We wrote a paper about this back in 2008:

Ian Foster, Yong Zhao, Ioan Raicu, Shiyong Lu. "Cloud Computing and Grid Computing 360-Degree Compared", IEEE Grid Computing Environments (GCE08) 2008, co-located with IEEE/ACM Supercomputing 2008.

You can download it from http://datasys.cs.iit.edu/publications/2008_GCE08_Clouds_Grids.pdf

I know some things have changed, but the paper has become a classic, and the answer to your question is nevertheless there, and still holds true today,.

They are the same, the grid is almost free, the cloud in some cases is not (public)

I think cloud computing is more commercial than grid.

And the elasticity is the great differential.

Sometimes (few times) i saw grid and cloud computing working together.

Grid was designed by the researchers to conduct their tasks over high power and high storage capability.

This eventually becomes base for cloud. And when we commercialize those resources/capabilities we came up with the concept of cloud.

But I do agree with the concept that we can contribute to Grid but not easily to cloud.

I also read an article with the following conclusions:

1) grid efforts often focus on running very large applications in batch mode while the cloud isn't a batch system and has a focus in the web application delivery space as opposed to data centers

2) grid typically requires expert knowledge in setting up and in execution (the end users are at least proficient with distributed computing) while cloud does not require any proficiency at least from the end user (he does not have to care about the mapping of task into resources, etc.)

3) it can be argued that cloud computing can be offered on top of grid computing by offering on-demand resource provisioning

Both cloud and grid are more or less same.... The crux of cloud computing is the effective User Interface/Interaction which you don't have in grid computing.

We can differentiate grid computing from cloud computing in the light of visualization of resources available with a cloud computing infrastructure.

I would consider GRID and Cloud as complementary services. There Public "free offerings for both in most countries (at least for academic purposes). Furthermore the term Cloud is quite generic and may imply one of the following.

IaaS: Infrastructure as a Service.

PaaS: Platfrom as a Service

SaaS: Software as a Service

SToraS: Storage as a Service.

Grid Computing can also be considered as PaaS

Check http://arxiv.org/ftp/arxiv/papers/0901/0901.0131.pdf where I. Foster et al. ellaborates on a comparison between Grid and Cloud computing systems.

This is an interesting article even if a bit old http://arxiv.org/ftp/arxiv/papers/0901/0901.0131.pdf.

I don't want to repeat my colleagues, you can find plenty of documents describing features and differences of the two paradigms.

Let me point out that today Grid and Cloud can be used in a combined way to give a set of specific Serivices to the final user (the ones that he needs) in terms of software and infrastructure.

Foster's paper suggested by A. Gouglidis is a very good reference on this topic. A while ago I wrote about this in my Ph.D. thesis. Here is an excerpt :

On the one hand, a quick comparison shows many similarities between both initiatives, something that years ago led some people to suggest that the cloud was nothing but the grid, simply presented from a new, market-oriented, perspective. On the other hand, it could be claimed that, although grids and clouds are both large scale distributed initiatives and therefore share many basic characteristics, cloud computing introduces several key aspects, creating a whole new paradigm.

There are important differences, especially when considering architectural aspects. Grid computing, on the one hand, is about independent institutions all around the world sharing resources in an harmonious way, and therefore it deals not only with large scale distribution and scalability issues but incorporates elements regarding security, trust between partners, decentralized control and so on. Cloud computing on the other hand, bases its infrastructure on large scale distributed data centers, in most cases centrally managed by an unique company (like Amazon in the case of EC2). These centralized computing resources are nevertheless of a very large scale, and therefore its managers are faced with scalability, distribution and other related issues, as in the case of the grid. However, this more controlled (to some extent) environment allows cloud service providers to develop more reliable low-level infrastructures and therefore focus on high-level service related issues and its market-oriented model. Nevertheless, most grid and cloud problems are still the same. Both need to be able to manage large scale (yet somehow different) facilities. They both need to define methods by which users/consumers discover, request and use resources provided by the system. Additionally, they both need to provide the users/consumers with the necessary mechanisms to develop the often highly parallel computations that execute on those resources.

The high expectations created by the grid community at its conception led over the years to disappointment and criticism, when many realized that the required efforts to create the grid were too great. In this sense, Cloud computing could be regarded as a compromise, where some of the technologically more challenging problems that grid computing created are eliminated by basing the cloud on a simpler infrastructure. This apparent simplification has led to the creation of new interesting platforms that provide grid-like services (specially if we remember the electric power grid analogy), with very promising scientific but most of all commercial results. But in a world where everything in computer science is moving towards a totally distributed, shared model (the popularity of, for instance, the Wikipedia project and the peer-to-peer systems are living proofs of this trend) it does not seem logical to expect that data center based, centralized structures such as most current commercial clouds (Amazon EC2, Google App Engine, etc.) should not in time evolve into a more widely distributed, resource-sharing approach. The recent advances in multi-cloud technology can be seen as proof of this trend.

I hope this helps :)

I would summarize the differences between the two as the following.

- Grid contains joined pools of resources in order to integrate the computing resource between organizations, where the infrastructure of Cloud usually belong to only one organization. It can be public (allow access from outside the organization, usually commercial), or private.

- Grid is usually for community usage where (public) Cloud is for commercial

- The usage of Grid is to submit program and data into the system and it will manage submitted jobs to be executed by using the appropriate resource (CPU, storage, etc.), where Cloud provides three paradigms of services Iaas, Paas, and Saas

- Applications on Cloud run on the virtual environment which provides lower upfront cost (for Cloud user) and more utilization of the server (for Cloud provider). For user, the cost is pay per usage, while you have to pay for the whole data center for Grid.

- Grid has the concept of virtual organizations, which restrict the access of user, while Cloud provide full privileges for the virtual environment.

That's all I can think of now.

Cloud is a business model. It is a different way of selling. F.e. in on premise product (email) you buy CD with software, machines to run it, hire people to install it and maintain it. This model has a number of disadvantages. You have to hire and train IT people to have proper expertise. And this is not cheap. You have to buy and renovate your hardware that is not cheap ether. You have to make backups and upgrades that are time consuming. So if you sum this up and count TCO (total cost of ownership) you will see that cloud services are cheaper. And that is the true essence for customer. For developer it is to make this prosperous business :) It is done by multi-tenancy - single application instance can be reused for requests from multiple customers. You are trying to make it as efficient as possible on your site , so TCO for you is lower than money you earn from customers.

Grid computing is mean of parallel computation for big data. You can install it on your computer or cluster if have one :) Or you can rent it via some cloud service.

In summary, Grid and Cloud deal distributed resources to run parallel applications. Grid is a general term, and Cloud (more specific term) involves hardware virtualization, which permits deploy several computational cenarios with operational system and applications. I think that the main feature about Cloud is virtualization.

To some extent, keywords and research topics follow the trends, so, at present, it is more common to hear about Cloud computing than about Grid computing. Both are quite general terms, and each of them comprises a lot of sub-topics and can be considered from different points of view.

In general, a Grid is a set of geographically distributed computing resources (typically CPU, memory and storage) able to cooperate, through a suitable middleware, to achieve some goal. Resources are as a rule non-dedicated, heterogeneous and loosely coupled. However, besides the basic Grid platforms, more specialized ones have been proposed during years, for examples more oriented to the execution of parallel, big jobs, or best suited for applications with demanding QoS requirements.

On the other hand Clouds can be seen as an evolution of the old computing centers, where resources are fully virtualized and delocalized.

Each node of a Grid platform can be, in principle, any machine, from a very simple one (e.g. a single PC or workstation) to more complex ones (e.g. local clusters, heterogeneous CPU/GPU platforms and so on). As a particular case a node may also be a Cloud platform.

In this sense Grid computing can be considered more general than Cloud computing.

We could also have a Grid composed of several Cloud platforms. On the opposite side, in principle (but it rarely happens) a Cloud platform could be constituted by a very high number of distributed resources, following a Grid-like approach.

Both Grid and Cloud approaches rely on a number of common enabling technologies, e.g. high speed interconnecting networks. Both try to answer to the need of accessing remote resources in a ‘ideally’ fully transparent manner . Both rely on techniques for

discover and select the best suited resources (Grid nodes or Cloud platforms) for a given user.

Among the most important differences I mention:

- Grid platforms are decentralized and loosely connected, whereas Cloud systems are usually concentrated in a single site or at most in few sites, tightly connected;

- Grids rely on open source middleware, whereas Clouds use proprietary software;

- The use of Grid resources is usually free, whereas Cloud services are paid by users (however among the various Grid flavors there are also the so-called ‘economic Grids’, in which resources have a cost and users an available budget; economic Grids can be viewed as an intermediate step from Grid to Cloud, nevertheless in my opinion is not correct to consider Clouds as an evolution of Grids, since they are simply two different things );

- In Grid platforms the emphasis is put on the collaboration among resources to solve a big problem, or to achieve a very high throughput, whereas in Cloud platforms the emphasis is in the provision of on-demand services.

Summarizing, I think that Grid and Cloud approaches are different, although they try to answer to partially similar needs and rely in part on similar technologies. Grid computing is in my opinion more general, whereas the main goal of Cloud platforms is to supply on-demand a number of services (e.g. IaaS, PaaS, SaaS).

We wrote a paper about this back in 2008:

Ian Foster, Yong Zhao, Ioan Raicu, Shiyong Lu. "Cloud Computing and Grid Computing 360-Degree Compared", IEEE Grid Computing Environments (GCE08) 2008, co-located with IEEE/ACM Supercomputing 2008.

You can download it from http://datasys.cs.iit.edu/publications/2008_GCE08_Clouds_Grids.pdf

I know some things have changed, but the paper has become a classic, and the answer to your question is nevertheless there, and still holds true today,.

Simply you can say .... Cloud computing is integrated form of Grid ....

Thanks for all your interesting responses.

I'm surprised, because - as many of you define clouds in various manners - virtually none of you mentioned the feature that - IMHO - is the most basic and a "conditione sine qua non" of a cloud - *virtualization*!

And - related to it - flexible allocation of resources.

Obviously, none of them is prohibited on a grid!

On the other hand, a simple and easy-to-use interface is common for clouds and grids. Please note, that most grid users are not computer guys, but... physicists!

>from Bartlomiej Kubica

>most grid users are not computer guys, but... physicists!

and chemical, bio, earth scientists, even from the field of medicine...

I feel that this can viewed in this way....Grids are mostly made for distribution of resources, whereas, cloud is made for supply of resources. a grid of clouds can be made to supply resources. In clouds the resources are in form of clusters at data-centers, whereas, in grids, the resources separately connect to help solve a distributed problem.

Grid and cloud is varied based on a method called virtualization. In grid, virtualization is multiple resources or other entities are grouped together and viewed as single one whereas in cloud, a single entity is divided into multiple instances each for one consumer.

http://arxiv.org/ftp/arxiv/papers/0901/0901.0131.pdf

Resource distribution: Cloud computing is a centralized model whereas grid computing is a decentralized model where the computation could occur over many administrative domains.

Ownership: A grid is a collection of computers which is owned by multiple parties in multiple locations and connected together so that users can share the combined power of resources. Whereas a cloud is a collection of computers usually owned by a single party.

Examples of Clouds: Amazon Web Services (AWS), Google App Engine

Examples of Grids: FutureGrid

Dropbox, Gmail, Facebook, Youtube, Rapidshare, etc are all examples of cloud computing services

Cloud Computing - http://en.wikipedia.org/wiki/Cloud_computing

Cloud computing is a computing paradigm shift where computing is moved away from personal computers or an individual application server to a “cloud” of computers. Users of the cloud only need to be concerned with the computing service being asked for, as the underlying details of how it is achieved are hidden. This method of distributed computing is done through pooling all computer resources together and being managed by software rather than a human.

The services being requested of a cloud are not limited to using web applications, but can also be IT management tasks such as requesting of systems, a software stack or a specific web appliance.  

Grid Computing - http://en.wikipedia.org/wiki/Grid_computing

Multiple independent computing clusters which act like a “grid” because they are composed of resource nodes not located within a single administrative domain. (formal)

Offering online computation or storage as a metered commercial service, known as utility computing, computing on demand, or cloud computing.

The creation of a “virtual supercomputer” by using spare computing resources within an organization

Cloud computing

With cloud computing, companies can scale up to massive capacities in an instant without having to invest in new infrastructure, train new personnel, or license new software. Cloud computing is of particular benefit to small and medium-sized businesses who wish to completely outsource their data-center infrastructure, or large companies who wish to get peak load capacity without incurring the higher cost of building larger data centers internally. In both instances, service consumers use what they need on the Internet and pay only for what they use.

The service consumer no longer has to be at a PC, use an application from the PC, or purchase a specific version that's configured for smartphones, PDAs, and other devices. The consumer does not own the infrastructure, software, or platform in the cloud. He has lower upfront costs, capital expenses, and operating expenses. He does not care about how servers and networks are maintained in the cloud. The consumer can access multiple servers anywhere on the globe without knowing which ones and where they are located.

Grid computing

Cloud computing evolves from grid computing and provides on-demand resource provisioning. Grid computing may or may not be in the cloud depending on what type of users are using it. If the users are systems administrators and integrators, they care how things are maintained in the cloud. They upgrade, install, and virtualize servers and applications. If the users are consumers, they do not care how things are run in the system.

Grid computing requires the use of software that can divide and farm out pieces of a program as one large system image to several thousand computers. One concern about grid is that if one piece of the software on a node fails, other pieces of the software on other nodes may fail. This is alleviated if that component has a failover component on another node, but problems can still arise if components rely on other pieces of software to accomplish one or more grid computing tasks. Large system images and associated hardware to operate and maintain them can contribute to large capital and operating expenses.

Source: IBM

@Muhammad Javed

"Cloud computing is a computing paradigm shift where computing is moved away from personal computers or an individual application server to a “cloud” of computers."

Please! PLEASE! No, "paradigm shifts", no buzzwords like that. This topic was intended t *clarify* the notions, to be "as precise as posible", not to produce even more noise.

Also, how is it going to be a "paradigm shift"? A "paradigm shift", if we are to use the notion, menas a revolutionary, very important change of basic views, approaches, technologies, etc.

And moving computing away from a personal computer was done already in mainframe architectures and application serers - *decades* ago, literally! Cloud computing is just making "one step more", by adding the virtualization and fleexibility (which are good things, obviously).

I beg you once more: only facts, only precise, clear notions, no marketing, no commercials!

If you like an one-sentence answer, I would consider cloud computing as an upgraded version of grid computing plus virtualization, and the computing infrastructure is under a single administration.

THE fact is, when you are applying for a fund, by supplying your proposal's title with terms including "cloud" instead of "grid", your chance of obtaining support would be way higher. I do not agree with the so-called distributing or supplying resources since they are in a sense conceptually interchangeable. Apart from a bit of technical improvements such as virtualization, these two terms share a common fundamental technical contributions. To my impression, "Cloud" is more of a floating business concept since the time Amazon fould their server clusters have a commercial value.

To add more, Grid computing means resource sharing. Cloud computing means services sharing. Both belong to the domain of distributed computing.

In cloud computing, we get the application(SaaS) or platform(PaaS) or Computing resource(IaaS) as services by having access through internet. No need for the client to have anything other than the Browser with the internet connectivity.

If you want to access the grid, you have to register and after obtaining the Certificate for authentication from the approved Certifying Authority, you are allowed with single sign on facility. You must have the client code loaded in your client machine to access grid resources. The access of resources is one of the main differences between grid and cloud computing.

We use the Grid middleware such as Globus Toolkit or gLite or UNICORE for job submission utilizing High Performance Computing resources. The resources had been registered. The standard APIs are available to get the information about resources, have security management, manage data for manipulation, manage resources, transfer data and code to resources, execute job and to transfer the output after data manipulation, The middleware enables these functions and the task is completed. By using meta scheduler, the functions are simplified. Grid computing takes place in a secured environment. In cloud also we can provide the security, but at an extra cost only.

In cloud environment, the open source IaaS provider enables the resources available to the user. To submit a job to the resources, only the user has to prepare the execution environment for job submission. So many other works such as SLA, security management, transport management, job monitoring etc., to be taken care of by the user. Else, we have to pay a large amount for these purposes. There is no standard Cloud Resource Broker available at free of cost. In grid environment, GT 5 has been integrated with Gridway meta-scheduler. The job submission has to be done by the user and all the other activities are automatically performed ad the results are obtained by the user. GT and Gridway are at free of cost.

In grid, we have meta scheduler as part of it where as in cloud, there is no meta scheduler available at free of cost.

Cloud is much more popular because of the virtualization that enables dynamic resource provisioning for any requirement. Mostly, the physical resources are used in Grid computing.

Because of Software Defined Networking and Network Function Virtualization, the network resources are managed efficiently and economically. We have virtualized computing resources, virtualized Storage resources and virtualized Network resources. We must have the meta scheduler to manage these three resources efficiently to succeed in distributed computing.

In addition, read the following paper also as suggested by Ioan Raicu.

Ian Foster, Yong Zhao, Ioan Raicu, Shiyong Lu. "Cloud Computing and Grid Computing 360-Degree Compared", IEEE Grid Computing Environments (GCE08) 2008, co-located with IEEE/ACM Supercomputing 2008.

Please see some differences at:

http://www.researchgate.net/publication/224587188_Intrusion_Detection_for_Grid_and_Cloud_Computing?ev=prf_pub

Article Intrusion Detection for Grid and Cloud Computing

Thanks again for all the answers. So, everyone has a completely different opinion from the others... ;-)

It seems, an important source of confusion is the ambiguity of the notion of GRID.

Maybe we should distinguish:

- a ``technological grid'' - a system consisting of heterogeneous computers, connected via Internet,

- an ``organizational grid'' - a system, consisting of machines belonging to several parties, cooperating as a single system.

Now, we can say the cloud can be implemented over a ``technological grid'', but is differently organized from the ``organizational grid''.

The terms sound clumsilly, but possibly you can come up with better ones...

Links about grid computing:

User Maturity Based Trust Management for Grid Computing

https://www.researchgate.net/publication/4332698_User_Maturity_Based_Trust_Management_for_Grid_Computing?ev=prf_pub

Towards Advance Reservation in Large-Scale Grids

https://www.researchgate.net/publication/4331931_Towards_Advance_Reservation_in_Large-Scale_Grids?ev=prf_pub

Intrusion Detection for Computational Grids

https://www.researchgate.net/publication/220726782_Intrusion_Detection_for_Computational_Grids?ev=prf_pub

Design and Evaluation of a Grid Computing Based Architecture for Integrating Heterogeneous IDSs

https://www.researchgate.net/publication/4302712_Design_and_Evaluation_of_a_Grid_Computing_Based_Architecture_for_Integrating_Heterogeneous_IDSs?ev=prf_pub

Grid-M : Middleware para Integrar Dispositivos Móveis, Sensores e Grids

https://www.researchgate.net/publication/254861948_Grid-M__Middleware_para_Integrar_Dispositivos_Mveis_Sensores_e_Grids?ev=prf_pub

A Grid-based Intrusion Detection System

https://www.researchgate.net/publication/224633737_A_Grid-based_Intrusion_Detection_System?ev=prf_pub

Grid Middleware for Mobile Decision Support Systems

https://www.researchgate.net/publication/4240527_Grid_Middleware_for_Mobile_Decision_Support_Systems?ev=prf_pub

Towards a Middleware for Mobile Grids

https://www.researchgate.net/publication/224645660_Towards_a_Middleware_for_Mobile_Grids?ev=prf_pub

Towards Grid-based Intrusion Detection

https://www.researchgate.net/publication/224645659_Towards_Grid-based_Intrusion_Detection?ev=prf_pub

Towards a Grid of Sensors for Telemedicine.

https://www.researchgate.net/publication/221031346_Towards_a_Grid_of_Sensors_for_Telemedicine?ev=prf_pub

Defending Grids Against Intrusions.

https://www.researchgate.net/publication/220875450_Defending_Grids_Against_Intrusions?ev=prf_pub

Conference Paper User Maturity Based Trust Management for Grid Computing

Conference Paper Towards Advance Reservation in Large-Scale Grids

Conference Paper Intrusion Detection for Computational Grids

Conference Paper Design and Evaluation of a Grid Computing Based Architecture...

Conference Paper Grid-M : Middleware para Integrar Dispositivos Móveis, Senso...

Conference Paper A Grid-based Intrusion Detection System

Conference Paper Grid Middleware for Mobile Decision Support Systems

Conference Paper Towards a Middleware for Mobile Grids

Conference Paper Towards Grid-based Intrusion Detection

Conference Paper Towards a Grid of Sensors for Telemedicine

Conference Paper Defending Grids Against Intrusions

One more link about grid of agents:

Grids of agents for computer and telecommunication network management.

http://www.researchgate.net/publication/220105727_Grids_of_agents_for_computer_and_telecommunication_network_management?ev=prf_pub

Article Grids of agents for computer and telecommunication network management

Links about Cloud Computing:

Optimizing Green Clouds through Legacy Network Infrastructure Management

https://www.researchgate.net/publication/259392181_Optimizing_Green_Clouds_through_Legacy_Network_Infrastructure_Management?ev=prf_pub

Decision-Theoretic Planning for Cloud Computing

https://www.researchgate.net/publication/259399614_Decision-Theoretic_Planning_for_Cloud_Computing?ev=prf_pub

Privacy-preserving Identity Federations in the Cloud - A Proof of Concept

https://www.researchgate.net/publication/259715881_Privacy-preserving_Identity_Federations_in_the_Cloud_-_A_Proof_of_Concept?ev=prf_pub

Provisioning, Resource Allocation, and DVFS in Green Clouds

https://www.researchgate.net/publication/259913665_Provisioning_Resource_Allocation_and_DVFS_in_Green_Clouds?ev=prf_pub

Risk-based Dynamic Access Control for a Highly Scalable Cloud Federation

https://www.researchgate.net/publication/254561229_Risk-based_Dynamic_Access_Control_for_a_Highly_Scalable_Cloud_Federation?ev=prf_pub

Challenges of Operationalizing PACS on Cloud Over Wireless Networks

https://www.researchgate.net/publication/254366302_Challenges_of_Operationalizing_PACS_on_Cloud_Over_Wireless_Networks?ev=prf_pub

A Review of PACS on Cloud for Archiving Secure Medical Images

https://www.researchgate.net/publication/255964728_A_Review_of_PACS_on_Cloud_for_Archiving_Secure_Medical_Images?ev=prf_pub

Environment, Services and Network Management for Green Clouds

https://www.researchgate.net/publication/254263747_Environment_Services_and_Network_Management_for_Green_Clouds?ev=prf_pub

Comparison of a Multi output Adaptative Neuro-Fuzzy Inference System (MANFIS) and Multi Layer Perceptron (MLP) in Cloud Computing Provisioning

https://www.researchgate.net/publication/241677983_Comparison_of_a_Multi_output_Adaptative_Neuro-Fuzzy_Inference_System_%28MANFIS%29_and_Multi_Layer_Perceptron_%28MLP%29_in_Cloud_Computing_Provisioning?ev=prf_pub

Multi-Tenancy Authorization System with Federated Identity for Cloud-Based Environments Using Shibboleth

https://www.researchgate.net/publication/257200931_Multi-Tenancy_Authorization_System_with_Federated_Identity_for_Cloud-Based_Environments_Using_Shibboleth?ev=prf_pub

Toward an architecture for monitoring private clouds.

https://www.researchgate.net/publication/220144318_Toward_an_architecture_for_monitoring_private_clouds?ev=prf_pub

Experimental Assessment of Routing for Grid and Cloud

https://www.researchgate.net/publication/254347753_Experimental_Assessment_of_Routing_for_Grid_and_Cloud?ev=prf_pub

Simulator improvements to validate the Green Cloud Computing approach.

https://www.researchgate.net/publication/221430419_Simulator_improvements_to_validate_the_Green_Cloud_Computing_approach?ev=prf_pub

Customer Security Concerns in Cloud Computing

https://www.researchgate.net/publication/228451862_Customer_Security_Concerns_in_Cloud_Computing?ev=prf_pub

A Cloud Computing Solution for Patient's Data Collection in Health Care Institutions.

https://www.researchgate.net/publication/221222788_A_Cloud_Computing_Solution_for_Patient%27s_Data_Collection_in_Health_Care_Institutions?ev=prf_pub

SLA Perspective in Security Management for Cloud Computing

https://www.researchgate.net/publication/232631054_SLA_Perspective_in_Security_Management_for_Cloud_Computing?ev=prf_pub

Conference Paper Optimizing Green Clouds through Legacy Network Infrastructur...

Conference Paper Decision-Theoretic Planning for Cloud Computing

Article Privacy-preserving Identity Federations in the Cloud - A Pro...

Article Provisioning, Resource Allocation, and DVFS in Green Clouds

Conference Paper Risk-based Dynamic Access Control for a Highly Scalable Clou...

Conference Paper Challenges of Operationalizing PACS on Cloud Over Wireless Networks

Article A Review of PACS on Cloud for Archiving Secure Medical Images

Article Environment, Services and Network Management for Green Clouds

Article Comparison of a Multi output Adaptative Neuro-Fuzzy Inferenc...

Conference Paper Multi-Tenancy Authorization System with Federated Identity f...

Article Toward an Architecture for Monitoring Private Clouds

Conference Paper Experimental Assessment of Routing for Grid and Cloud

Conference Paper Simulator improvements to validate the Green Cloud Computing approach

Article Customer Security Concerns in Cloud Computing

Conference Paper A Cloud Computing Solution for Patient's Data Collection in ...

Article SLA Perspective in Security Management for Cloud Computing

To see the differences, baybe it's easier to explain in which scenario each one is used:

- Grid: Usually for batch-like execution of tasks with large computation demands (e.g. weather forecast simulations); the grid will split the task into subtasks, schedule their execution on one or several infrastructures, and send back the result*.

- Infrastructure as a Service (IaaS) cloud: Used as a source of virtual resources that resemble physical ones (virtual machines, virtual storage, virtual networks). On top of those resources you can build a grid, for example.

- Platform as a Service (PaaS) cloud: Used as a runtime/container for software components (servlets, python scripts...). Maybe this one resembles a bit Grid systems, however there is a difference: grids execute batch computation tasks; a PaaS runs/hosts a software component so it can work as an interactive service for long periods of time (like a servlet getting http requests). Also, a PaaS can try to scale the component (creating new replicas), but it won't 'split' like a grid can do with tasks.

- Service as a System (SaaS) cloud: Basically, any application provided by a third-party with a web interface ;) . GMail is an example.

I hope it helps!!

*(Note: if you think that the 'Grid' definition can be applied to many big data systems... I'd say you are right ;) )

I think the 2008 paper by Foster et al. is very helpful in comparing grid and cloud computing. In computing time flies, of course, and we hear much more about cloud computing today and less about computational grids. As the paper notes, large companies like Amazon, Google and Microsoft now offer computing on demand using only a credit card for charges. Virtualization has advanced a great deal (e.g. VMWare) and may scientists are making good use of the new commercial offerings. The NSF Ocean Observatories Initiative makes use of virtual machines on commercially available clouds, but also relies on VMs running on computational resources located at system Cyberinfrastructure Points of Presence (CyberPoP's) when security is particularly important and management of large quantities of data are necessary. From the point of view of Foster et al. grids and clouds remain important, and are beginning to share more and more software through virtualization.

With regard to "paradigm," I generally recommend substituting "pachyderm" for "paradigm." It makes about as much sense. I think, looking back, that plate tectonics was a paradigm shift, but it took a very long time for it to be recognized for what it is. It's way to early to say that "cloud computing" will stand the test of time as a major change in the way we're dealing with scientific problems.

Since both terms are 'fuzzy', and definitions of either depend on what the different people want to sell (or get funding for), I'd have to go with Heng Yu: these are mostly marketing terms, like Web 2.0, etc.

There are no definitions of either (especially Cloud Computing) because no one really wants to have one, or no party wants to accept the definition of another party, because they're both 'selling' or 'pitching' something different.

That doesn't mean the attempts at definition above are inaccurate. It's all accurate to some extend, but only describes some (usually arbitrary) aspect of both terms, that may or may not hold for various situation, but not probably not in general.

Scientists like to qualify, quantify, categorise, and draw distinctions. For Cloud/Grid Computing this is a futile excercise.

For example, all the 'positive' things in Muhammed Sharif's answer can also apply to Gird Computing applications/implementations, and for each point I can argue the case for the opposite for Cloud Computing. I can do the same for his bit on Grid Computing (Systems).

In my view, Cloud Computing AND Grid Computing are both snazzy marketing terms for 'Distributed Computing with bells on'. We all know what Distributed Computing means, but the term doesn't have 'zing' to it. And developments in Distributed Computing led to the term Grid Computing, and then to the term Cloud Computing, when a new snazzy word was needed for get funding or to sell stuff/services. In the end it is still rooted in Distributed Computing with this, that, or the other emphasis or development.

If scientists didn't have to beg for money from funding agencies, or companies didn't need to market/sell some of the services they wish to provide, we wouldn't need all these snazzy terms, and we'd probably still call it Distributed Computing.

I agree, both terms are used as buzzwords, but it does not mean, they make no sense. Grids are very different from clusters (heterogenous hardware and OS, importance of use of open standards, security problems, common failures).

As for clouds, it seems, most important featues are:

- virtualization,

- privacy issues,

- billing problems.

Virtualization can be used in clusetrs and grids, but the other two seem pretty new problems. That's my opinion, after the long discussion above...

Neither grids nor clusters require heterogeneous hardware and OS. As you said because of virtualisation, but also because of portals and interfaces for specific applications. In fact, none of the distinctions you mention truly describe either grids, clusters or clouds.

You're welcome to your opinion naturally, but are you sure you want to claim privacy and billing problems/issues exclusively or as most important for cloud computing, but not for grids, or distributed computing in general? I.e., cloud computing is something with virtualisation with privacy and billing issues and problems? Seems a bit silly.

And this is the problem in an academic context. A lot of seemingly arbitrary lines are drawn to distinguish between 'systems' that, to my mind stubbornly, refuse to be defined by them. If that is the case, wouldn't it be more honest to admit that the attempt a classification has failed?

Grids, clusters, blades, clouds, etc. They aren't technical terms. They don't define systems. They aren't meant to distinguish one from the other either. They are buzzwords, meant to market an idea: we're doing something new; give us money/funding/recognition/whatever. That's all.

Fine, nothing wrong with it, happens all the time. Just let's not get carried away and assume that it actually means something in reality. In the end all these things are just new(ish) ways and applications and combinations thereof of distributed computing.

In my opinion ofcourse.

Grid and Cloud Computing concepts derived from the Distributed Computing domain. Grid Computing is based on Virtual Organization (VO) to provide resources mainly for scientific and Research problems while Cloud Computing is based on Virtualization to provide resources as services for commerical problems.

Cloud and Grid computing have both emerged from the ‘‘computing as a utility’’ paradigm. Nevertheless, whereas Grid computing evolved to a cooperation model where resources are shared between (sometimes spread around the world) organization or between departments in the same organization, Cloud computing is often based on a business model where users have on demand access to elastic resources on a pay-per-use model.  (https://www.researchgate.net/publication/263012600_Introduction_to_the_special_issue_on_Grid_and_Cloud_Computing_Current_Advances_and_New_Research_Trends)

Article Introduction to the special issue on Grid and Cloud Computin...

There is apparently few difference separating cloud computing and grid computing, Cloud Computing and Grid Computing are both forms of distributed computing. Cloud computing and grid computing include some common features such as Heavy use of abstraction, or masking the actual complex process taking place inside the system, as well as instead providing users with a simpler and easy to use interface, Scalability, or varying computing loads according to load demands. Both cloud computing and grid computing tend to draw more resources from the pool when the processes need so, and surrender the capacity or resources not needed to the common pool, for other users, Multi-tenancy and multitask, or the ability to perform many different tasks simultaneously.

Grid makes use of the high end computing facility with the platform suitable for executing your application. If the computing infrastructure does not possess the essential r execution environment of the application of users, the resource is not allocated to the job submitted. The job will go to pending state. In grid, the middleware takes care of the job execution and storage of the result. By virtualizing the resources, grid may be extended to cloud services. Grid is the base for the cloud.   

Cloud is the refinement of grid services combining virtualization and web services. Customization, configuration  and externalization  are the terms used in cloud. All these are happening in grid also.  Multi tenancy is a new concept  in cloud.

In cloud, it is possible to get the computing hardware facility as IaaS and you can configure the system with your OS and other necessary APIs meant  for your application. The required execution environment may be created using IaaSa and the job may be successfully executed. Using PaaS, we can develop the application of our requirement using the available APIs of PaaS.  SaaS is the application developed by vendors and you can access it for that specific application purpose only. In cloud, we can easily access the computing resource, storage resource and network resource(SDN) separately by means of virtualization and you need only a browser to access IaaS ot PaaS or SaaS.

Grid computing system is a widely distributed resource for a common goal. It is Brother of Cloud Computing and Sister of Supercomputer. We can think the grid is a distributed system connected to a single network. This types of computing work with the large volume of files. Basically, it is a cluster types system. So people call it cluster computing.

Source: https://www.cloudwebhostingtips.com/grid-computing-vs-cloud-computing-supercomputer/

Grid computer tends to be more geographically disperse and heterogeneous by nature. Grid network also has various types. A single grid is like dedicated connection but a common grid perform multiple tasks.

The size of the grid is large. So grid computing is like supercomputing. It consists of many network, computer, and middleware. Grid computer is dedicated to some specific function of the large volume of data. In the grid process, each task divided into a various process. All the process starts execution simultaneously on a different computer. As a result, very few seconds needs to execute and enjoy the flavor of supercomputing.

For more:https://www.cloudwebhostingtips.com/