Thứ Sáu, 16 tháng 8, 2013

Clearing the fog: The future of cloud computing - Talk Mobile | Android Central

by Rene Ritchie, Daniel Rubino, Kevin Michaluk, Phil Nickinson

There are many visions for the future of computing, and by-and-large they all involve some form of distributed storage and processing. Sure, our computers, tablets, and smartphones will continue to evolve and become more powerful, but so too will cloud services and the internet infrastructure that connects it all.

Whether we're looking at connected headset displays, constantly-syncing smartphones, or present-anywhere virtual machines, the cloud is going to play a significant role in computing in the years ahead.

The future of the cloud isn't going to be without troubles, however. What will it take to get our data everywhere? Are our networks and devices and the services themselves up to the challenge of syncing gigabytes upon gigabytes of data?

If we can have all our data everywhere, do we even need a powerful computer anymore, or can we offload it all onto the cloud and just have a thin client for interaction? And what of our media - is it ours, and will we ever be able to get it our way instead of Hollywoods?

Let's get the conversation started!

22
loading...
6
loading...
27
loading...
0
loading...

If it weren't for decent bandwidth, we wouldn't even be talking about the cloud. Can you imagine trying to sync Dropbox over a US Robotics 14.4k modem? Or even downloading this page? We couldn't have the modern cloud without modern networks.

A decade ago, cloud services weren't a thing that we interacted with on a daily basis. "The cloud" wasn't even a term, though it did exist - its role was making data more efficient. Case in point: the BlackBerry Network Operations Center. The NOC managed traffic between device and server by streamlining and compressing data that was sent to devices on the BlackBerry network.

The NOC was necessary then, back when you were lucky if your phone was capable of 100Kbps downloads, and even luckier if your carrier ever came close to providing that. Today we live in a world of LTE that can hit 50 megabits down, or even 75 or 100. The speeds are absurd, rivaling what's available for most home wired connections.

Of course, that's if you have a good LTE connection, which not everybody does. 4G connectivity is spreading, but rural and poor areas will lag behind for the foreseeable future.

Today, for the most part, cloud services are pretty good, though there's always room for improvement. Sometimes it's hard to tell where the improvement needs to happen. Is it the service itself that's faltering, my local internet connection, or something in-between? I don't always know who to blame when a show streaming off Netflix has to downgrade its quality or stop to buffer, or when the music I'm streaming stutters and cuts out.

Is it the service itself that's faltering, my local internet connection, or something in-between?

That said, there's nothing wrong with throwing more bandwidth and more servers at the cloud. A lot of the lag problems in the early years of Android have been solved today by throwing increasingly powerful hardware at it until it could keep up; the same can be done for cloud services. Hell, that's what Twitter had to do when the service took off: more servers, more bandwidth, problem solved.

Bandwidth isn't holding back adoption of cloud services. It's being comfortable with the concept of the cloud, or understanding just what the cloud offers, that's stopping people from jumping on board. But even then we've seen a massive uptake in the past few years - thanks in no small part to a massive uptick in bandwidth. The cloud has momentum now - and clouds aren't easy to stop.

To me, that's what the future is. You have your data, you have your identity, but the fashion will change all the time.

- Dalton Caldwell / Founder and CEO, App.net

“Computer!” It’s become a cliché. What Scotty said in his brogue, Spock in his monotone, or Kirk in his swagger. What commanded the attention of a vast computational core, and caused it to come squawking out of any terminal, anywhere on the Enterprise. It was the thinnest of clients. And it was a concept that everyone from Sun Microsystems to Oracle to Google would embrace and try to extend for decades.

It’s a compelling idea. Giant computers, hidden away, surfaced anywhere a light, convenient interface can be found. It’s the dream of every IT administrator who’d rather work from poolside than from a desk. It’s the hope of every writer and student who wants to carry the barest minimum hardware still maintain constant connectivity.

It’s a future that’s always coming but never quite arriving.

And like HTML5 apps, it’s a future that’s always coming but never quite arriving, and for a similar reason: the web just isn’t good for everything, and will never be as good for everything.

Take Google’s Chrome OS for example. The Chromebook Pixel is probably the most beautiful thin client ever conceived. And one of the stupidest. It’s a powerful machine that costs as much as the Adobe suite, yet can’t run Photoshop. It has a retina display that won’t display Premiere or After Effects. It’s utterly dependent on the internet in an age when the internet is still utterly undependable.

There are, should be, and increasingly will be software and services that make sense to be cloud-based. Not just things like messaging or browsing, which are born of the web, but anything that benefits from collaboration or cooperation. Other than that, nope, sorry. Code it natively so I can access all the power and performance native enables.

Rather than a thin client, what seems to be evolving is a client with a nice physique. Not emaciated, not, it’s a right-sized machine backed up by an amazing cloud.

Chrome OS is ahead of its time, and hopefully Google will someday meld its strengths with the native strengths of Android. Dropbox, having already bought Mailbox, may one day field their own cloud-based operating system. So might Facebook, who’s already got a bunch of apps and strong cloud infrastructure. And Amazon has server-sided browser in Silk, who knows what’s next?

The clients that run any or all of those wouldn’t be thin either, they’d just be remarkably fit, and that’s what we want from the future.

The road to having all of your information everywhere through the cloud is a long and tedious one, but like all things that deal with computing, it’s moving faster than ever. The problem is multifaceted: there are security concerns, availability of storage, bandwidth, cost, file integrity, and incorporating these services into our current systems and behaviors.

There are two paths that can be taken: third-party clouds like Dropbox or SugarSync, and first-party clouds like those created by Apple, Google, and Microsoft. To be frank, the easiest way to get to the all-data-everywhere state will be buying into one ecosystem and its first-party cloud. Dropbox, Box, and all the rest of the third-party gang offer loads of flexibility, but it's tough to beat the integrated nature of the first-party cloud.

Hop on with Microsoft and Windows 8.1 will back up nearly everything on your computer to SkyDrive, and you'll be able to access all of that from a Windows Phone. Use a Mac? Get an iPhone and iCloud will seamlessly sync your photos, iWork documents, music, calendar, and more across your whole iEcosystem.

But integrated first-party clouds lock you into that ecosystem. Once you have a Windows 8 laptop paired through SkyDrive with an Xbox One and your Windows Phone, it's hard to justify the switching cost to another platform. There's a SkyDrive app for Android, but is it going to be as integrated of an experience as on Windows Phone? Will a BlackBerry 10 smartphone work well in a household that otherwise runs on Google services?

Integrated first-party clouds lock you into that ecosystem.

Competition is what's driven us to have multiple operating systems, and it's what's driving us to a future with clouds dedicated to those operating systems. It's in the best interests of any manufacturer or developer to make the cost of switching too high, either by loss of features or simply making it inconvenient for customers. It's not great for consumers, but that's just the way it is and will continue to be.

The day that all of your stuff is everywhere is nigh. It is certainly where all of these companies are moving towards today and tomorrow, with vast resources being devoted to the endeavor. It will take time, because for these services to be truly everywhere will require platform updates at a core level. And you'll probably only be able to reap the benefits by buying entirely into the platform ecosystem.

Make no mistake, it’s the folks in the mansions in California who control what you watch on your TV, your tablet and your phone. It’s not Apple. It’s not Google. It’s not Microsoft. It’s (mostly) not Netflix.

We’ve all experienced the frustration of searching Netflix for the latest movie and finding nothing. You still can’t subscribe to online-only versions of the premium cable channels. For as great as HBO Go and the like are, you still have to have a cable subscription. You, dear user, are at their mercy. And if Hollywood has anything to say about it, that’s how things will remain.

If Hollywood has anything to say about it, that’s how things will remain.

There is hope. In 2013 we saw the first series produced by Netflix hit the internet. Complete seasons released all at once, for you to watch at your leisure. Binge on a dozen episodes in a single day, or pace yourself. Doesn’t matter to Netflix.

With a much lower profile (and far fewer dollars at stake) are any number of independent producers that put out their own content on YouTube, Vimeo, or other platforms. They should be commended, and we should all support them. But the simple fact is that they’re swimming upstream, and Hollywood is the bear waiting to bite the head off any salmon unlucky enough to poke its head out of the water.

In a perfect world, all video would be available anywhere, at any time. YouTube has made publishing and promoting online video easier than ever. But YouTube (or any other service) usually doesn’t own that content. And so we’re back at the same formula we started with, with those of us staring at the screen at the dead end of a long, dark road.

There are a lot of little, independent lights working to illuminate that road. But it’s a long road to light, and they’ve got a lot of work ahead of them.

Those of us with young children are raising boys and girls who will never know a world without Netflix. Without YouTube. Without the means to watch any video online, day or night. And, one day, hopefully, they’ll know a world in which the studios give us what we want -- their content, whenever and however we want it.

We've been saying it all along: the cloud is the future of computing. But it's not the be-all-end-all of computing. While there's no doubt that the cloud will be more and more involved in our daily computing needs, especially as mobile networks grow stronger, faster, and larger, it's not going to replace our devices as we know them.

Our laptops, smartphones, tablets, and whatever other new forms of personal computing hardware arise in the coming years will continue to become more powerful and more personal, and the cloud is only going to help in that. If it's serving up additional processing power when it's needed, having all of our content on tap at a moments notice, or simply getting us our emails, the cloud will be there.

But the cloud is going to need better infrastructure if it's going to be with us in new and omnipresent ways. More servers with better redundancy, greater bandwidth, and broader connectivity will all play a part in expanding the reach of the cloud.

We're only in the early days of the cloud. What does the future hold?


Source : feedproxy[dot]google[dot]com

Không có nhận xét nào:

Đăng nhận xét