Software to Data model Lenos Vacanas Stelios Sotiriadis

Software to Data model Lenos Vacanas, Stelios Sotiriadis, Euripides Petrakis Technical University of Crete (TUC), Greece www. intelligence. tuc. gr Workshop on Adaptive Resource Management and Scheduling for Cloud Computing, ARMS-CC 2015

Agenda • • • Cloud computing The need… Inter-Cloud Software to data model Experimental Analysis Future steps

Cloud computing

Cloud computing • An infrastructure (including hardware and software) for delivering bespoke and on-demand services via the Internet – Service Elasticity – Including: Saa. S, Paa. S, Iaa. S, hybrid models • Vendors develop data-centers (DCs) to accommodate users requests for certain resources – – – Faster provisioning of systems and applications Ease to integrate Cost over consumption services Flexibility by the use of virtualisation Mobility access …

The need… • Customers need to transfer services among clouds. • Currently, clouds of the same provider do not support a coordinated distribution of resources. – Amazon example • DCs are spread at different geographically locations • Doesn’t support collaboration of clouds for e. g. outsourcing requests, improve redundancy etc. (intra-) • Requests for users to select their cloud location, thus low Qo. S level based on random user choices – HP, Intel, Yahoo aim to interoperable clouds • Vendor-oriented solution • Limited access to the testbed

Inter-cloud • Inter-cloud expands cloud datacentre capabilities by providing a flexible initiative for sharing resources • The standards for an Inter-cloud environment include that services from cloud providers could be geographically distributed, coordinated and dynamically scaled in order to offer a good quality of service • Inter-cloud is a dynamic distributed system topology. • The viewpoint of the study is that an Inter-coud environment is composed of different cloud virtual infrastructures

Cloud and Inter-Cloud An inter-cloud service expects to act as an intermediate in communication between difference clouds. 7

Open. Stack • Open. Stack Software delivers a massively scalable cloud operating system. • Open. Stack is a global collaboration of developers and cloud computing technologists producing the ubiquitous open source cloud computing platform for public and private clouds. • Founded by Rackspace Hosting and NASA, Open. Stack has grown to be a global software community of developers collaborating on a standard and massively scalable open source cloud operating system.

Open. Stack Community

Cloud provisioning models • Iaa. S (Infrastructure) – Virtualization, dynamic scaling, Internet connectivity, utility computing (e. g. a VM) • Saa. S (Software) – Apps hosted in vendors (e. g. a Google doc), automatic updates, compatibility • Paa. S (Platform) • • • – Outgrowth of Saa. S: Apps are available through the Internet (e. g a VM with ready installed Software Enablers) Whats the difference? – Iaa. S: underlying infrastructure | Paa. S: Development platform Private Cloud & Public Cloud – No difference in terms of architecture but in terms of accessibility. – Private: Single Organisation – Public: Rendered over the net Hybrid provisioning models? 10

Cloud in sensitive domains: Healthcare provisioning • Why cloud computing & what is the problem? – Standards, regulations, recommendation • legislation, ISO standards – Security standards (ISO 27000) • there are severe restrictions to data transfer, storage, aggregation and analysis – Cloud computing: • remote services (data management in distant datacenters) 11

Problem area: Healthcare provisioning Data Data Big. Data Data Data Data Data 12

Hybrid Cloud Deployment: Software to Data • Hybrid Cloud: – A reverse cloud approach that offer the required framework (Public Cloud) to allow Software Enablers to be initiated at clients’ sites and to be hosted in a Private Cloud • Software to data cloud model – Provider Edge – Consumer Edge Healthcare private cloud Provider of Platform services Deployment of for developers Provision of Software Blocks as utility services 13

FI-STAR at a glance • Future Internet Social and Technological Alignment Research (FISTAR) project – attempting to identify suitable software to data solutions based on Generic Enabler technology provided from FIWARE – to establish early trials in the health domain • FI-STAR will develop and deploy 8 use case Future Internet (FI) applications FI-STAR Provider: FI-WARE FI-STAR Use Case developers Provision of Private Cloud Deployment of Software Blocks as utility services 14

15

Building blocks of FI-Applications • Generic Enablers (GEs) are considered as software modules: – offer various functionalities along with protocols – interfaces for operation and communication. • Using GEs FI-STAR will develop Specific Enablers that highlight: 1. 2. 3. 4. Light-weight applications Robust security and privacy functionality Optimal usage of network resources Virtualization of resources 16

The Inter-Cloud Model

The Software to Data (S 2 D) • S 2 D is designed as a modular cloud Saa. S to allow easy deployment and configuration using API interfaces. • It is designed to allow migration configuration using either a web-based user interface or an xml containing all the needed information.

The Software to Data (S 2 D) • Modes: • The mode being responsible for transferring and deploying the software from public to private cloud. • The mode responsible for monitoring the usage of the software (for instance its uptime). • The mode that allows the usage of the S 2 D as an API that oer the user an alternative way containing all the needed information in an XML document.

Performance analysis

Conclusion and future work • • • This work presented an inter-cloud service architecture by utilizing Open. Stack, FI-WARE DCRM GE and AWS APIs. The current version is draft and supports basic functionalities in order to integrate an inter-cloud GE that is deployed into a cloud. This includes RESTFul APIs and schemes along with the AMPQ models. We will explore Open. Nebula, AWS and VCloud along with Cloud. Stack RESTFul interfaces and APIs in order to expand the inter-cloud services. This work is part of the Future Internet Social Technological Alignment Research (FI-STAR) project, which is a Future Internet Private Public Partnership (FI-PPP) run by the European Commission

For questions or access to the service: • Please contact me on s. sotiriadis@intelligence. tuc. gr • www. intelligence. tuc. gr Thanks!