECD
SCDCHDUI 1
U3
U1
U2 HD0.9
0.8
0.8
0.6
1.0
0.8
Calculation of user incidents:
SCD = 0.68
ECD = 0.80
UI = 185
UI: User incidents
CHD: Calls that reach the helpdesk
α: Scaling factor
The ratio of both densities can be used to predict the unkown number of total user incidents (UI) from the number of known incidents (CHD)
Example
HICSS 39 20. April 2023
The initial model has to be extended to reflect these findings
Helpdesk quality and social network density now determine the user action upon an incident
HICSS 39 20. April 2023
Further research
Validation of the model with real-life data
Evaluation of the applicability of other SNA measures
Consideration of multiple hops
Consideration of dynamic aspects