Information Technology Reference
In-Depth Information
as we have collaborated with industry partners and domain experts from
the e-science domain while identifying them. Zinn et al. migrated an exist-
ing application based on scientific workflows from the domain of astronomy
to Microsoft Windows Azure (Zinn et al., 2010). The existing application
we migrate to the cloud for the purpose of evaluating our approach is also
based on scientific workflows. Deelman et al. evaluated the cost of running
e-science applications in the cloud, focusing on the trade-off between dif-
ferent workflow execution modes and provisioning plans, and came to the
conclusion that the costs highly depend on the selected deployment strategy
(Deelman et al., 2008). We do not explicitly consider costs but provide recom-
mendations and guidelines with respect to the deployment strategy.
Amazon proposes a phase-driven approach consisting of six phases for
migration of an application to its cloud infrastructure (Varia, 2010). The
data migration phase is subdivided into a selection of the concrete Amazon
AWS service and the actual migration of the data. Amazon provided recom-
mendations regarding which of their data and storage services best fit for
storing a specific type of data; for example Amazon Simple Storage Service
(Amazon S3, http://aws.amazon.com/s3/) is ideal for storing large write-once,
read-many types of objects. As the methodology proposed by Amazon
focuses on Amazon AWS data and storage services only, we abstracted from
this methodology and integrated the guidelines in our proposal. In addi-
tion to several product-specific guidelines and recommendations (Microsoft,
2013a, 2013b), Microsoft provided a Windows Azure SQL Database Migration
Wizard (http://sqlazuremw.codeplex.com) and the synchronization service
Windows Azure SQL Data Sync (http://www.windowsazure.com/en-us/
manage/services/sql-databases/getting-started-w-sql-data-sync/). We reused
some of these tools, tutorials, and wizards and refer to them during the data
migration phase.
For the App Engine, Google is offering the tool Bulk Loader (http://bulk
loadersample.appspot.com), which supports both the import of CSV and
XML files into the App Engine Data Store and the export as CSV, XML, or
text files. The potentially required transformations of the data during the
import are customizable in configuration files. In addition, Google supports
the user when choosing the appropriate data store or service and during its
configuration (Google, 2013b). Moreover, they provide guidelines to migrate
the whole application to Google App Engine (Google, 2013a). We refer to
the tools during the migration phase and abstract from the vendor-specific
guidelines and recommendations to integrate them in our tool.
Salesforce provides data import support to their infrastructure via a
web user interface or the desktop application Apex Data Loader (http://
sforce-app-dl.sourceforge.net). Another option to migrate and integrate
with cloud providers such as Salesforce is to hire external companies that
specialize in migration and integration, such as Informatica Cloud (http://
www. informaticacloud.com). In addition to the tools or external support,
Salesforce provides data migration guidelines (salesforce.com, 2013). We
Search WWH ::
Custom Search