The Oracle Cloud Platform a PaaS service provides a shared and elastically

The oracle cloud platform a paas service provides a

This preview shows page 21 - 24 out of 96 pages.

The Oracle Cloud Platform, a PaaS service, provides a shared and elastically scalableplatform for consolidation of existing applications and new application development and de-ployment. It supports standards-based shared services and elastic scalability on demand andincludes database functionality based on Oracle Database and Oracle Exadata Database Ma-chine.It features middleware technology based on Oracle Fusion Middleware and OracleExalogic Elastic Cloud.Problem 3.Read the IBM report [41] and discuss the workload preferences for private andpublic clouds and the reasons for the preferences.The IBM study [41] asked customers to rate the 25 different workloads listed in Figure 4they have already deployed or were planing to deploy on public and private clouds. Figure5 summarizes the leading attractions to a public cloud. Figures 6 and 7 from [41] show thepreferred private and public cloud workloads, respectively.Problem 4.In Section 3.7 we introduced the concept of energy-proportional systems andwe saw that different system components have different dynamic ranges.Sketch a strategyto reduce the power consumption in a lightly-loaded cloud and discuss the steps for placing acomputational server in a standby mode and then for bringing it back up to an active mode.21
Background image
Figure 4: Workloads discussed in the IBM survey.We have seen in Section 3.7 that even at a very low load a server uses close to 50% of theenergy consumption when delivering peak performance. The basic philosophy is to maintainall servers in an optimal region, a region where the relative performance is high while therelative power consumption is low. To achieve this goal, the applications running on a serverworking outside its optimal region should be migrated to other servers and the server shouldbe switched to a sleep mode when its power consumption is negligible.Problem 5.Read the paper which introduced the concept of dataspaces [27] and analyze thebenefits and the problems with this new idea. Research the literature for potential applicationof dataspaces to scientific data management in a domain of your choice, be it the search forthe Higgs boson at CERN, structural biology, cancer research, or another important researchtopic which involves data-intensive applications.According to the paper “A New Abstraction for Information Management”semantic inte-22
Background image
Figure 5: Public cloud drivers.grationis a measure of how closely the schemas of the various data sources have been matched.In other words, how well the types, names, units, meanings, etc. of the data in the sourcesare matched up. At the high end of the spectrum, all data conforms to a single agreed-uponschema. At the low end, there is no schema information at all. In between lay various dataintegration solutions and approaches based on semi-structured data and controlled vocabu-laries. This dimension indicates the degree to which semantically rich query processing anddata manipulation can be provided across a group of data sources, with higher degrees of
Background image
Image of page 24

You've reached the end of your free preview.

Want to read all 96 pages?

  • Summer '17
  • ALBERT DOMINIC
  • The Land, cloud service, Cloud delivery models, Cloud Computing delivery

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture