Page 308 - Cloud computing: From paradigm to operation

1 Framework and requirements for cloud computing

Table I.2 – Inter-cloud use case: Performance guarantee against an abrupt increase in load

Use case
Derived requirements for the The capability is required to support:
cloud capability – Policy negotiation including SLA management among the multiple CSPs
within a pre-established group (i.e., federation);
NOTE – Policy refers to a way for a CSP to provide services in terms of
presumed reliability of a machine, including its backup scheme and target
service levels. The policy may be different with each CSP. To maintain the same
quality of service to the CSC even when the CSP changes, the difference should
be negotiated beforehand and settled. This process is referred to as policy
negotiation. The same note applies to the other use cases.
– Self-performance monitoring at a CSP. If the performance degrades, the CSP
should initiate the next configured actions;
– Discovery, reservation, use and release of cloud resources in a dynamic
manner (i.e., not relying on the pre-configuration) on other CSPs within a
federation;
– Application invocation over the reserved resources on other CSPs within a
federation;
– Alteration and reversion (i.e., switchover and switchback) of CSC access
from one CSP to another CSP in a dynamic manner (i.e., not relying on the
pre-configuration) within a federation;
– Exchange of monitoring and auditing information among the multiple CSPs
within a federation;
– Exchange of authentication information about CSC (user/enterprise)
authentication status among the multiple CSPs within a federation.

I.3 Performance guarantee regarding delay (optimization for user location)
Table I.3 shows inter-cloud use case for performance guarantee regarding delay.

Table I.3 – Inter-cloud use case: Performance guarantee regarding delay

Use case

Use case title Inter-cloud use case: Performance guarantee regarding delay
Relevant roles CSP and CSC

Use case description – CSPs guarantee their service performance (in particular, network delay
and response time), even when a CSC moves to a remote location (e.g.,
on a business trip), by using cloud resources provided by another CSP
located close to the CSC on a temporary basis.
– When degradation in the response time is detected for a CSC at a CSP,
available resources are autonomously discovered and reserved in
another CSP that is near the CSC based on the user's location
information.

300