Re: "Cloud Computing" Smears (Was: Google Wave)
> On Mon, 10/19/09, John Williams wrote: > >> Never underestimate the power of human error. As this > >> debacle demonstrates. > > (me, IIRC) Which particular debacle would that be? > I was referring to the Sidekick debacle: > http://www.eweek.com/c/a/Mobile-and-Wireless/Microsoft-Claims-Sidekick-Data-Will-Be-Restored-This-Week-491196/ Whoops! I have a friend who isn't using that service, but somehow lost all his hundreds of contacts from his phone, and they weren't retreivable from his back-up site either. I, OTOH, have all my contacts on paper...(and I'd be smug about it except I don't have hundreds, just dozens, which is quite managable). There _was_ something about a 'hole in security of cloud computing' in a recent MIT Tech update (but I only get the headlines, not the full story, and doubt I'd understand without some major studying, which I just don't have time for at this point...). Debbi Borderline Luddite? Maru :) ___ http://mccmedia.com/mailman/listinfo/brin-l_mccmedia.com
Re: "Cloud Computing" Smears (Was: Google Wave)
On Mon, Oct 19, 2009 at 12:42 PM, Deborah Harrell wrote: >> From: John Williams > > >> Never underestimate the power of human error. As this >> debacle demonstrates. > > Which particular debacle would that be? I was referring to the Sidekick debacle: http://www.eweek.com/c/a/Mobile-and-Wireless/Microsoft-Claims-Sidekick-Data-Will-Be-Restored-This-Week-491196/ ___ http://mccmedia.com/mailman/listinfo/brin-l_mccmedia.com
Re: "Cloud Computing" Smears (Was: Google Wave)
> From: John Williams > Never underestimate the power of human error. As this > debacle demonstrates. Which particular debacle would that be? We gotcher health care, Afghanistan, Eyerak, and balloon boys... Take yer pik! Debbi Whaddya Expect From A Family On "Wife-Swap" Twice? Maru ___ http://mccmedia.com/mailman/listinfo/brin-l_mccmedia.com
RE: "Cloud Computing" Smears (Was: Google Wave)
Julia wrote: > (Jo Anne -- a RAID is a Redundant Array of Independent Disks, where the data > is stored on multiple disks and checked for accuracy on some regular basis. > If one drive goes down, either the data should be duplicated somewhere, or > there should be enough information stored on another disk or disks to > reconstruct what was lost. Off-site backup is still recommended for things > like fire, floods and tornadoes, and don't anyone laugh about the tornadoes, > m'kay?) Thanks, Julia. I figured it was something like that and not a can of bug spray. Amities, Jo Anne evens...@hevanet.com ___ http://mccmedia.com/mailman/listinfo/brin-l_mccmedia.com
Re: "Cloud Computing" Smears (Was: Google Wave)
On Sun, Oct 18, 2009 at 9:42 AM, Bruce Bostwick wrote: > > On Oct 18, 2009, at 12:25 AM, Max Battcher wrote: > >> On 10/18/2009 0:38, John Williams wrote: >>> >>> On Sat, Oct 17, 2009 at 8:36 PM, Julia Thompson wrote: Er. In that sort of a situation, I myself would set up a RAID for storing the data, *much* less chance for losing it. >>> >>> RAID does not protect from rm -rf / , which (some variant of) is my >>> guess at what happened. Although now they are saying most of the data >>> is recovered, so maybe it got munged in a reversible way. >> >> Any "cloud" service at this point is going to be tens, if not hundreds, of >> servers. (Major services easily run in the thousands of servers, and if you >> count "virtual" servers the biggest services are using millions of servers >> already.) At this point any outage that is going to affect a service as >> whole is generally going to be a lot subtler (and possibly a lot "nastier", >> such an accidental viral infection due to an underlying bug/exploit in the >> service) than a rm -rf /. >> >> At least, assuming the system admins are doing their jobs correctly rm -rf >> / to a single server is extremely unlikely to cause massive outage or >> damage... (As a service gets large enough hard drives are expected to fail >> randomly, and surprisingly frequently, and services should be designed >> around that problem...) > > And, as with a RAID except on a much larger scale, there's built in > redundancy and error correction, so the system tends to self-heal. About > the only threat is viral mechanisms that propagate through the system. Never underestimate the power of human error. As this debacle demonstrates. ___ http://mccmedia.com/mailman/listinfo/brin-l_mccmedia.com
Re: "Cloud Computing" Smears (Was: Google Wave)
On Oct 18, 2009, at 12:25 AM, Max Battcher wrote: On 10/18/2009 0:38, John Williams wrote: On Sat, Oct 17, 2009 at 8:36 PM, Julia Thompson wrote: Er. In that sort of a situation, I myself would set up a RAID for storing the data, *much* less chance for losing it. RAID does not protect from rm -rf / , which (some variant of) is my guess at what happened. Although now they are saying most of the data is recovered, so maybe it got munged in a reversible way. Any "cloud" service at this point is going to be tens, if not hundreds, of servers. (Major services easily run in the thousands of servers, and if you count "virtual" servers the biggest services are using millions of servers already.) At this point any outage that is going to affect a service as whole is generally going to be a lot subtler (and possibly a lot "nastier", such an accidental viral infection due to an underlying bug/exploit in the service) than a rm -rf /. At least, assuming the system admins are doing their jobs correctly rm -rf / to a single server is extremely unlikely to cause massive outage or damage... (As a service gets large enough hard drives are expected to fail randomly, and surprisingly frequently, and services should be designed around that problem...) And, as with a RAID except on a much larger scale, there's built in redundancy and error correction, so the system tends to self-heal. About the only threat is viral mechanisms that propagate through the system. I'm just territorial about my data, is all. I tend to like knowing where it's stored and who has access to it, and have some control over its persistence in some cases. There are some applications for which I think cloud storage might serve my needs, and others for which I consider it unsuitable. "Oh yeah? Well, I speak LOOOUD, and I carry a BEEEger stick -- and I use it too!" **whop!** -- Yosemite Sam ___ http://mccmedia.com/mailman/listinfo/brin-l_mccmedia.com
Re: "Cloud Computing" Smears (Was: Google Wave)
On 10/18/2009 0:38, John Williams wrote: On Sat, Oct 17, 2009 at 8:36 PM, Julia Thompson wrote: Er. In that sort of a situation, I myself would set up a RAID for storing the data, *much* less chance for losing it. RAID does not protect from rm -rf / , which (some variant of) is my guess at what happened. Although now they are saying most of the data is recovered, so maybe it got munged in a reversible way. Any "cloud" service at this point is going to be tens, if not hundreds, of servers. (Major services easily run in the thousands of servers, and if you count "virtual" servers the biggest services are using millions of servers already.) At this point any outage that is going to affect a service as whole is generally going to be a lot subtler (and possibly a lot "nastier", such an accidental viral infection due to an underlying bug/exploit in the service) than a rm -rf /. At least, assuming the system admins are doing their jobs correctly rm -rf / to a single server is extremely unlikely to cause massive outage or damage... (As a service gets large enough hard drives are expected to fail randomly, and surprisingly frequently, and services should be designed around that problem...) -- --Max Battcher-- http://worldmaker.net ___ http://mccmedia.com/mailman/listinfo/brin-l_mccmedia.com
Re: "Cloud Computing" Smears (Was: Google Wave)
On Sat, Oct 17, 2009 at 8:36 PM, Julia Thompson wrote: > Er. In that sort of a situation, I myself would set up a RAID for storing > the data, *much* less chance for losing it. RAID does not protect from rm -rf / , which (some variant of) is my guess at what happened. Although now they are saying most of the data is recovered, so maybe it got munged in a reversible way. ___ http://mccmedia.com/mailman/listinfo/brin-l_mccmedia.com
RE: "Cloud Computing" Smears (Was: Google Wave)
Er. In that sort of a situation, I myself would set up a RAID for storing the data, *much* less chance for losing it. I'd just do that anyway. In fact, the computer that's still in a box and is destined to replace the one I'm using right now has a RAID, because I seem to have a knack for catastrophically losing hard drives that baffles my husband entirely. (He has more problems with his PDAs than I do, so I guess there's *some* sort of balance) I think I've lost 2 or 3 in the past 6 years, and any data that wasn't backed up, which is kind of rough for an information junkie. For *that* sort of application, I'd go with a decent number of disks in the array for any one set of data. My own problem with cloud computing is, if the magical set of wires between me and my data has a glitch, I can't get to my data, and we end up with Grumpy Julia, which is not pleasant for anyone directly involved. (Jo Anne -- a RAID is a Redundant Array of Independent Disks, where the data is stored on multiple disks and checked for accuracy on some regular basis. If one drive goes down, either the data should be duplicated somewhere, or there should be enough information stored on another disk or disks to reconstruct what was lost. Off-site backup is still recommended for things like fire, floods and tornadoes, and don't anyone laugh about the tornadoes, m'kay?) Julia ___ http://mccmedia.com/mailman/listinfo/brin-l_mccmedia.com
"Cloud Computing" Smears (Was: Google Wave)
Bruce Bostwick wrote: Part of my concern with the concept in general is the fairly glaring admin/management deficiency described in this article: http://dailyqi.com/?p=10576 I've been avoiding most articles on this subject because there is a lot of FUD out there and very little real truth. Some of that is T-Mobile blamestorming and some of that is the usual sorts that like any opportunity to smear Microsoft. It's very easy to focus on the negative case studies in "cloud computing" and miss the 99.9% of the time when stuff works as it is supposed to. No one buys digital ads for articles about "the status quo works, go back to sleep". Danger was acquired in February of this year and I wouldn't be surprised that the majority of the infrastructure in question predated the acquisition. Even big companies like Microsoft can't magically change infrastructure with the snap of a finger... Furthermore, to my knowledge, Microsoft/Danger have been explicitly mum about what precisely the technological glitches were that lead to the failures. It's certainly easy to presume that "there were no backups at all", but at this point it is still hearsay, at best, and my money is on slander. I've heard that some of the affected customers have already started to get some of their personal data back and the press release from Microsoft declares that they are "confident" that they will restore the majority of it, which seems to contradict the "no backups at all" theory pretty well. (I doubt that they would remain "confident" if they were combing disks in clean rooms for good sectors...) Certainly Microsoft isn't entirely blameless, you would assume a technical audit would be an early priority in any acquisition. Presumably stability issues would be a huge priority and reliability engineers would be some of the first gated into a acquisition project. More particularly, I think that T-Mobile isn't nearly as blameless as they would like to believe or portray themselves as. Getting back to that "it's who you ask the questions" of problem, T-Mobile was the first call in that chain (their name is branded on the product!) and if their answer at any point was "we don't know about our service's reliability" or "our service is absolutely reliable" without connection to reality (and without in turn encouraging customers to talk to Danger if they wanted deeper answers), then they are absolutely a part of the blame and a part of the problem. All of which isn't to say that your fears, Barry, are unwarranted or that caution doesn't apply. More that I think that journalists (and almost especially "tech" journalists) seem to be having a harder and harder time reflecting technical reality and I think there is a need for some mechanism to break the tedious Hype then Fear/FUD/Doom/Gloom cycle. To me this is exactly the sort of story that breaks that doesn't get a healthy grain of salt... -- --Max Battcher-- http://worldmaker.ne ___ http://mccmedia.com/mailman/listinfo/brin-l_mccmedia.com