Migrating e-Science Applications to the Cloud: Methodology and Evaluation - Cloud Computing with e-Science Applications - page 109

Assessment
Analysis & Design
Test
Migration,
deployment &
support

10%

10%

30%

50%

FIGURE 5.6
Amount of time spent per migration phase.

In the fifth step, the analysis of the gathered and processed data took place.
Altogether, we recorded seven problems. Five of the recorded problems had
a high priority; the remaining two had middle priority. Two of the occurred
problems were due to bugs in the graphical user interface of the tool, one
with middle and one with high priority. Two problems were caused by miss-
ing features, also one with middle and one with high priority. The rest of the
problems, all with high priority, were caused by lack of appropriate informa-
tion available to the user, as in the example of Table 5.2. The analysis of the
identified problems with respect to their priority and the cause of the prob-
lems showed that the main weakness of the Cloud Data Migration Tool was a
lack of information provided to the user. Further improvements toward this
direction are therefore required in the future.
The analysis of the time spent per migration phase is summarized in
Figure 5.6. As shown in the figure, half of the time was actually spent in the
Test phase, which as explained in Section 5.4 is not directly supported by
our methodology (and therefore also not by the Cloud Data Migration Tool).
While this identifies a deficiency in our proposal, it can also be attributed at
least in part to the acceleration of the other phases by the use of the Cloud
Data Migration Tool. In any case, what can be identified is a clear need for the
incorporation of the remaining two phases (Test and Optimization) in our
methodology and as a result their support by the Cloud Data Migration Tool.
Finally, for the implementation of steps 6 and 7 of the ITIL CSI process
(presentation and use of the information and implement improvements,
respectively), we are currently in the process of incorporating the lessons
learned by this case study in further research work.

5.6 Conclusions
The popularity of cloud computing has led to significant research in
cloud-enabling applications, that is, migrating whole systems or only parts