--
Jerry Feldman <g...@blu.org>
Boston Linux and Unix
PGP key id:3BC1EB90
PGP Key fingerprint: 49E2 C52A FC5A A31F 8D66 C0AF 7CEA 30FC 3BC1 EB90
In the meantime, take a look at Cleversafe(.com)
--dan
_______________________________________________
Discuss mailing list
Dis...@blu.org
http://lists.blu.org/mailman/listinfo/discuss
I understand the value of backups, and I understand the value of
virtual computing, and parallel computing, and other things. Could
someone (and I apologize if this is OT to the thread) explain to me
what exactly the buzzword "cloud computing" means, exactly? My BS
detector flashes every time someone says it, but I have been wrong
once (okay, *maybe* twice) in the past. Is it just a way of saying
that you have a distributed, parallel app whose individual nodes can
come online (or go offline) dynamically without interrupting the
service? Is it strictly used to reference data storage - as in the
original email in this thread - or does it mean more than that?
Thanks,
Dan
On 01/20/2012 04:01 PM, Daniel C. wrote:
> On Fri, Jan 20, 2012 at 2:35 PM, Jerry Feldman <g...@blu.org> wrote:
>> Theoretically, a cloud is a virtual storage device where the actual storage
>> media should be in several different locations fully mirrored.
> I understand the value of backups, and I understand the value of
> virtual computing, and parallel computing, and other things. Could
> someone (and I apologize if this is OT to the thread) explain to me
> what exactly the buzzword "cloud computing" means, exactly? My BS
> detector flashes every time someone says it, but I have been wrong
> once (okay, *maybe* twice) in the past. Is it just a way of saying
> that you have a distributed, parallel app whose individual nodes can
> come online (or go offline) dynamically without interrupting the
> service? Is it strictly used to reference data storage - as in the
> original email in this thread - or does it mean more than that?
>
Maybe I'm late to the IT party, but how do you mean that it's going
back to the way it used to be? I've never really experienced (that I
can remember) a work environment where most of the data (the important
stuff, at least) isn't housed in some central repository - usually a
web or standalone DB server. Am I misunderstanding what you mean, or
have I just not been paying attention to how things work where I've
worked?
-Dan
I see "cloud" as inherently implying leasing/renting access to other
peoples' equipment. In some sense, it is an extension of
outsourcing. Except you may only outsource pieces of a long chain of
steps rather then the entire system and there is an implication that
you can adjust your outsourcing on the fly. In the mainframe days,
there were certainly service bureaus; but I don't think the
granularity of the outsourcing was a fine back then as is being
discussed now. I think "cloud computing" and "software as a service
(SaaS)" are related as well. Particularly if you think in terms of
different pieces of middleware being done as SaaS. I've seen people
refer to "private clouds" where you use many of the same technologies
to quickly reconfigure systems; but you actually own (or at least
long-term lease access to) the equipment.
Bill Bogstad
After having done backups for years, this kind of thing is a concern of mine.
Just using a 'cloud' data center doesn't mean it is duplicated over wide
area. Amazon's outage of their East Coast US data center access last
year showed the hole in many of their clients thoughts. To get geographic
diversity, one must ACTIVELY put data in various data centers, and must
code to ensure wide area replication.
It makes sense to me that they might provide a service that provides
'automatic geographic diversity' but there isn't one that I know about. They do
however, provide information about how to do this.
Do various services use this? You must ask them and then we 'believe'
their answers. Not much of a way to 'trust but verify' like the big customers
could do.
If you are 'big enough' to run your own data centers, this has been an
issue 'forever'. EMC, IBM, and other vendors sell SAN and other software,
and options in their backup systems, database systems, etc that would allow
this, but you must still provide a secure 'network' (either private, VPN,
or secure tunnels) to ensure secure connectivity and pay for 'sufficient
bandwidth' with reliability to be able to support the service.
Other cloud services may provide 'automated geographic diversity' but
as far as I know Eucalyptus and Amazon don't (they share an almost
identical API - they are co-developers of the API they use).
After working for 'big companies' for many years, disaster recovery and
business continuation assurance is a 'big deal'. One major oil, while I
was there, had multiple geographically diverse data centers. To keep
from having 'enough idle capacity' laying around to back themselves up,
they purchased cold data center capacity from a major DR company.
At the time, we still made backup tapes, shipped them offsite to a
separate company facility (I was told they owned an old salt mine somewhere
in Kansas where we shipped tapes from all data centers, but I never
saw it or knew where it was exactly. We shipped out containers of
9 track tapes or cartridges, and got 'old' containers back.)
When doing a 'disaster test', we had a 'disaster date' in the past, we had
containers shipped to our 3rd party recovery site (the DR company's data
center), and went there. We could restore, test, be audited, and
clean up (write
erasing data patterns multiple times to all disk drives touched), and have
the tapes ready to ship back to the 'salt mine' in 48 hours.
It took several times to ensure the 'system worked'. Some data centers never
had a good 'test'. The one where I was took 3 times before we had a 'fully
successful' test. -- All that being said, doing good disaster recovery is hard.
We got some stupid T-shirts that said 'the only thing worse than Disaster
Recovery is no Disaster Recovery'. That has a lot of truth in it.
One time I had bosses whine about the cost involved. I suggested we
just don't do it. But first, go get at least 3 bids from large insurers to find
out the cost of an insurance policy to cover the business risk involved.
If the policy costs less than doing DR and business continuity planning,
testing, including the overhead costs of equipment, service, and people.
The lower cost option should win. They also need to be able to explain
their decision to auditors and shareholders. ... I never heard back from
that suggestion. But it was heart felt, not tong in cheek. ... sorry
a real soap box that I carry around in my baggage.
All in all, cloud computing decision is similar to the DR decision above.
Many companies are choosing 'cloud vendors' without understanding,
or choosing actively or passively deciding to ignore, the value and liability
options related to it. They just see the apparent 'cost cutting' portion.
All that being said, I am not against cloud computing. Just need to
understand what it really means before betting the company on it.
Now, to your question about 'what happens'. IMHO, the data center
will probably be dismantled, and the data will, if they are good, be
discarded by over writing or better low level re-formatting, drives.
Worst case, a new owner would get it, and will probably format it so they
can put their data on it instead. They are 'probably' not interested in
your data. If you are sufficiently paranoid (like we all 'should' be),
keeping data encrypted is a good idea, IMHO. Most cloud based
web sites have no need. You are putting your data on the web without
encryption, so why encrypt? (A rhetorical question, there are good
reasons both pro and con).
Oh yes, one of the 'very large' companies I worked for had a 7 year
cycle of 'centralization' vs 'diversification'. That was true even back
in the mainframe days. They had 2 major data centers in the 'centralized'
times, and about 13 in 'decentralized' times. Depends on whether
'cutting cost' or 'responsive to customer departments' was the focus.
Better networking generated a major 'centralization' especially
when 'pc's or 'workstations' put more computing power on the desktops.
Sorry for the overly long response.
><> ... Jack
Whatever you do, work at it with all your heart... Colossians 3:23
In the old-old days, only the largest institutions could afford computers. Smaller organizations leased time on their iron. That's just how things were done.
And that's what you do today when you buy time on Amazon's or Google's data centers: you lease time on their iron. The nature of the iron has changed. It's now a huge cluster of relatively small units instead of a room-sized monolith but the use model is the same and just as importantly the reason is the same.
--Rich P.
Cloud computing is just the scalable externally hosted offering of any
service which was traditionally hosted in-house.
For example, backups, email, storage... Best and clearest example of cloud
computing that I know of is ... Amazon, for example, running massively
parallel compute jobs for some companies I work for, because the companies'
needs are massively spiky. They need 1000 compute hours completed in the
shortest possible time, and then they'll have no need for any compute hours
for 3 weeks. Etc.
I'd like to see that explanation myself.
> My BS detector flashes every time someone says it, but I have been wrong
> once (okay, *maybe* twice) in the past.
That happened to me about two months ago. It was actually a relief,
knowing that I had finally made a mistake, and that I could stop
wondering when it would happen. BTW, my "BS detector" (which has an
analog indicator) has been stuck at the mechanical stop limit since the
start of the Reagan administration.
> Is it just a way of saying that you have a distributed, parallel app whose individual nodes can
> come online (or go offline) dynamically without interrupting the
> service? Is it strictly used to reference data storage - as in the
> original email in this thread - or does it mean more than that?
I think it's a way of saying "I hired someone else to do it", with
specifics left to the imagination of whomever hears the statement. For
practical purposes, it's a way of saying "Don't hold me accountable for
any mistakes made in the future: I hired <company x> to take the blame
for them".
Bill
--
Bill Horne
339-364-8487
To put it in very charged terms, "cloud computing" is outsourcing hardware and the associated IT staff. Instead of you purchasing equipment and maintaining it yourself you pay someone else to do it for you under the assumption that it will save you money.
--Rich P.
No criticism of Bil, but if I were placed in charge of a system that
depended on being able to adjust an outsourced component or process 'on
the fly', the first thing I would do would me to test how quickly it
could be transitioned to another vendor or an in-house backup. I have
seen several instances where a salesman's assurances of quick recoveries
came to a full stop on-the-fly-in-the-ointment.
In each case, the "root cause" was the fact that the System
Administration team did not have the training needed to prepare them for
a multiple-failure scenario: each vendor assumed that they were not
responsible for /anything/ but the smallest possible interpretation of
what /they/ were legally obligated to provide, with all questions of
integration, presentation, security, recoverability of meta-data, and
total-time-to-repower being left to the customer. In effect, they left
the companies that had purchased their services holding the bag, looking
for help, and facing questions from stockholders about how they got into
the mess.
I liken computing in the "Cloud" to the process of flying in clouds: it
demands careful preparation, extensive training, and well-maintained
hardware /before/ it's a reliable way to get from "A" to "B". In the
case of pilots who have "Instrument" certification, that training
includes the use of basic instruments such as a compass and airspeed
indicator to navigate without being able to see anything outside the
aircraft, since advanced electronics and autopilots can't be relied on
during systemic failures. In like manner, using "Cloud" computing
requires well-trained operators and managers whom have practiced the
needed steps to recover, and are prepared to deal with multiple failures
at the same time: I have seen something as trivial as a clip-on utility
lamp cause a server shutdown because it added just enough electrical
load to be the "straw that broke the camel's back", and popped a circuit
breaker which cast an entire room into semi-darkness, at the same time
it revealed that an apprentice electrician had inserted a "Delta"
breaker into a panel wired for "Wye" service, thus interrupting only two
legs of a three-phase power system, and leaving the lights in series
with the servers and CRT's. Some very interesting effects were visible
on the monitors and lights during the time when the responsible parties
waited in the almost-dark for me to dig a flashlight out of my case and
then lead them to the door.
We recovered the data - at the time, the IS departement was a
soup-to-nuts organization, responsible for /every/ aspect of the
company's IT, including having nightly backups in the safe, and we took
a lot of pictures of the electrical panel before the electricians
arrived, just to avoid questions from the insurance carrier. The damage
was limited to a few skinned fingers while we rushed around unplugging
everything, and one failed ballast in an overhead fixture, plus (of
course) some expensive server disks.
The moral of the story is that too many corporations switch to "Cloud"
computing as a way to justify using junior-level employees in
senior-level positions, and they tend to find out the hard way that
managing a "Cloud-based" system requires /more/ expertise and training,
not /less/.
FWIW. YMMV.
Bill
--
Bill Horne
339-364-8487
_______________________________________________
I never thought I would meet another human who could out-cynicism me.
It is a pleasure to make your virtual acquaintance, sir.
-Dan
You should read my take on SOPA: http://billhorne.com/.
Bill, who did 25 years at Verizon, and learned from the best
[A cynic is] a man who knows the price of everything and the value of
nothing"
- Oscar Wilde
"By the time the suckers figure out they paid twice the value, I'll be
rich and gone"
- Me
--
Bill Horne
339-364-8487
> your data (or your car)
> would be held hostage.
Greetings Jerry, list members.
Most of the responses were about cloud services for the corporate, where
in that case (hopefully ) there is some real technical consideration of
the "what-ifs".
for the SMB market however, Microsoft 365 and maybe Google-apps (not
sure about the latter) are a huge success. At previous lines of work I
saw them selling very well. Small IT firms really loved the fact that
they will not need to manage any local exchange servers and still earn
money.
Lawyers, Doctors, CPAs gladdy put their exchange servers and shared MS
Work/Excel/PPS on the cloud, with little understating as for what this
means - it is just like taking all your boxes of documents and giving
them to someone else,and being able to access them as long "they" allow
you. In this case, Microsoft and Google have very good PR and people
would trust them even if you portray bad scenarios.
It will only be possible to make SMB size companies to re-think these
moves if stories like " and even-though I had Internet service working,
I could not access my Email or my clients Payroll summary " will start
showing up.
--
Guy Gold
If you are talking specifically about cloud-based backup services, then
the answer is that it is a backup, not primary service. If the provider
goes away, you find another.
The general answer applicable to all cloud services is that you apply
redundancy, just as you would in any other situation. Unfortunately the
proprietary nature of most SaaS applications make this impossible. At
best, your provider might provide a way of backing up your data outside
of their cloud. (As Google does for many of its services.) But having
your data without the app may be of limited value, at least in the short
term.
Ideally this is why open source is still important, and your first
choice for a SaaS should be a hosted open source application. The next
best option is to deploy an open source application yourself to a PaaS
or IaaS provider.
I've been hoping that we'd see more open source apps in hosted form, and
we have seen some, but not really wide spread. Take for example virtual
PBXs. It's entirely feasible that we could have seen an "Asterisk
hosting" market develop much like web hosting, but it didn't happen.
There is maybe one vendor that I know of that provides Asterisk hosting.
The rest stick a proprietary GUI on top, or use an entirely proprietary
solution. If things don't work out with your PBX provider, there is no
way to download your config and prompts and upload them to another
provider.
> ...should be in several different locations fully mirrored. ...
> Companies like Amazon, Google, Microsoft, IBM, HP are huge and have
> multiple datacenters so if one datacenter gets destroyed by a
> hurricane, tornado, or a bomb, the other data centers continue
> without much of an issue.
Netflix uses Amazon to host some of its services, but I believe they
don't use them exclusively. Ultimately that's what any business with
critical infrastructure in the cloud needs to do: use multiple vendors.
Common APIs, tools like Eucalyptus (http://www.eucalyptus.com/ mentioned
elsewhere in this thread), and projects like OpenStack
(http://openstack.org/) suggest this may be more practical in the future.
-Tom
--
Tom Metro
Venture Logic, Newton, MA, USA
"Enterprise solutions through open source."
Professional Profile: http://tmetro.venturelogic.com/
Oddly all of the responses to this question seem to assume you know what
"cloud" means in this context.
Technology Review has a good article on the origin of the term cloud
computing:
Who Coined 'Cloud Computing'?
http://www.technologyreview.com/business/38987/
Unfortunately a subscription is required, but I'll include a few key
bits here:
At the time, telecom networks were already referred to as the cloud;
in engineering drawings, a cloud represented the network.
So if "cloud" is just a visual metaphor for the Internet, then "cloud
computing," in the strictest sense, just means any computing resource
that is accessed over the public Internet.
The full definition is not universally agreed to:
After the country's former IT czar, Vivek Kundra, pushed agencies to
move to cheaper cloud services, procurement officials faced the
question of what, exactly, counted as cloud computing. The government
asked the National Institutes of Standards and Technology to come up
with a definition. Its final draft[1], released this month, begins by
cautioning that "cloud computing can and does mean different things to
different people."
1.
http://csrc.nist.gov/publications/drafts/800-144/Draft-SP-800-144_cloud-computing.pdf
Another TR article summarizes and analyzes the above NIST definition:
http://www.technologyreview.com/business/38738/
The draft document defines cloud computing as "a model for enabling
ubiquitous, convenient, on-demand network access to a shared pool of
configurable computing resources (e.g., networks, servers, storage,
applications, and services) that can be rapidly provisioned and
released with minimal management effort or service provider
interaction."
The definition specifies five "essential" characteristics of cloud
computing: self-service; accessibility from desktops, laptops, and
mobile phones; resources that are pooled among multiple users and
applications; elastic resources that can be rapidly reapportioned as
needed; and measured service.
The article also defined the related buzzwords: IaaS, PaaS, and SaaS.
If you're curious about the history, the "Who Coined 'Cloud Computing'?"
article goes on to dig up documented evidence for the term dating back
to 1996. The first use by Sean O'Sullivan, CEO of (now defunct) startup
NetCentric, followed a few weeks later by a mention in an internal
Compaq document by George Favaloro. (O'Sullivan and Favaloro had contact
and Compaq invested in NetCentric. So there is dispute over who used the
term first.)
> My BS detector flashes every time someone says it...
Your detector is well justified, given it is a marketing term (quoting
again from "Who Coined 'Cloud Computing'?"):
Cloud computing still doesn't appear in the Oxford English Dictionary.
But its use is spreading rapidly because it captures a historic shift
in the IT industry... Other technology vendors, such as IBM and
Oracle, have been accused of "cloud washing," or misusing the phrase
to describe older product lines.
Like "Web 2.0," cloud computing has become a ubiquitous piece of
jargon that many tech executives find annoying, but also hard to
avoid. "I hated it, but I finally gave in," says Carl Bass, president
and CEO of Autodesk, whose company unveiled a cloud-computing
marketing campaign in September. "I didn't think the term helped
explain anything to people who didn't already know what it is."
[O'Sullivan and Favaloro] Both agree that "cloud computing" was born
as a marketing term. ... What they were hunting for was a slogan to
link the fast-developing Internet opportunity to businesses Compaq
knew about. "Computing was bedrock for Compaq, but now this messy
cloud was happening," says Favaloro. "And we needed a handle to bring
those things together."
Now try out your detector on the term "Service Oriented Architecture."
Have fun finding a clear definition for it. :-)
TR has a bunch of other articles on the topic:
http://www.technologyreview.com/business/?id=20
Including:
The Cloud Imperative (by former local Simson L. Garfinkel)
http://www.technologyreview.com/business/38710/
Treating computing as a utility, like electricity, is an old idea. But
now it makes financial sense--a historic shift that is reshaping the
IT industry.
...
The facts are really simple: although every organization on the
Internet essentially is using some cloud-based service, they should
use more. The economies of scale are becoming mind-blowing. Someone
who wants to go buy a rack of servers probably hasn't done the math.
They also have an article on Facebook's "open hardware" servers, which
I'll discuss in a separate post.
> Is it just a way of saying that you have a distributed, parallel app
> whose individual nodes can come online (or go offline) dynamically
> without interrupting the service?
I would say that although many cloud services include those attributes,
none of them are strictly required to fit my definition, but it looks
like you are pretty close to the NIST definition. (Distributed could be
argued is required if you count the fact that user's location and the
server's location are different as being distributed. I wouldn't.)
-Tom
--
Tom Metro
Venture Logic, Newton, MA, USA
"Enterprise solutions through open source."
Professional Profile: http://tmetro.venturelogic.com/