Cloud computing has started gaining traction among software product and service providers. Regardless, defining the cloud is still a difficult task, since the concept is nebulous and can be defined both in terms of technology infrastructure and business solutions.
Rather than articulating a single statement that attempts to outline the myriad perspectives inherent to the Cloud, the definition can be elucidated by answering a few questions:
Really, what is it?
Simply stated: the cloud provides “computing resources on demand.” For example, if you need a computer to host your personal blog, you can fill out a form at a cloud-service provider website and they will give you one. You’re not really buying or renting a particular computer or piece of hardware; you are provisioning what is called a “virtual machine.” Once you’ve acquired this virtual machine, you are charged according to how often it is used. Amazon’s billing model works like a utility: just as your electricity company charges you every month for the amount of electricity you use, Amazon charges you for how often your computing resources are used (interestingly, I’ve heard of people with particularly unpopular websites getting charged $.92 on their monthly Amazon bill). Google and Microsoft give you a certain amount of usage for free before they start charging for use.
Who provides it?
Amazon provides web servers through its Elastic Compute Cloud (EC2) service and disk space for file storage through their Simple Storage Service (S3). Google and Microsoft are taking a different approach. With Google’s App Engine and Microsoft’s Azure, you don’t get direct control over virtual machines. Rather, you are provided a platform where you can upload code, images, html, and other assets that define a web application. Microsoft and Google automatically acquire the virtual computing resources for you and will install your application on them.
All of these services can be managed through web-based interfaces that allow you to upload your applications and monitor virtual computer usage. Additionally, companies that provide physical machines and bandwidth for lease, such as RackSpace, are also starting to get into the cloud-based platform business.
What are the advantages?
It’s cheap. Google gives you enough free bandwidth for 5 million page-views a month before they start charging for use. Microsoft hasn’t yet provided pricing details for their service, but it is expected to resemble Google’s pricing model.
Cloud-based platforms are also extremely scalable; if your website traffic goes from two page views on one day to three million the next, all of the virtual machines required to handle the extra load are automatically acquired, without any interaction required on the part of anyone’s IT department.
What are the risks?
Like all things internet-related, cloud services can fail. If your cloud provider does fail and your website goes down, you probably won’t find anyone you can reach on the phone for support. Theoretically, cloud service providers build enough redundancy into their infrastructure to make failures and down-time obsolete; in practice, this has not proven to always be the case.
Microsoft and Google currently limit the capabilities of programs installed in the cloud (for example, you can’t access the file system); thus developers must conform to the frameworks provided by these services. Almost any type of web application can be accommodated, but anything that is particularly specialized requires a dedicated physical machine.
Security concerns are currently a huge barrier to entry for many businesses. It is all but impossible to guarantee where sensitive data will be located; it could be spread out across massive data centers around the country. Due to the lack of control over the geographical location of the machines that store data, it is impossible to apply protocols for storing sensitive information (such as industry-standard PCI, SAS70 and ISO9000 compliance requirements). There is also a risk that some of the smaller cloud-service providers might go out of business or be acquired; what, then, happens to all the valuable information they stored?
Prepositions: things you can do to a cloud
Certain phrases are starting to arise that refer to things that are “in the Cloud” or things that come “from the Cloud.” These statements allude to alternate meanings of Cloud Computing. Via our Gmail accounts or our FaceBook profiles (or Flickr streams, etc.), there is much information about us being stored in the ethereal internet infrastructure of Cloud Computing. When referring to these particular forms of information, people will sometimes refer to “the Cloud.” They are substituting the qualities of the information being stored for the more accurate depiction of the way it is being stored. Having said that, “the Cloud” is now coming to be defined as a personalized version of the internet where people can store their videos, photos, thoughts, etc. and also peruse the same variety of information associated with others.
What’s the future of Cloud Computing?
At the risk of over-using an already questionable pun: “the sky’s the limit.” Cloud computing already provides the advantage of cheap or free web hosting. In the future, mobile and desktop applications will start to leverage the cloud to provide ubiquitous access to our files and information. For example, imagine creating a Word Document in Microsoft Word on your computer, saving your changes, and then opening up the very same document on an entirely different computer later, without having to transfer the file between machines. The document can be stored in the cloud, tied to your login (also stored in the cloud), and then instantly accessed through a desktop application, a mobile device, or a web-browser.
This type of instantly-accessible and globally persistent data storage will change the way we think about information. No longer will the things we create or consume on a computer be tied to a particular piece of hardware; they will be available at all times from anywhere.
Defining the Cloud
Cloud computing has started gaining traction among software product and service providers. Regardless, defining the cloud is still a difficult task, since the concept is nebulous and can be defined both in terms of technology infrastructure and business solutions.
Rather than articulating a single statement that attempts to outline the myriad perspectives inherent to the Cloud, the definition can be elucidated by answering a few questions:
Really, what is it?
Simply stated: the cloud provides “computing resources on demand.” For example, if you need a computer to host your personal blog, you can fill out a form at a cloud-service provider website and they will give you one. You’re not really buying or renting a particular computer or piece of hardware; you are provisioning what is called a “virtual machine.” Once you’ve acquired this virtual machine, you are charged according to how often it is used. Amazon’s billing model works like a utility: just as your electricity company charges you every month for the amount of electricity you use, Amazon charges you for how often your computing resources are used (interestingly, I’ve heard of people with particularly unpopular websites getting charged $.92 on their monthly Amazon bill). Google and Microsoft give you a certain amount of usage for free before they start charging for use.
Who provides it?
Amazon provides web servers through its Elastic Compute Cloud (EC2) service and disk space for file storage through their Simple Storage Service (S3). Google and Microsoft are taking a different approach. With Google’s App Engine and Microsoft’s Azure, you don’t get direct control over virtual machines. Rather, you are provided a platform where you can upload code, images, html, and other assets that define a web application. Microsoft and Google automatically acquire the virtual computing resources for you and will install your application on them.
All of these services can be managed through web-based interfaces that allow you to upload your applications and monitor virtual computer usage. Additionally, companies that provide physical machines and bandwidth for lease, such as RackSpace, are also starting to get into the cloud-based platform business.
What are the advantages?
It’s cheap. Google gives you enough free bandwidth for 5 million page-views a month before they start charging for use. Microsoft hasn’t yet provided pricing details for their service, but it is expected to resemble Google’s pricing model.
Cloud-based platforms are also extremely scalable; if your website traffic goes from two page views on one day to three million the next, all of the virtual machines required to handle the extra load are automatically acquired, without any interaction required on the part of anyone’s IT department.
What are the risks?
Like all things internet-related, cloud services can fail. If your cloud provider does fail and your website goes down, you probably won’t find anyone you can reach on the phone for support. Theoretically, cloud service providers build enough redundancy into their infrastructure to make failures and down-time obsolete; in practice, this has not proven to always be the case.
Microsoft and Google currently limit the capabilities of programs installed in the cloud (for example, you can’t access the file system); thus developers must conform to the frameworks provided by these services. Almost any type of web application can be accommodated, but anything that is particularly specialized requires a dedicated physical machine.
Security concerns are currently a huge barrier to entry for many businesses. It is all but impossible to guarantee where sensitive data will be located; it could be spread out across massive data centers around the country. Due to the lack of control over the geographical location of the machines that store data, it is impossible to apply protocols for storing sensitive information (such as industry-standard PCI, SAS70 and ISO9000 compliance requirements). There is also a risk that some of the smaller cloud-service providers might go out of business or be acquired; what, then, happens to all the valuable information they stored?
Prepositions: things you can do to a cloud
Certain phrases are starting to arise that refer to things that are “in the Cloud” or things that come “from the Cloud.” These statements allude to alternate meanings of Cloud Computing. Via our Gmail accounts or our FaceBook profiles (or Flickr streams, etc.), there is much information about us being stored in the ethereal internet infrastructure of Cloud Computing. When referring to these particular forms of information, people will sometimes refer to “the Cloud.” They are substituting the qualities of the information being stored for the more accurate depiction of the way it is being stored. Having said that, “the Cloud” is now coming to be defined as a personalized version of the internet where people can store their videos, photos, thoughts, etc. and also peruse the same variety of information associated with others.
What’s the future of Cloud Computing?
At the risk of over-using an already questionable pun: “the sky’s the limit.” Cloud computing already provides the advantage of cheap or free web hosting. In the future, mobile and desktop applications will start to leverage the cloud to provide ubiquitous access to our files and information. For example, imagine creating a Word Document in Microsoft Word on your computer, saving your changes, and then opening up the very same document on an entirely different computer later, without having to transfer the file between machines. The document can be stored in the cloud, tied to your login (also stored in the cloud), and then instantly accessed through a desktop application, a mobile device, or a web-browser.
This type of instantly-accessible and globally persistent data storage will change the way we think about information. No longer will the things we create or consume on a computer be tied to a particular piece of hardware; they will be available at all times from anywhere.