Information Technology Reference
In-Depth Information
this interface, the program can call the services
provided to upload and download digital resources.
The other way is to use it as a Mashup service. The
application can embed it in the web page directly
and manage the digital content.
In this cloud, we use UUID as the identifier of
a file. Because of the uniqueness, the application
program can generate the UUID itself without
creating a duplication problem. UUID makes the
application of the Mashup service to upload file
asynchronously. That is, the application can save
the digital resource immediately without waiting
for the resource ID returned from WFMS. There-
fore, the performance of the application can be
improved. The uniqueness of UUID also enables
the distributed deployment of WFMS servers,
which will be implemented at the next stage.
Although the digital contents are in the format
of files, for the security reason, these files cannot
be accessed directly. Rather, the files must be
downloaded for which there are two methods..
One is to include the whole file in the response
of the request. The other is to copy the file into
a temporary directory with a random name and
provide a link. In conjunction, there is a routine
to clean the temporary files periodically. Overall,
the first method is appropriate for the seldomy
downloaded files that consume a lot of memory.
The second method is appropriate for parallel
download processes, which can decrease the
workload a lot.
logs locally, and send them to ULAS when the
system is in low workload, the performance of the
system can be improved. By keeping the logs on
the remote server, concern for the security of the
data might be relieved. In addition, keeping the
logs on several ULAS clouds is also supported.
Corresponding to the Log Component in sec-
tion 3.1.4, there are two kinds of logs supported
in the ULAS. ULAS can analyze the two kinds
of logs and present the results in a chart. Addi-
tionally, ULAS supply the service of adding logs
through ActiveMQ. It also provides SaaS service.
Therefore, a user can also upload log files to the
system, and then ULAS can help analyze the log
files and give visualized charts.
5.5. TAGSYS
As the common information retrieval technol-
ogy cannot meet the requirement of queries with
different backgrounds and intents, personalized
services appear to provide different services to
different users. Personalization can improve the
service quality and access efficiency, thus to pull
more visitors. More so, a personalized annotation
system is used to collect annotations of a user,
predicate users' preferences and recommend
resources according to users' current preferences
and taste (Sen et al. 2009).
TAGSYS is a personalized resource annota-
tion, management and recommendation system
based on tags, score, comments and usage status.
It provides the support to the process of resource
utilization. In the traditional digital resource man-
agement systems, the value of a digital resource
will not change. With the help of TAGSYS, the
users' feedback can enrich the semantics of the
digital resources and help with other users' learn-
ing processes. The dynamic generation of tags
can help to generate the up-to-date taxonomy. By
analyzing the annotations of the users, TAGSYS
can recommend related digital resource to users.
The services provided by TAGSYS include the
management of annotations and recommendation
5.4. ULAS
The logs are very important to track the users'
behaviors and system changes. There are several
disadvantages to keep the logs locally. Writing the
logs to local disk requires file IOs, which will de-
crease the performance of the system significantly.
Secondly, if the local disk is crashed, the logs may
be corrupted at the same time. Therefore, it will
bring difficulties to re-construct the system data.
We use ActiveMQ to transfer the logs to the
ULAS cloud. Because ActiveMQ can keep the
Search WWH ::
Custom Search