Apple officially launched its much-hyped iCloud suite of services at its Worldwide Developer Conference Monday, and although the capabilities are sure to be the talk of the town among consumers, it’s Apple’s cloud infrastructure that makes it all work. Apple CEO Steve Jobs said as much during his WWDC keynote by closing with an image of — and shout-out to — the company’s new iDataCenter in Maiden, N.C. Details about the technology that will power iCloud have been sparse, but those who’ve been watching it have uncovered some interesting information that sheds some light on what Apple is doing under the covers. Probably the most-interesting data is about the iDataCenter itself. It has garnered so much attention because of its sheer scale, which suggests Apple has very big plans for the future of iCloud. As Rich Miller at Data Center Knowledge has reported over the past couple of years, the iDataCenter: Will cover about 500,000 square feet — about five times the size of the company’s existing Silicon Valley data center. Cost about $1 billion to build, which is about twice what Google and Microsoft generally invest in their cloud data centers. Puts a focus on high availability, including clustering technology from IBM, Veritas and Oracle. Was set to open in spring after delays postponed an October launch. Is only one of two similarly sized data centers planned for the site.
The other big Apple infrastructure news came in April, with reports that the company had ordered 12 petabytes of worth of Isilon file storage from EMC. It hasn’t been confirmed where all that storage will be housed — in the iDataCenter, in Apple’s Newark, Calif. data center or in the new space it has leased in Silicon Valley, or spread among the three facilities — but its mere presence suggests Apple is serious about storing and delivering files of all types. As Steve Jobs noted during the keynote, iCloud is the post-PC-world replacement for syncing everything — photos, audio, documents and more — across all your Apple devices. The company even rewrote the core MobileMe functions as iCloud apps and, much like Google with Google Apps, is giving them away for free. Despite all that storage capacity, though, Apple won’t be housing individual copies of everybody’s media files. Even 12 petabytes would fill up fairly fast with the combined audio of Apple users worldwide, which is why Apple’s focus is still on local storage for iTunes. This way, instead of storing millions of individual copies of Lady Gaga’s “Poker Face” for individual customers, Apple can house minimal copies of each individual song and sync purchased files to devices based on purchased licenses. Even iTunes Match merely applies iTunes licenses to files within users’ personal libraries that weren’t originally purchased via iTunes, rather than uploading each track into the cloud before syncing. This differs from both Amazon’s and Google’s cloud-based music services, which literally store your music in cloud. That could help explain why Apple will charge only $24.99 a year for the iTunes Match service instead of charging customers per gigabyte. This model and the huge storage infrastructure will come in handy, too, should Apple step up its cloud-based video services, which bring even greater capacity issues to the table, as well as those around encoding for delivery to specific device types. (A skeptic might say that Apple’s reliance on local storage is antithetical to the cloud’s overall them of access anywhere (not just on your Apple devices), but that’s a story for another day.) But Apple’s cloud story doesn’t start and stop with iCloud and its related services; in fact, the cloud touches almost every aspect of pretty much every new service and feature discussed during the WWDC keynote. Every time Apple is syncing anything — from application data to system settings to media — it’s touching Apple’s new cloud computing infrastructure. That’s why Jobs highlighted the iDataCenter in his keynote and why Apple recently hired noted cloud data center expert Kevin Timmons from Microsoft. When you’re selling as many different types of devices as Apple does, the real value of the cloud is in syncing data among devices and users, and that requires a robust cloud infrastructure.
Jim Haughwout - July 25, 2011 [we basically disagree with the author - the "feel" of what iCloud is about is far more important than any definition.]
Cloud computing has definitely moved into the mainstream. You now see commercials from Microsoft, Cisco, IBM and others every evening on prime time Cable TV. CNBC has created a Cloud Computing Special Report for investors to learn more about it. Even government agencies are now moving to cloud-based solutions.
Unfortunately,one of the most touted reasons we see for using cloud computing – that it provides universal access to data and applications from the Internet – has nothing to do with what cloud computing actually is. This is simply what web-based applications have been doing since the 1990s. True cloud computing offers a whole lot more. In October 2009, The National Institute of Standards and Technology (NIST) published an excellent definition of cloud computing that calls out five essential characteristics that separate clouds from simple remotely hosted, web-based computing models:
1. On-demand self-service
A consumer can unilaterally provision computing capabilities, such as server time and network storage, as needed automatically without requiring human interaction with each service’s provider
2. Broad network access
Capabilities are available over the network and accessed through standard mechanisms that promote use by heterogeneous thin or thick client platforms (e.g.,mobile phones, laptops, and PDAs).
3. Resource pooling
The provider’s computing resources are pooled to serve multiple consumers using a multi-tenant model, with different physical and virtual resources dynamically assigned and reassigned according to consumer demand. There is a sense of location independence in that the customer generally has no control or knowledge over the exact location of the provided resources but may be able to specify location at a higher level of abstraction (e.g., country, state, or datacenter). Examples of resources include storage, processing, memory, network bandwidth, and virtual machines.
4. Rapid elasticity
Capabilities can be rapidly and elastically provisioned, in some cases automatically, to quickly scale out, and rapidly released to quickly scale in. To the consumer, the capabilities available for provisioning often appear to be unlimited and can be purchased in any quantity at any time.
5. Measured service
Cloud systems automatically control and optimize resource use by leveraging a metering capability1 at some level of abstraction appropriate to the type of service (e.g., storage, processing, bandwidth, and active user accounts). Resource usage can be monitored, controlled, and reported, providing transparency for both the provider and consumer of the utilized service.
I know, some of these terms are mouthful – especially to those who do “live and breathe” technology. However, they remove so much of the work and complexity that has so frequently made management of computing so painful and costly:
On-demand Self-Service (Think “Now”). With on on-demand self-service, you do not need to ask your provider to execute an “IT project” to enable you to use your application (or update it) to support a new business development. You can do whatever you need, when you need it – without the cost and delay of overhead managing your vendor.
Broad Network Access (Think “Convenience”).This lets you work wherever you need, whenever you need – from your work or home computer, netbook, tablet, or smartphone. Traditionally, this was done through browser, to bypass the need to install local software. However, the rise of (cloud-based) App Stores now allows us to install richer applications to access our data – wherever we are, on-demand.
These first two characteristics are what most people think of when talking about cloud computing. However, it is the next three characteristics that make true clouds stand out:
Resource Pooling (Think “Black Box”). Somewhere far away IT people are managing shared, redundant infrastructure across many data centers. They manage maintenance, business continuity, elimination of failures and bottlenecks, etc. You gain all of the benefit of these large-scale investments in time and resources – but without the need to do any work.
Rapid Elasticity (Think “No Limits”). You no longer have to worry about capacity planning. If you suddenly get a surge in traffic (due to an emergency or unexpected popularity) the computing resources you need are automatically – and immediately – available. You avoid slow-downs, timeouts and outages that waste time, cause frustration and turn away customers.
Measured Service (Think “Value”). Pay only for what you use, and no more. Rather than paying 100% for servers that you only use at 20% utilization, you pay for the exact number of resources you use, when you use them. The ideal cloud providers charge usage in terms that everyday people – not just IT systems administrators – understand and value.
When explaining these cloud computing characteristics to those whose “day jobs” are not in tech, I like to use the electricity analogy. When you buy a new television, you do not call the power company and ask them to initiate a project to set up your television. You simply plug it in and begin using it. If you don’t like where it is in your house, you unplug it, move it to a different room, and plug it in again. At the end of the month, you don’t pay for the power company’s generator and labor investments; you pay for the extra kilowatt-hours your television used.
Services that meet all five of these characteristics are much more valuable than those that simply provide access from anywhere. That’s why cloud computing has so much potential to transform the software and IT markets.