There are many definitions out there on the internet , but I am asking here , just because someone may answer it briefly from his/her practical experience with these technologies.

How does cloud computing ( Here I mean Open-stack , Amazon ec2 , Google Compute Engine), grid computing and parallel computing differ and what is going to be the future trend , who is going to be the winner. My own understanding is that , inside a private cloud , it may be possible to run workloads that traditionally require a computing grid.

closed as primarily opinion-based by ceejayoz, yoonix, Iain, Michael Hampton Jan 8 '16 at 18:52

Many good questions generate some degree of opinion based on expert experience, but answers to this question will tend to be almost entirely based on opinions, rather than facts, references, or specific expertise. If this question can be reworded to fit the rules in the help center, please edit the question.

  • "Cloud" isn't solidly defined and what we know as cloud can be grid/parallel. It's incorrect to treat each term as mutually exclusive like you're doing, and there won't be "the winner" as there are different use cases. – ceejayoz Jan 8 '16 at 18:15
  • Your edit doesn't change my comments. It's possible on EC2 - a "cloud" host - to set up a grid computing infrastructure, as well as a parallel computing infrastructure. A "private cloud" isn't anything technologically different than a "public cloud", it's just running on your own hardware rather than hardware you rent. You're throwing terms around that are essentially meaningless in the context of the question. – ceejayoz Jan 8 '16 at 18:29
  • Got clear about it. eecs.berkeley.edu/Pubs/TechRpts/2009/EECS-2009-28.pdf – Ijaz Ahmad Khan Jan 17 '16 at 19:32

They differ based on what the marketing and sales team are trying to sell...

In all honesty, at least to me:

Cloud - the name for Dynamic private hosted services - e.g. You have the ability to deploy a "service" and not care about the physical place it is located, you can deploy more in the future, they can be in different Data centres in different countries... it is completely irrelevant to you, it should "just work" without you having to do anything more than you would on a single network at your own office.

Parallel computing is basically the process of splitting a compatible workload among many nodes - e.g. not every task is compatible, but, certain algorithms/operations are (e.g. bitcoin mining).... you could have many machines/objects that are designed to take a workload from a central point and split it between themselves.

Grid computing... I believe interchangeable with parallel.

  • Ok , thanks , and you also want to say that there is only marketing difference between cloud and grid. But i guess that the grids provided by cloud providers inside cloud for grid workloads , are currently not as good as the physical grids and something has to be done to achieve a grid inside a cloud that is as good as a physical grid :) – Ijaz Ahmad Khan Jan 8 '16 at 18:41
  • In my case , a job on 8 virtual cores takes 24 hours while on 8 physical cores , it takes 4 hours. – Ijaz Ahmad Khan Jan 8 '16 at 18:44
  • 1
    @IjazKhan If you have a real problem to solve, you should ask a question about the real problem. The question you've asked here will not help you make progress on that problem. – Michael Hampton Jan 8 '16 at 18:52

Not the answer you're looking for? Browse other questions tagged or ask your own question.