Question

What is the difference between Cloud, Cluster and Grid? Please give some examples of each as the definition of cloud is very broad. As answered in another question, can I call Dropbox, Gmail, Facebook, Youtube, Rapidshare etc. a Cloud?

What are the examples of Cluster and Grid as well?

Was it helpful?

Solution

Cluster differs from Cloud and Grid in that a cluster is a group of computers connected by a local area network (LAN), whereas cloud and grid are more wide scale and can be geographically distributed. Another way to put it is to say that a cluster is tightly coupled, whereas a Grid or a cloud is loosely coupled. Also, clusters are made up of machines with similar hardware, whereas clouds and grids are made up of machines with possibly very different hardware configurations.

To know more about cloud computing, I recommend reading this paper: «Above the Clouds: A Berkeley View of Cloud Computing», Michael Armbrust, Armando Fox, Rean Griffith, Anthony D. Joseph, Randy H. Katz, Andrew Konwinski, Gunho Lee, David A. Patterson, Ariel Rabkin, Ion Stoica and Matei Zaharia. The following is an abstract from the above paper:

Cloud Computing refers to both the applications delivered as services over the Internet and the hardware and systems software in the datacenters that provide those services. The services themselves have long been referred to as Software as a Service (SaaS). The datacenter hardware and software is what we call a Cloud. When a Cloud is made available in a pay-as-you-go manner to the general public, we call it a Public Cloud; the service being sold is Utility Computing. We use the term Private Cloud to refer to internal datacenters of a business or other organization, not made available to the general public. Thus, Cloud Computing is the sum of SaaS and Utility Computing, but does not include Private Clouds. People can be users or providers of SaaS, or users or providers of Utility Computing.

The difference between a cloud and a grid can be expressed as below:

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

  2. 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.

Examples of cloud computing services: Dropbox, Gmail, Facebook, Youtube, RapidShare.

OTHER TIPS

Cloud: is simply an aggregate of computing power. You can think of the entire "cloud" as single server, for your purposes. It's conceptually much like an old school mainframe where you could submit your jobs to and have it return the result, except that nowadays the concept is applied more widely. (I.e. not just raw computing, also entire services, or storage ...)

Grid: a grid is simply many computers which together might solve a given problem/crunch data. The fundamental difference between a grid and a cluster is that in a grid each node is relatively independent of others; problems are solved in a divide and conquer fashion.

Cluster: conceptually it is essentially smashing up many machines to make a really big & powerful one. This is a much more difficult architecture than cloud or grid to get right because you have to orchestrate all nodes to work together, and provide consistency of things such as cache, memory, and not to mention clocks. Of course clouds have much the same problem, but unlike clusters clouds are not conceptually one big machine, so the entire architecture doesn't have to treat it as such. You can for instance not allocate the full capacity of your data center to a single request, whereas that is kind of the point of a cluster: to be able to throw 100% of the oomph at a single problem.

There's some pretty good answers here but I want to elaborate on all topics:

Cloud: shailesh's answer is awesome, nothing to add there! Basically, An application that's served seamlessly over the network can be considered a Cloud application. Cloud isn't a new invention and it's very similar to Grid computing, but it's more of a buzzword with the spike of recent popularity.

Grid: Grid is defined as a large collection as machines connected by a private network and offers a set of services to users, it acts as a sort of supercomputer by sharing processing power across the machines. Source: Tenenbaum, Andrew.

Cluster: A cluster is different from those two. Clusters are two or more computers who share a network connection that acts as a heart-beat. Clusters are configurable in Active-Active or Active-Passive ways. Active-Active being that each computer runs it's own set of services (Say, one runs a SQL instance, the other runs a web server) and they share some resources such as storage. If one of the computers in a cluster goes down the service fails over to the other node and almost seamlessly starts running there. Active-Passive is similar, but only one machine runs these services and only takes over once there's a failure.

Cloud is a marketing term, with the bare minimum feature relating to fast automated provisioning of new servers. HA, utility billing, etc are all features people can lump on top to define it to their own liking.

Grid [Computing] is an extension of clusters where multiple loosely coupled systems are used to solve a single problem. They tend to be multi-tenant, sharing some likeness to Clouds, but tend to rely heavily upon custom frameworks that manage the interop between grid nodes.

Cluster hosting is a specialization of clusters where a load balancer is used to direct incoming traffic to one of many worker nodes. It predates grid computing and doesn't rely on a homogenous abstraction of the underlying nodes as much as Grid computing. A web farm tends to have very specialized machines dedicated to each component type and is far more optimized for that specific task.

For pure hosting, Grid computing is the wrong tool. If you have no idea what your traffic shape is, then a Cloud would be useful. For predictable usage that changes at a reasonable pace, then a traditional cluster is fine and the most efficient.

Cloud: the hardware running the application scales to meet the demand (potentially crossing multiple machines, networks, etc).

Grid: the application scales to take as much hardware as possible (for example in the hope of finding extra-terrestrial intelligence).

Cluster: this is an old term referring to one OS instance or one DB instance installed across multiple machines. It was done with special OS handling, proprietary drivers, low latency network cards with fat cables, and various hardware bedfellows.

(We love you SGI, but notice that "Cloud" and "Grid" are available to the little guy and your NUMAlink never has been...)

my two cents worth ~

Cloud refers to an (imaginary/easily scalable) unlimited space and processing power. The term shields the underlying technologies and highlights solely its unlimited storage-space and power.

Grid is a group of physically close-by machines setup. Term usually imply the processing power (ie:MFLOPS/GFLOPS), referred by engineers

Cluster is a set of logically connected machines/device (like a clusters of harddisk, cluster of database). Term highlights how devices are able to connect together and operate as a unit, referred by engineers

Licensed under: CC-BY-SA with attribution
Not affiliated with StackOverflow
scroll top