Author Archives: pcalcada

Stay tunned…

Posted on by
Reply

For the past two months, although it has been very difficult to continue the writing, the CloudViews.Org has not stopped its work. As we promised in our ending notes on the CloudViews.Org Cloud Computing Conference 2009 the main subject for 2010 will be Interoperability, but before we start thinking on the layout of the new event we have to clarify some organizational issues. We want to use the base layout of 2009′s event but we have to articulate it with other projects that we are planning. All the projects will have as main goal to promote and help moving forward the Cloud Computing paradigm in an open and free way, but as I think all of you are able to imagine that is something not very easy to achieve.

On the writing side of CloudVIews.Org we already have some posts which are almost finished and ready to publish, but the subjects are so many and with the ideas flowing on the same path it’s becoming very hard to select the ones to finish first. If you want to continue your readings I leave here some of the sites and subjects that I’ve been occupied with and that have captured my attention:

So please stay tuned and as an appetizer here are some titles of next posts (not necessarily in this order):

  • Moving forward on Cloud Computing – Say bye bye to your LAN.
  • Entering the era of relations.
  • Cloud Security – Are we really moving forward?
  • Identity 2.0 – a simple (completed) mashup.

Is the Cloud as the vendors are claiming?

Posted on by
Reply

IT news from Australia is presenting a work made by the UNSW School of Computer Science. In this study they try to verify if the Cloud is as the Vendors are claiming that it is: elastic, reliable, dynamic, fault tolerant, high available..

http://www.itnews.com.au/News/153451,stress-tests-rain-on-amazons-cloud.aspx

These are very interesting points and independent tests are fundamental to create trust among Cloud users and providers.

Although I think this kind of test should be more focused on specific applications. As  studies show, some of the potential problems found on the Cloud could be solved by the developers.

Using these ideas as starting point I’m working on a project that should be presented as a parallel event on the CloudViews.Org Cloud Computing Conference – 2010. This project, and the whole CloudViews.Org Cloud Computing Conference 2010 will be presented very soon, but If any one is interested in these topics, feel free to contact me directly.

Amazon introduces the AWS Private Cloud, and the question that arises is if this new service enables the deployment of real Private Cloud..

Posted on by
Reply

Amazon has introduced today the new AWS Virtual Private Cloud service, more details on it could be found here: http://aws.amazon.com/vpc

On the follow up of this new service, some doubts have arisen about the real type of the cloud built with this new service, and also if this kind of cloud could even be difined as Private Clouds.

Using the NIST definition:

” Private cloud. The cloud infrastructure is operated solely for an organization. It may be managed by the organization or a third party and may exist on premise or off premise. ”

It’s very clear, a Private Cloud could be on premise or off premise. The next question is what should be considered the premises of the cloud. Should it be the wall of the datacenter, the box of the server or the VM where the OS is running? For one company their premises could be a single room, whereas to another it could be the entire building, in the cloud computing paradigm the private cloud is, in my perspective, a sandbox, which has boundaries well definite and on which we trust. These boundaries are the set of features that are available (CPU, network, storage, development API, etc) and the trust comes by means of contracts, cloud provider reputation, etc…

It’s easy to understand that this kind of definition are not closed to discussion, but in my prespective, the service introduced by Amazon should be consider a very interesting tool to build effective Private Clouds.

Google Chrome OS – more thoughts about it

Posted on by
1

I have to admit that this type of short post could transmit the idea of a follow-the-hype initiative, but that was not the intention. So following some feedback on the Google Groups I’ve decided to clarify some of my initial thoughts.

First, I have to clarify on what part of Cloud Computing I was talking about. As I’ve written here (http://www.cloudviews.org/2008/12/cloud-computing-the-internet-becomes-our-computer/) the SaaS layer is the one, in my perspective, that could have more relevance to the user’s daily work or in the user’s perspective of the Cloud. And it’s precisely there where I think that products such as the Google Chrome OS and the Palm WEB-OS could have a strong impact. Not because they are thin clients to access the Cloud but because they will put the cloud working for users.

The perfect sandbox or ecosystem for SaaS applications is the web browser, but because its development is so dependent on the development of the underlying OS, I think that having both developed in an integrated way could be a major step forward.

To be able to have a perfect Could OS we still need to develop simple things as:

Those are simple examples that I think could be easily developed if they are built in a integrated way.

It’s obvious that in a commercial perspective, and being Google one of the top developers on the SaaS market, to have a complete Could ecosystem is something that they would love to have, and it’s precisely here that they should be careful and try not to close this platform and make it only available to their products. Taking the VMware (Virtualization) example, don’t forget that we are now trying to discussed which Virtual Machine Open Format is more suitable to prevent the vendor lock-in problem.