January 11, 2011 at 8:11 pm
Comments posted to this topic are about the item The Importance of My Cloud Data
January 12, 2011 at 3:41 am
I've been telling people that for ages, ever since "the cloud" appeared. Same with server virtualisation. When an outage occurs in "the cloud" (or at least a part of it) even if it's nothing to do with "your" system, you're still affected by it and if you're a small company trying to do big things with little funding, who is the supplier going to fix first ?
If you've a virtualised server environment and the hardware fails, you'll no doubt have a more resilliant system than a traditional bank of servers but ultimately if you have to bring that hardware off-line, it's everything off-line. Yes it can fail over to alternative hardware but it could still be a single point of failure. Ever had a catastrophic disc controller crash / hardware failure on a storage array ? It's not necessarily the time it takes to recover, but the loss in business or the ability to provide services when everything is relying on "one" piece of kit rather than separate.
How many of the cloud providers are going to be in business in a few years time and what happens if they get taken over - how easy is it to change from one to another, how easy is it to maintain control of your data.
It's all a trade off in terms of cost, support, resilliance and security.
It's not the be all and end all that some have proclaimed it to be.
January 12, 2011 at 4:02 am
The Sidekick data loss was a classic for this:
January 12, 2011 at 4:34 am
As someone pointed out to me once, the problem with clouds is, they tend to blow away.
January 12, 2011 at 5:32 am
Just before the holidays my son visited from school and at one point went to play a game on one of our PC's - but the game would not play. After a little investigation it turned out that the latest McAfee update "claims" that the game was infected with a virus, a trojan to be exact. The game is about 15 years old and has run for us that long without a problem. No problem I thought, just add the game to McAfee's exclusion list and we would be good to go. NOT!
McAfee now stores its exclusion list on its own servers and users cannot add exclusions locally anymore. In fact, you have to submit the EXE to McAfee and they say it takes about a month to get the EXE into the exclusion list. That was about 6 weeks ago and we are still waiting. Worse, when I contacted McAfee recently, the support tech didn't know how to check if they were even looking at the game!
This, like the cloud, is "progress" apparently. This, like the cloud is the latest "big thing".
Its baffling to me, and has been ever since the cloud concept reared its ugly head - why would people want to store vital data a million miles away on a server where you are now just one of millions, and likely not much of a priority???
The fact that we can "do" something technically is NO measure that we should do something. Indeed, imagine a world where no one stores data locally - you just saw the future and its pretty ugly. When the cloud goes down, the world goes down. And we did this for what good reason??? Because we can. Not that we should.
January 12, 2011 at 6:21 am
Amen... blandry and Steve,
You guys bring to light proof of the MBA Evolution that shows the conversion of the old mantra "the customer being right and important" to Today's mantra "The best value for labor provided". The paradigm shift changes the focus from quality production of goods that result in slow growth and increased value through volumes of loyal customers, to short term gains for stock holders. (When you apply stock to someone's compensation it's only natural that they focus on this.)
I recently spent over three months "Taking a Greater Hand in my Healthcare" because the large insurance institution's developers hadn't taken all customer's situations into account. We deal with a chronic disease (MS) in our family and when my employer eliminated everything but the High Deductible plan I needed to open an HSA. Since I'd had one in the past but closed it out 4 years ago when we switched to my wife's lower cost plan there was no way for me to open it back up. Sounds easy enough but I escalated my calls to the help desk and finally had to tailgate at the insurance company to get escalation to someone who said they thought they could help me. Things with the HSA have finally turned around but I now understand what "Taking a greater Hand in my HealthCare" REALLY MEANS.
Once a company begins to focus on short term gains they group improvements by statistics and not customers. If you're not the million dollar monster in the room it is difficult to have your problem heard or attended to. It is up to each of us as consumers however to keep bringing this to the forefront. That's why here in the States we now have an advocate at the cabinet level. (Though talk about taking the position of Sysiphus.)
An ironic sidenote on my problem. I opened my case with their helpdesk on November 3rd 2010 and heard back from the supervisor they escalated my issue to on 1/10/2011. The information she had to resolve my problem was still wrong.
January 12, 2011 at 6:37 am
Makes me think of our Oracle Service Requests...
After two months of uploading traces I still can't get my problem understood correctly.
What would happen if they lost my data? Would it take 2 weeks to perform a restore?
No, thanks.
-- Gianluca Sartori
January 12, 2011 at 6:44 am
Hotmail. That's a free service, right? People always want to get something for nothing. You get what you pay for. If you get it for free, what kind of support or guarantees can you expect?
Now, if you're paying for a service, that's different. Then, you ahve to be aware of the SLA's and the fine print in the SLA's and the licensing agreements. I know people who own small businesses and they like the fact that they can run their technology free or nearly free. But, they are oblivious to the fact that their services can and will go down and be unavailable when they need it most.
January 12, 2011 at 7:33 am
The cloud, and it's issues, are as old as the mouse. However I have found that my mouse is more reliable than any cloud service I have used.
For example: I got talked into opening and using a Hotmail account at work. Six months later all my Hotmail went from yahoo to M$ and anything older than 90 days was gone.
The cloud tries to convince us that it works like the Internet/Darpanet servers were designed to work. A fully redundant data storage network with colocation that allows access to all information anywhere on the network, and that can continue to operate even if entire sections of the network are lost.
Unfortunately, the only reality our current cloud provides is a fancy name for the smoke being blown by those that would like to sell us something they can not deliver.
January 12, 2011 at 8:09 am
I'm afraid I agree with just about every comment. And for the sake of completeness, I will add that the odds of a security failure are multiplied many times over. I do believe that there is a time and place for almost anything, even cloudish services, but for a business, the problem is the huge number of compounded failure modes.
As most (or all) of us know, at best, these services are sold to businesses by the overenthusiastic, and too often, both sides of the sale are manned by the under-informed. Big firms, of which I am intimate with a few, need to find new products and new pitches, and the new pitch part of that equation is far, far more important than any real "newness" of the product. The cloud falls in this category.
What sensible sounding approaches I have heard regarding "clouds" are those where the principles are pulled into a private data store. At which point one is discussing pretty much the same tools (and concerns) we all have regarding resiliency, redundancy and availability--not a means of ducking responsibility for those three requirements.
I don't care what the contract says, when it comes down to the wire, the cloud purveyor will want to bet that his business can survive the falure of your business--and likely continue to make that same bet until you're gone or you're both over the cliff. The cloud purveyor believes he has to bet on himself--he believes his personal and business survival is more important to him than your survival. Moreover, there is no honor on Wall Street and, consequently, very little in big business. They expect to lose some customers, and that customer they are willing to lose might be you.
January 12, 2011 at 9:31 am
I think that the new database cloud services are useful in that they provide standardized and lower costs platforms for those corporations whose operations or business model already leverage it. However, I do think it's been over-hyped. Recall back in 2000 when the conventional wisdom was that all application development going forward would be web-based and free open source operating systems would replace Windows and Mac.
"Do not seek to follow in the footsteps of the wise. Instead, seek what they sought." - Matsuo Basho
January 12, 2011 at 9:32 am
OCTom (1/12/2011)
Hotmail. That's a free service, right? People always want to get something for nothing. You get what you pay for. If you get it for free, what kind of support or guarantees can you expect?
Tom, There are two service levels with Hotmail, the free version and the annual subscription. As an annual subscriber I do pay for stability and should have some accountability and remedy should there be a problem. These were removed in the EULA however.
January 12, 2011 at 9:50 am
I have a Hotmail account, but I also have a hosted MS Exchange account for professional purposes. The problem with Hotmail, GMail, and other similar services is that it's dominated by spam, multimedia attachments, and all the other pedestrian traffic going back and forth between a billion users. For the same reason, I prefer LinkedIn over FaceBook. In terms of reliability, I don't think that Hotmail and FaceBook are held to the same standard; they serve their purpose in the broader picture, but they're "lower end" services.
"Do not seek to follow in the footsteps of the wise. Instead, seek what they sought." - Matsuo Basho
January 12, 2011 at 10:06 am
Hey Steve
Interesting thought. Made me think about how we would handle availability within our control. Another thought popped as I finished reading your article.
Can you replicate between different vendor clouds? :unsure:
January 12, 2011 at 10:10 am
I just want to point out that Microsoft has indeed posted the cause of the outage. It was not due to SQL Server, or load balancing, but due to an automated test script that accidentally deleted real user accounts instead of only deleting the test accounts that it created. More info on the Windows Live blog:
--
KM
Viewing 15 posts - 1 through 15 (of 45 total)
You must be logged in to reply to this topic. Login to reply