Page 255 - DCAP405_SOFTWARE_ENGINEERING

Software Engineering

Notes
integrity, and authentication requirements, along with location of data, availability, and
recoverability.
Moreover, the industry has unique and dynamic regulatory, legal and compliance issues
to address before switching to cloud services. Bankers apprehend that computing in “the
cloud” is risky, as it involves outsourcing the data of its customers to third-party cloud
service operators.
In order to take advantage of the emerging powerful Internet-based business solutions,
what is needed is an IT technology architecture that combines the merits of the public
cloud with the security and data-privacy of the private cloud.
The appropriate business strategy seems to be to outsource relatively less sensitive data to
the public cloud infrastructure service with cryptography and simple password access,
along with dedicated servers with firewalls and intrusion detection devices and other
updated safety features for housing ultra-secure critical data centres that demand strong
authentication for access.
Strategic Policy

There should be a clear strategic policy for cloud computing and management, prioritising
data that can be entrusted to the cloud operator, with clearly defined Service Level
Agreements (SLA) with milestones and a set time-frame, backed by a comprehensive
governance structure.

While choosing the cloud service provider, it is important to look into the firm’s financial
stability, ability to improve functionality and service levels and integrate data across
different technology platforms and cloud services.

The policy-based key management, with industry-standard encryption, is emerging as
the cryptography model for better control on data in the cloud as the common encryption
key management techniques are susceptible to vulnerability.
Cloud security services are emerging to address data security, privacy and compliance
risks, as well as prevention of data theft, ensuring disaster management, and detecting
compliance violations with robust server security for virtualised data centres.

14.6 Summary

To design any system, we require a mathematical model. Similarly, for DB Systems there are
several models being used as given below:
Relational Model

Network Model
Hierarchical Model
Out of all the above models, the Relational Model is used by most of the database management
packages, including Oracle, Ingres, Sybase, etc.

14.7 Keywords

CODASYL/DBTG: (Conference on Data Systems Languages/Database Task Group) A standard
specifying how to store and retrieve data to and from a network database.
Data Model: An abstract representation that captures the essence of relation with the data items.

248 LOVELY PROFESSIONAL UNIVERSITY