A lire sur: http://www.itworld.com/cloud-computing/277228/real-limits-cloud-computing?source=ITWNLE_nlt_saas_2012-05-16
It's important to see past the hype machine
May 14, 2012, 1:33 PM —
When Google Drive was initially announced to be integrated with
ChromeOS-running machines, the general consensus in the media seemed to
be that this functionality was going to harbor a new age of personal
cloud computing.
I, for one, am not so sure.
It is increasingly apparent that the level of hype around cloud computing has reached new and dizzying levels, and personal and business storage seems to be the latest "hot" implementation through which the Glories of Cloud are proclaimed.
But there are some serious limitations with the notion of cloud computing that seem to be blithely ignored amidst all of the hype.
First, there's the very real issue of bandwidth limitation. If I, a business owner, find the prospect of storing my data cheaply in the cloud and not having to worry about building a local infrastructure (forgetting, for a moment, some of the obvious problems that such an approach implies), then let's go with that a moment, shall we? If you have any sizable data load, you're going to have big problems with upload speed.
Figuring an upstream speed of 10 Mbps (if you're lucky), it would take about 12.5 days to push up a full terabyte of data… and a mere 10 GB would take just under two hours at those speeds. That's not exactly optimal.
Yet services like Box.net, Dropbox, and Google Drive are expecting us to live with just upload bottlenecks. Unless you have the resources to afford a really big pipe to the Internet, cloud storage (at least on the initial upload) is going to take a long time.
And then there's the problem that home internet users are increasingly confronting: data caps. Designed to prevent illicit downloads, as well as slow down the growing trend to "cut the cable" many people are doing to drop traditional cable services in favor of Internet-delivered content on Netflix, Hulu, and Apple TV, many ISPs are putting data caps into place.
You can agree or disagree with their motives all you want (and I, for the record, disagree most strongly), but the real presence of data caps put a serious damper on using the cloud for storage. Granted, business Internet accounts have no such caps, but the premium that must be paid to get fast business-grade access is quite high with most providers.
With storage so relatively cheap, I can set up terabytes of data locally and not have to worry about losing data or filling up for a long time. And what about the benefits of disaster recovery and multi-device access?
There, I will admit, is the real lure of cloud storage. And I do, in fact, use a free Dropbox account to share files across PCs, laptops, and portable devices. My DR strategy is still local, though, because there's too much "important" data to store online.
It may be pedantic to bring up such mundane issues as bandwidth and data capping to knock on cloud storage, but they must not be lost in all of the hype.
I, for one, am not so sure.
It is increasingly apparent that the level of hype around cloud computing has reached new and dizzying levels, and personal and business storage seems to be the latest "hot" implementation through which the Glories of Cloud are proclaimed.
But there are some serious limitations with the notion of cloud computing that seem to be blithely ignored amidst all of the hype.
First, there's the very real issue of bandwidth limitation. If I, a business owner, find the prospect of storing my data cheaply in the cloud and not having to worry about building a local infrastructure (forgetting, for a moment, some of the obvious problems that such an approach implies), then let's go with that a moment, shall we? If you have any sizable data load, you're going to have big problems with upload speed.
Figuring an upstream speed of 10 Mbps (if you're lucky), it would take about 12.5 days to push up a full terabyte of data… and a mere 10 GB would take just under two hours at those speeds. That's not exactly optimal.
Yet services like Box.net, Dropbox, and Google Drive are expecting us to live with just upload bottlenecks. Unless you have the resources to afford a really big pipe to the Internet, cloud storage (at least on the initial upload) is going to take a long time.
And then there's the problem that home internet users are increasingly confronting: data caps. Designed to prevent illicit downloads, as well as slow down the growing trend to "cut the cable" many people are doing to drop traditional cable services in favor of Internet-delivered content on Netflix, Hulu, and Apple TV, many ISPs are putting data caps into place.
You can agree or disagree with their motives all you want (and I, for the record, disagree most strongly), but the real presence of data caps put a serious damper on using the cloud for storage. Granted, business Internet accounts have no such caps, but the premium that must be paid to get fast business-grade access is quite high with most providers.
With storage so relatively cheap, I can set up terabytes of data locally and not have to worry about losing data or filling up for a long time. And what about the benefits of disaster recovery and multi-device access?
There, I will admit, is the real lure of cloud storage. And I do, in fact, use a free Dropbox account to share files across PCs, laptops, and portable devices. My DR strategy is still local, though, because there's too much "important" data to store online.
It may be pedantic to bring up such mundane issues as bandwidth and data capping to knock on cloud storage, but they must not be lost in all of the hype.
Aucun commentaire:
Enregistrer un commentaire