INTRODUCTION TO GRID CLOUD COMPUTING U Jhashuva 1

INTRODUCTION TO GRID & CLOUD COMPUTING U. Jhashuva 1 Asst. Professor Dept. of CSE

TOPICS • Grid Computing – Definition – Characteristics – Architecture – Services – Applications • Cloud Computing – Architecture – Services – Applications • Grid vs. Cloud

GRID COMPUTING Definition: A grid computing is a computing environment where sharing of computing power, may be computing cycle, resources, and storage is taken place to accomplish the task efficiently and quickly

GRID COMPUTING Grid Information Service R 2 database R 3 R 5 Resource Broker Grid Resource Broker Application R 4 RN R 6 R 1 Resource Broker Grid Information Service Figure : A Bird view of global grid (from Rajkumar buyya presentation)

GRID COMPUTING Characteristics: • Large Scale : – A grid is able to deal with any number of resources (either system or memory). It can range from few to millions. • Global Distribution: – Grid computing resources can be in any location in the global. Cont. .

GRID COMPUTING • Loosely Coupled: – Main characteristics of grid computing is loosely coupled architecture. • Distributed Job Management: – It has the characteristics to split the job for various resources or computers to avoid the work pending in the queue for long time. Cont. .

GRID COMPUTING • Consistent Access: – A grid is designed in such a way to give the consistent data and also other resources accessing ability. • Resource Coordination: – Resources in a grid must be coordinated in such a way to provide cluster computing capabilities. Cont. .

GRID COMPUTING • Dependable Access: – A grid should assume the quality of service standards in order to provide an assure way of sharing the resources for the requested user • Transparent Access: – A grid has a capability to see as a single virtual computer even it has multiple resources across the global Cont. .

GRID COMPUTING • On-Demand Computing: • It provides an efficient way to access the resources that cannot be cost effectively or conveniently located in local area • Heterogeneity : – A grid can host both hardware and software that can be varied in data, sensors, devices, networks and so on.

GRID COMPUTING Core Middleware Application Development and Deployment Environment Resource management and scheduling Job submission, info services, Storage access, Trading, Accounting, License Distributed Resources Security Services Coupling Services Authentication, Single sign-on, secure communication System level Grid resources Desktops, servers, clusters, networks, applications, storage, devices + resource manager + monitor Autonomic/ Grid Economy User-Level Middleware Grid programming environment and tools Languages, API, libraries, compilers, parallelization tools Adaptive Management User level Grid applications Web Portals, Applications,

GRID COMPUTING Requirements: • Step 1: – The integration of individual s/w & h/w components into a combined networked resource (single system image cluster). • Step 2: – Low-level middleware to provide a secure and uniform access to services provided by different resources. Cont…

GRID COMPUTING • Step 3: – User-level middleware to support application development and aggregation of distributed resources. • Step 4: – The construction of Grid applications and Web portals.

GRID COMPUTING Categories of Grid Computing:

GRID COMPUTING • Computational Grid: – provides a secure way of accessing the resources from a huge collection (pool). This kind of grid is much useful for high throughput applications and computation intensive computing. • Collaboration Grid: – with the popularity of Internet, there is a huge demand for better collaboration. Using the grid, advance collaboration is possible. Cont. …

GRID COMPUTING • Utility Grid: – is not limited for sharing data and computation cycles but also software and resources can be shared. It provides main services like sharing Software and other special equipment. • Network Grid: – Provides fault-tolerant and high performance communication services. During communication each grid node act as a data router between two communication points. Cont. …

GRID COMPUTING • Data Grid: – provides an environment to support data discovery, data storage, data handling, data publication and data manipulation. It is stored in different types of databases and file systems.

CLOUD COMPUTING Definition: A cloud is a type of parallel and distributed system consisting of a collection of interconnected and virtualized computers that are dynamically provisioned and presented as one or more unified computing resources based on service-level agreements established through negotiation between the service provider and consumers.

CLOUD COMPUTING Characteristics: • No up-front commitments • On-demand access • Nice pricing • Simplified application acceleration and scalability • Efficient resource allocation • Energy efficiency • Seamless creation and use of third-party services

CLOUD COMPUTING Fig : Cloud Computing Architecture (from mastering cloud computing text book)

GRID Vs. CLOUD • Business Models: – While in Grid business models are usually based on bilateral agreements between academic institutions, provision of resource in Clouds requires more differentiated business models. • Resource Management: – While Grids rely on batch systems, utilization of virtualization technologies represents the resource management solution for the Clouds.

GRID Vs. CLOUD • Resource Provision Models: – Grid resource provisioning models are based on virtual organisations where the relationships are established offline. In Clouds usage of SLAs, compliance, and trust management is essential.

GRID Vs. CLOUD • Resource Availability: – In Grids resource sharing relies on the best effort manner, sometimes resources are not available and sometimes there are plenty of resources which are idle. Clouds rely on massive elasticity in Clouds. Challenging issues in Clouds are to find the balance between wasting resources due to the virtualization overhead and standby modes of devices on the one hand, and pooling of resources to facilitate efficient consumption of resources and reducing energy consumption on the other.

Thank you ANY QURIES ?

REFRENCES 1. http: //www. csiindia. org/digital_magazine/November/Nov. ht ml 2. Mastering in Cloud Computing 1. Rajkumar Buyya , Christian Vecchiola , S. Thamarai Selvi 3. Grid vs Cloud – A Technology Comparison 1. Ivona Brandic, Schahram Dustdar, Vienna University of Technology