1) Code related issues lead to critical failure. For a revenue generating system this can have immediate effects on the bottom line. For non-revenue systems this can affect how many support hours are burned fixing problems.
2) Financial impact from lawsuits or non-compliance can be costly. If a security bug or design flaw in your application leads to PCI non-compliance, you may find your cost of doing business with credit providers can rise dramatically.
Actually a separate MP3 player is a must when in the gym. Trying to carry around even the smallest smartphone is a chore. Either you carry it in your hand or get one of those dorky looking armband harnesses. As for the sound quality, it's good enough for me. I enjoy my music to the extent that I've purchased a good set of headphones for my home computer.... but I don't think much about the nuance of sound color when running a 5K.
I used to work at a shipping company. They used to provide higher volume customers with a PC that ran their shipping software. One of my responsibilities was to maintain this equipment at the customer sites (upgrade software, fix PCs, etc.). So the company assigned me to the Florida West Coast (Naples, Ft. Myers, etc.). I arrived at the facility with a list of about a dozen customers to see that week. My first task was to get the spare equipment to replace/install at customer sites. There was supposed to be a small room with all the equipment. I went to the room but it was empty. I called the home office... They said something to the effect that "It's a large room with a whole bunch of PCs, monitors, keyboards. You can't miss it." So I looked again. Even though it was a good sized warehouse, there were only four or so rooms. Nope, couldn't find all that equipment. Finally got in touch with the manager there.... Yup, my predecessor had loaded all the equipment into a truck and taken it away.
I called my office. And yeah, I figured that something suspicious had happened but I had to play it dumb (can't go around accusing someone of theft if I wasn't certain). Call went something like: "Hey, the former admin took all the equipment away. Where did he take it?"
"What do you mean he took the equipment?"
"I understand that he loaded everything into a truck last week and drove off. Let me know where and I will see about moving it back."
"What do you mean he took the equipment?"
"The facility manager said he took the whole day loading everything up into a Ryder truck. Then he drove off."
"Where did he take them?"
"I don't know. I just got here today. "
The beauty of his move was that he maintained all the inventory... So when it came time to see how much equipment was supposed to be there, everything showed as empty or at customer sites or disposed off... A roomful of brand new equipment was marked as "Disposed" or "Sent back"...
Today's kids start with computers that are already large systems with complex operating systems, millions of times more memory than we have *disk* when we started, that are difficult to understand at a low level.
That's a good point, and largely agree that todays systems are more complicated... But there's more to it, I think. Back then, when we played with our Commodore 64s and Atari 800XLs (and Ti 99/4A and CoCos and whatnot), it was easy to fix a problem.. You just rebooted. If you screwed up something by POKEing some reserved area, you just rebooted (well, unless you had a PET and could destroy the machine with an errant POKE).
It's coming around to that point again. With older systems, and cheaper systems, it's easy to experiment again. I love Linux for this reason.. It does let me poke around in the internals of the OS and break things and fix things... Whether it's mucking around with init scripts or actually rebuilding the kernel, it does bring back that "see what this does" attitude.
I keep a few Linux instances running on some VMWare and KVM based servers on my home network. The desktop systems run vncserver and I can access the sessions remotely from any system in the house. Though I run some of the same apps locally, there are enough reasons to run them on the central server.
1) The types of apps I need are not available easily on the client. For example, I use some photography related apps under Ubuntu. These are free and easily available via the Software Manager. The same quality of apps are not available under Win7. For example, there are some HDR utilities I use in Ubuntu that work quite well. Similar software under Win7 or MacOSX costs $40 or so.
2) The netbooks I've started to use don't have the power needed to run some of the larger apps. Though my main laptop (CentOS 5.5) can handle it, I have some Atom based systems that have issues running a JDE or full blown dev environment.
3) I have *many* client devices. At last count I have 10 laptops in the house. These run CentOS, Ubuntu, MacOSX, Win7, WinXP and Fedora. This is unusual for most households, but reflects the type of environment I'm seeing in smaller businesses. No matter what client I use I can run my set of apps.
I have a few hundred files worth of Atari and Commodore data from the 80s. In their original format they took up a few cases of floppies. It was actually pretty easy to read them with some emulation software under Linux. Now, alas, I don't have any floppy drives on any of my machines. VLC still plays the audio files and Imagemagick can read the image files. All told about a gig worth of data. Though I'm usually pretty good about purging old cruft, all of it takes up such a small amount of space that I still keep them hanging around.
Learning math is just as difficult as learning any other subject or content material.
You think so?
One of the things I promised I would never do is to assume that my area of study is harder than another's. Or, assume that because I do not grasp something easily it is a difficult subject.
So I started as an art major in college and was a straight A student for a dozen classes. Lots of that was spent sitting by the lake with a guitar in my hand hoping that some girl would think I was cool (didn't work). Well, there was also an hourly visit to the campus gallery to look at art and write some bullshit story about it. Most amazing was the ability to take my *opinion* and get an 'A' grade because I mentioned shamanism or diversity or Warholian ethos or myth cycles. Cool thing? It was my *interpretation* so could never be wrong.
I switched from Art to English (there's was an English major chick who I had this mad crush on). You know those art history papers I wrote? Amazingly I could use the same themes in English. Same approaches too.. Look at the political clime at the time (hey that rhymes), make a statement that all art is political (because the professor wants to rebellious against the others and say that art is not motivated by beauty) and voila, an 'A' paper. In one paper I managed to roll together gothicism (not black fingernail gothicism, but you know, the rebellion against the void type), the rise of photography, Godel, and Eastern mysticism into one 3,000 word essay. cool thing? It was interpretation and could never be wrong. Even the historical parts -- you know, Dante's influence on Tolkien or Conrad's on Vonnegut -- were so broadly argued that it could never be wrong....
Funny thing though... when I finally switched to computer science and mathematics major that shit didn't fly. I had to work my ass off to pull down those grades. For the first time in my college career, I got a C in a class. Up to that point it was all A's. MAC1101 was easy enough. So were the first couple calculus courses. Calc II was tough, but I scraped by. Calc 3 wasn't so bad. Differential equations kicked my ass. I still remember dreading the tests. I still remember trying to figure out how to find vector fields and tensors and determine saddle points and lots of stuff that I've forgotten. I read my notes from back then and I can't make heads or tails of it. If my "interpretation" of the math was incorrect, my answers were marked wrong.
So is one harder than another? I don't know. But I can tell you that it's certainly easier to get a high grade in an English or Art class than it was for Mathematics. At least for me.
No doubt.. I think they're looking at spellcheck as just another extension of calculators on math tests. The argument was that the rote computation didn't relate to understanding of mathematics concepts so calculators were benign.
In my day we had penmanship exams...:D And I imagine that at some point they'll allow full word processors to allow the penmanship-challenged to submit work without fear of their papers being illegible.
I didn't want to fall into the classic old geezer thinking that everything was harder back in the day...So I peeked at the curriculum for some of the local high schools. And damn, it was harder in my day. In my high school classes back then we learned about Turing and Godel and their impact on how computers are designed. We didn't write much code, but I remember blackboard sessions on sorting algorithms, queuing, floating point operations, etc..
So I wonder.. 25 years ago, did other adults look at the high school curriculum and think the same thing? In the 1960s there was a push for "new math" which apparently included set theory and base-n computation, both of which would be very helpful in computer science. And I can imagine that even though Simpson and Newton-Raphson methods were centuries old, the computers of the 1960s were not necessarily accessible to students.
It reminds me of a story by Roger Zelazny. There is a mythical creature that didn't have hands. It loved to play chess, but because of his lack of hands (and IIRC, lack of opponents), this mythical creature had to play chess games in his head. He got to be very good at mental chess.
The upside of this is that there are are some very bright high school students out there. Twenty five years ago the people who were interested in computers were just a handful. In my class there were five or so. In a given high school there are probably still that many but it's harder to spot them because typing classes are masquerading as computer science.
After a hurricane had wiped out power to Miami, I had to drive into a facility I maintained to get their email servers back online. It was critical for their remote employees to send in orders and time sheets. This was back before outsourced email services such as Google or Yahoo were available.
When I got there the power was still off. I had to rely on a 300W inverter plugged into the owner's truck battery. We ran a high gauge extension cord about 50' to the truck parked outside. Next we added a power strip to the end of the extension and plugged in the modem, server and monitor. On powering up the fuse for the cigarette adapter blew. We clamped up directly to the battery then. Powered up the monitor, modem, then PC. Everything worked for about 3 seconds until the BIOS splash screen turned on. Then it all went dead. The 300W inverter was not enough to power on both the server and the old CRT. We had the bright idea to charge a UPS for 30 minutes. With the monitor plugged into the UPS, we had just enough juice to see that the server has hanging on a bad filesystem. Then it died.
This is where it got fun.
I unplugged the monitor. As the system booted, I replayed in my head the steps I needed to bring the filesystem back. I knew that needed to login to maintenance mode first. I knew this by entering the root password then typing (blindly) "touch/tmp/foo; find/tmp -name foo". When I saw the hard drive light flicker when I pressed enter I knew I was at the shell.
I had to check the filesystems... I didn't remember what partition it was on, so on a piece of paper I wrote out an awk script that would peek through/etc/fstab, grab the relevant filesystems and the appropriate/dev entry, then pass that to stdout. I piped that output to a file then used that file to run fsck. All of this was done without seeing my commands or the output from those commands.
When the remote user was able to connect via mail then I knew it was working..
It wasn't particularly ingenious, but the circumstances made it memorable. Missing pieces of the room, navigating around downed trees to get to the site, complete darkness except for a door propped open on the other side of the room (server room was the farthest room in the office and had no windows or doors to the outside), hot hot hot hot hot (Florida weather), and users calling every five minutes trying to connect... Power came up later that day, but what an experience.
Yeah, I worked at UPS. It wasn't only the employees.
In our Hialeah sort facility we had a device called a BullFrog which sorted packages along a conveyor belt. In its initial stages it didn't work quite correctly and some packages made it through to the end and just *fell* about ten feet into a bin for reprocessing.
During the peak season (i.e., right now), thousands of packages would travel through chutes meant for hundreds of packages. During regular days there would be a jam every few minutes as oversize packages clogged the chutes. In peak it was much worse. Packages would crush up against each other and an employee would need to walk up the conveyor belt and clear the jams. Many packages were damaged at these times.
There are very explicit rules against damaging packages and penalties can include termination. However, the volume of packages during peak is enormous and loaders fall behind, package cars get overloaded and need to be "assisted" to add extra packages, and sometimes yes, there are some bad employees.
Is it better at other companies? Not necessarily. I received a package via FedEx that was accordianed. It looked like the box was compressed from about 12" to about 5". And they still delivered it in that condition.
I loved school. I took classes because I was really interested in them. Though I ended up with computer science/mathematics, on the way I took art classes, history, literature, philosophy, chemistry... most of them not required for my degree. And throughout elementary school, middle school, high school and college, I never cheated. Not even once.
Twenty years later, the teachers who invested the time in their classrooms are still influencing me today. Whether watching Lost and recalling Robinson Crusoe from English Lit, or hearing Lady Gaga and thinking about Warhol, or reading about Buffett and remembering an economics course, those few hours spent learning makes my life so much richer.
I actually feel an incredible sadness when I see kids rushing to get through college. Those years are so infinitely amazing that cheating your way through it is an affront to life itself. There will rarely be another time when you will be surrounded with so many vibrant minds.
And to those who make the excuse that they cannot afford to take courses outside their degree, know that I worked my way through school. It was hard. Damned hard. Going to work at 2AM, getting out at 8:30AM, then rushing to class.. Studying until 8PM, sleeping for a few hours... On weekends scraping together a few dollars to go out with friends (because that's part of it too). But it was all worth it. Given the chance again, I'd do it in a heartbeat.
OK, seriously, I've done a couple dozen of these 10 to 50 user installations. Half the time is spent at the beginning to determine what the customer needs and wants, and what the budgeting will be. Things invariably cost a lot more than the customer anticipated so your goal is to manage expectations. If you don't do that, your life will either become a living hell (if you will be providing long-term support) or you will leave behind an unhappy customer.
Some of the basic things that were not considered when customers brought me on:
Are there remote employees? Will they need VPN access? What platforms are they using to connect? Can you verify that the endpoints are secure?
What is the anticipated volume of mail? In this day, it's often much cheaper to outsource to Google for smaller installations, but in some cases it makes a lot of sense to keep in-house.
When hosting your own web server how much downtime is acceptable? Do you need 24/7 uptime or will you have maintenance windows? What if your primary site burns to the ground? Do you have the floor space and adequate cooling? How much traffic is anticipated at the beginning of the project? How much do you expect to grow?
What applications do you need in-house? Accounting packages? Company intranet? Database? How will you separate your LAN for security purposes? Do you take credit cards as part of business?
What infrastructure applications do you need? Can you afford downtime on these? How many ports/switches do you need? Wireless? Separate backup LAN? OOB management for your servers?
Before you even start pricing hardware, find out what your customer needs and wants and willing to pay for.
The "Cloud", as referenced here, is nothing more than the delegation of responsibilities...specifically those of infrastructure. That's it. It's not some mystical cure all. In fact, it's nothing more than a glorified way to outsource applications.
Well, no. The cloud they referenced was an "abstracted data-center infrastructure" and not necessarily a means of outsourcing applications. Yes, the downside/upside is that it eases moving workloads from internal to external clouds, but that's the point.
Now there are specific technologies which lend themselves to this concept ( those of virtualization, certainly ), but the overall goal is the same; the business doesn't want to worry about the infrastructure behind their app. They simply want it to work.
Is that a bad thing not to want to worry about the infrastructure? Traditionally servers are designed around the concept of a physical server. We used to name servers by rack number or some other geographic location. Virtual machines were often named according to what physical server they resided within. Cloud technology, once the marketing speak is burned away and the APIs get to a mature and standard state (i.e., an in-house or an outside hosted cloud looks the same to an application), would allow other ways of managing the hundreds of thousands of machines in large data centers.
For example, capacity planning is a big deal. One of the responsibilities of a system engineer is to ensure that workloads can run properly on the servers. When there is a planned outage on one server or an increased load due to seasonal or scheduled work, the admins have to juggle the resources of the servers. In a planned outage we may use VMWare VMotion or Workload Migration and swing the workloads across. But then we often have to worry about IP changes, hostnames, virtual host software levels, etc.. With a properly configured internal cloud, this is a non-issue. I can literally click a button and remove a physical server from the cluster and it's completely transparent to end-users. Need to add capacity? I SAN-boot a cloned disk and the new server is automatically part of the cloud and ready to take on work.
We used to build our environments around managing discrete servers. Even if we had streamlined the process, it was still very much centered around the physical box. For example, we can stand up a box in a manner of minutes using RHEL kickstart, but if we wanted to add high availability this often meant configuring heartbeat IPs, swing SAN disks,/etc/hosts files for private IP ranges, etc.. HA on a cloud is almost too trivial to detail.
Of course it's not there yet, but it's where the more recent virtualization technologies was 5 years ago (and yeah, virtualilzation has been out for decades, but it has only within the past decade really surged).
I know exactly what the problem is. Try passing different elevator options to the kernel at boot time. It helps with clock skew on virtual environments.
They will get cheaper. I picked up a 60G SSD from Newegg for under $100 after rebate. In a few months I expect that to be the normal price.
Is it worth it? Hell, yes. For systems where you need a lot of space or battery life isn't an issue, then they're probably not ideal. However, in a netbook they are amazing. I have a Samsung N120 with a 1.6GhZ Atom. With a standard HD, it was boggy. Resuming from suspend would take a minute. Launching apps would take 15 to 30 seconds. After installing the SSD it's like a new machine. Resume takes a few seconds. App launch times is a second or three. Browsing the web is snappier. I.e., anything that does multiple reads from the drive is much faster. If you replaced your standard laptop drive you may not notice it, but replacing a relatively slow HD in a netbook makes a huge difference. On top of it, my battery times climbed to at least 4 hours of constant use.
BTW, the SSDs run great with bcache/Linux. I'm putting together some benchmarks, but even before I run the numbers I can tell you that CentOS and Ubuntu on an Atom-based machine (a mini-pc form factor) runs incredibly.
I know that there are folks who enjoy wrestling as much as I enjoy sci-fi/fantasy but never made sense to me that they'd stick them both on the same channel. I don't know if they just needed *any* channel to air it and it was purely economical but I can imagine some exec thinking, "Well, the demographic for sci-fi/fantasy is mainly male. Therefore, they will like wrestling." I recently canceled all but the most basic cable and Internet after realizing that every time I wanted to watch something on sci-fi I found either some idiotic ghost busters reality show or wrestling.
"Never attribute to malice what can be explained by incompetence..."
The problem is likely some poor interface design. I've seen it used deliberately on some installers in order to sneakily add other products. It may follow a series of "Next" buttons that asks "Also install McAfee agent" or "Install Yahoo Toolbar"... In this case, the checkbox for the candidate may happen to be on the "Next" button of the previous screen.
It's a boring sentence trapped in a boring, verbose memo, so I found it a new home in a Philip K Dick story:
The movie was based on a Dick story, but that line was only in the screenplay (and modified by Rutger Hauer during filming)...
As to the corporate speak.. yeah, that's part of corporate culture. You have to remember that the people in the meetings are not always very well edu-macated. I remember attending meetings at UPS. Most of the managers were promoted through the ranks and started as truck drivers. Many had only a high school education. Many still spoke like dock workers (which is where they may have entered the company).
I was representing the TSG (Technology Support Group) and still remember a wonderful "discussion" that went something like this: "The pre-load crew wants to know what time we're starting on Monday" (Monday being the first day of the Peak season, which was typically the day after Thanksgiving break).
"Fuck 'em. They'll come in we tell them."
"Well, yeah, but we need to let them know when to come in."
"Fuck you. You don't have a fucking clue."
And this was just a normal meeting.
Corporate speak reminds me of listening to some Jamaicans (and yes, I'm Jamaican and maybe as guilty). In Jamaica much of the population did not have formal education beyond the equivalent to 8th grade. But you know, the less you know, the more you think you know. It was always entertaining to hear some of the less educated ones using and mis-using words.
"We're going to creativate new jobs by learning the children about technological progress in the classroom. We are going to learn them about the computer and the CRT and the hard drive and the programming. We are going to elevate their understanding and invigorate the learning. So Mr. Accountant with his calculator and his writing implements and accounting books, I as a man and as a computer operator -- yes, computer operator -- demand that these funds and moneys go to the Kingston school district."
No, really, this was really what I heard... Corporate speak sounds the same to me....
There are some advantages to running KVM. There's a lot of flexibility in storage configuration and pretty much any host supported storage device will work. You also get some management and reporting capability. But yeah, I understand your point. At work I run a lot of PowerVM systems, actually 99% of our AIX environment is an LPAR, but even the VIO servers and HMCs run an OS. If you don't go the OS route then the hypervisor tends to be limited on what it can support. As to recoverability, using an OS means you can use all the OS tools. E.g., dd or rsync or LVM snapshots can be used to backup an OS image. With a pure hypervisor you tend to rely on exporting a guest VM.
Xen and paravirt is great. You can certainly cram a lot of instances into a Xen system. The downside is that your client OS must support that virtualization. Now Xen has made some strides with Windows and other non-Linux workloads, but many still need to run in full virt.
I'm running multiple VM systems including VMWare Server, VMWare ESX, VMWare Workstation, xen, KVM and VirtualBox.
VMWare Server is going away and sort of a pain to manage. However, it was free and worked decently. I have since replaced it with VMWare Workstation on my desktop and laptop systems. I use VirtualBox on my Mac laptop because it's free and was the easiest/cheapest to get going.
On my servers I am running VMWare ESX, xen and KVM on AMD systems (mostly dual core, but a couple quad core systems in the mix).
VMWare ESX was the most finicky as to installation but has been pretty simple to manage. The remote console options are simple. The VSphere management client is Windows only though. There is support for command line administration, but it's somewhat of a bear. You can script around it though and many people have done so and provide scripts online. Check out the VMWare community pages. Support is so so..
Xen was my workhorse for the longest time, but since my primary OS is RedHat/CentOS and RH is moving towards KVM, I've also been moving to KVM. The GUI management tools work fine, but are not as polished as VMWare ESX. However, it very much makes up for it in being able to do just about everything from the command line. I can deploy an image with a single command and this works wonderfully for testing. Performance is awesome with both xen and KVM. Well, the caveat is that some network intensive stuff seems to be bottlenecking somewhere, but it only has a single gigabit NIC across 8 VMs. I'll be adding another NIC in the next couple weeks and either bonding the adapter or just splitting them up.
Be aware that client/guest images generally do not have video acceleration so many games will fail to load. If you're running VMWare Workstation on a laptop, or the more recent KVMs then there is some measure of acceleration, but not 100%. Also, sound can be finicky especially across the network.
The whole cloud concept is still being defined.. It's, um, cloudy....
That said, there are many reasons why some companies could benefit from the idea. For example, say that you want to build a global enterprise. Up until recently, this would take lots of money in order to have a physical "presence" in many different countries. With a global internet it's possible and it doesn't matter where the physical server is located as you can make the best use of local resources. Extend this concept a bit further and you realize that it doesn't matter where your physical server is located. Companies began to co-locate servers so they didn't need to have offices in Singapore to have a geographically ideal location for business, political, financial and even technical reasons.
The same challenges faced with local, company hosted resources are there with the cloud model. What happens when your company hosted server shuts down? what happens when your company network connections go down and your customers can't access your servers? What happens when an internal employee gets fired and sends out your customer list?
What the cloud concept can offer is reduced expense of hosting a large data center. For example, in the US land prices in areas where talent is located is generally high. It goes hand in hand. If a city is popular with technical folks then land prices will be high. The solution may be to host the personnel offices in one location and the servers somewhere in the cloud. Remember, building a data center is not a trivial thing. You need air conditioning, water, ducts, conduits in the floor, power, backup batteries and generators, fire suppression systems, etc..
No, clouds don't magically make standard IT problems disappear, but they can give a lot of flexibility in how you approach them.
I wonder what the laws are if you happen to forget the password? I use one-time passwords all the time. Some are 10 characters or more. I count on my ability to either reset the password or re-create the data. Politicians do it all the time. "It slipped my mind" or "It was ten years ago" or "I get so many papers that it's hard to remember what I signed".
Slashdot Mirror
A couple things could happen:
1) Code related issues lead to critical failure. For a revenue generating system this can have immediate effects on the bottom line. For non-revenue systems this can affect how many support hours are burned fixing problems.
2) Financial impact from lawsuits or non-compliance can be costly. If a security bug or design flaw in your application leads to PCI non-compliance, you may find your cost of doing business with credit providers can rise dramatically.
Actually a separate MP3 player is a must when in the gym. Trying to carry around even the smallest smartphone is a chore. Either you carry it in your hand or get one of those dorky looking armband harnesses. As for the sound quality, it's good enough for me. I enjoy my music to the extent that I've purchased a good set of headphones for my home computer.... but I don't think much about the nuance of sound color when running a 5K.
I used to work at a shipping company. They used to provide higher volume customers with a PC that ran their shipping software. One of my responsibilities was to maintain this equipment at the customer sites (upgrade software, fix PCs, etc.). So the company assigned me to the Florida West Coast (Naples, Ft. Myers, etc.). I arrived at the facility with a list of about a dozen customers to see that week. My first task was to get the spare equipment to replace/install at customer sites. There was supposed to be a small room with all the equipment. I went to the room but it was empty. I called the home office... They said something to the effect that "It's a large room with a whole bunch of PCs, monitors, keyboards. You can't miss it." So I looked again. Even though it was a good sized warehouse, there were only four or so rooms. Nope, couldn't find all that equipment. Finally got in touch with the manager there.... Yup, my predecessor had loaded all the equipment into a truck and taken it away.
I called my office. And yeah, I figured that something suspicious had happened but I had to play it dumb (can't go around accusing someone of theft if I wasn't certain). Call went something like:
"Hey, the former admin took all the equipment away. Where did he take it?"
"What do you mean he took the equipment?"
"I understand that he loaded everything into a truck last week and drove off. Let me know where and I will see about moving it back."
"What do you mean he took the equipment?"
"The facility manager said he took the whole day loading everything up into a Ryder truck. Then he drove off."
"Where did he take them?"
"I don't know. I just got here today. "
The beauty of his move was that he maintained all the inventory... So when it came time to see how much equipment was supposed to be there, everything showed as empty or at customer sites or disposed off... A roomful of brand new equipment was marked as "Disposed" or "Sent back"...
Today's kids start with computers that are already large systems with complex operating systems, millions of times more memory than we have *disk* when we started, that are difficult to understand at a low level.
That's a good point, and largely agree that todays systems are more complicated... But there's more to it, I think. Back then, when we played with our Commodore 64s and Atari 800XLs (and Ti 99/4A and CoCos and whatnot), it was easy to fix a problem.. You just rebooted. If you screwed up something by POKEing some reserved area, you just rebooted (well, unless you had a PET and could destroy the machine with an errant POKE).
It's coming around to that point again. With older systems, and cheaper systems, it's easy to experiment again. I love Linux for this reason.. It does let me poke around in the internals of the OS and break things and fix things... Whether it's mucking around with init scripts or actually rebuilding the kernel, it does bring back that "see what this does" attitude.
I keep a few Linux instances running on some VMWare and KVM based servers on my home network. The desktop systems run vncserver and I can access the sessions remotely from any system in the house. Though I run some of the same apps locally, there are enough reasons to run them on the central server.
1) The types of apps I need are not available easily on the client. For example, I use some photography related apps under Ubuntu. These are free and easily available via the Software Manager. The same quality of apps are not available under Win7. For example, there are some HDR utilities I use in Ubuntu that work quite well. Similar software under Win7 or MacOSX costs $40 or so.
2) The netbooks I've started to use don't have the power needed to run some of the larger apps. Though my main laptop (CentOS 5.5) can handle it, I have some Atom based systems that have issues running a JDE or full blown dev environment.
3) I have *many* client devices. At last count I have 10 laptops in the house. These run CentOS, Ubuntu, MacOSX, Win7, WinXP and Fedora. This is unusual for most households, but reflects the type of environment I'm seeing in smaller businesses. No matter what client I use I can run my set of apps.
I have a few hundred files worth of Atari and Commodore data from the 80s. In their original format they took up a few cases of floppies. It was actually pretty easy to read them with some emulation software under Linux. Now, alas, I don't have any floppy drives on any of my machines. VLC still plays the audio files and Imagemagick can read the image files. All told about a gig worth of data. Though I'm usually pretty good about purging old cruft, all of it takes up such a small amount of space that I still keep them hanging around.
Learning math is just as difficult as learning any other subject or content material.
You think so?
One of the things I promised I would never do is to assume that my area of study is harder than another's. Or, assume that because I do not grasp something easily it is a difficult subject.
So I started as an art major in college and was a straight A student for a dozen classes. Lots of that was spent sitting by the lake with a guitar in my hand hoping that some girl would think I was cool (didn't work). Well, there was also an hourly visit to the campus gallery to look at art and write some bullshit story about it. Most amazing was the ability to take my *opinion* and get an 'A' grade because I mentioned shamanism or diversity or Warholian ethos or myth cycles. Cool thing? It was my *interpretation* so could never be wrong.
I switched from Art to English (there's was an English major chick who I had this mad crush on). You know those art history papers I wrote? Amazingly I could use the same themes in English. Same approaches too.. Look at the political clime at the time (hey that rhymes), make a statement that all art is political (because the professor wants to rebellious against the others and say that art is not motivated by beauty) and voila, an 'A' paper. In one paper I managed to roll together gothicism (not black fingernail gothicism, but you know, the rebellion against the void type), the rise of photography, Godel, and Eastern mysticism into one 3,000 word essay. cool thing? It was interpretation and could never be wrong. Even the historical parts -- you know, Dante's influence on Tolkien or Conrad's on Vonnegut -- were so broadly argued that it could never be wrong....
Funny thing though... when I finally switched to computer science and mathematics major that shit didn't fly. I had to work my ass off to pull down those grades. For the first time in my college career, I got a C in a class. Up to that point it was all A's. MAC1101 was easy enough. So were the first couple calculus courses. Calc II was tough, but I scraped by. Calc 3 wasn't so bad. Differential equations kicked my ass. I still remember dreading the tests. I still remember trying to figure out how to find vector fields and tensors and determine saddle points and lots of stuff that I've forgotten. I read my notes from back then and I can't make heads or tails of it. If my "interpretation" of the math was incorrect, my answers were marked wrong.
So is one harder than another? I don't know. But I can tell you that it's certainly easier to get a high grade in an English or Art class than it was for Mathematics. At least for me.
Spell check doesn't fix bad writing though :P
No doubt.. I think they're looking at spellcheck as just another extension of calculators on math tests. The argument was that the rote computation didn't relate to understanding of mathematics concepts so calculators were benign.
In my day we had penmanship exams... :D And I imagine that at some point they'll allow full word processors to allow the penmanship-challenged to submit work without fear of their papers being illegible.
I didn't want to fall into the classic old geezer thinking that everything was harder back in the day...So I peeked at the curriculum for some of the local high schools. And damn, it was harder in my day. In my high school classes back then we learned about Turing and Godel and their impact on how computers are designed. We didn't write much code, but I remember blackboard sessions on sorting algorithms, queuing, floating point operations, etc..
So I wonder.. 25 years ago, did other adults look at the high school curriculum and think the same thing? In the 1960s there was a push for "new math" which apparently included set theory and base-n computation, both of which would be very helpful in computer science. And I can imagine that even though Simpson and Newton-Raphson methods were centuries old, the computers of the 1960s were not necessarily accessible to students.
It reminds me of a story by Roger Zelazny. There is a mythical creature that didn't have hands. It loved to play chess, but because of his lack of hands (and IIRC, lack of opponents), this mythical creature had to play chess games in his head. He got to be very good at mental chess.
The upside of this is that there are are some very bright high school students out there. Twenty five years ago the people who were interested in computers were just a handful. In my class there were five or so. In a given high school there are probably still that many but it's harder to spot them because typing classes are masquerading as computer science.
:)
After a hurricane had wiped out power to Miami, I had to drive into a facility I maintained to get their email servers back online. It was critical for their remote employees to send in orders and time sheets. This was back before outsourced email services such as Google or Yahoo were available.
When I got there the power was still off. I had to rely on a 300W inverter plugged into the owner's truck battery. We ran a high gauge extension cord about 50' to the truck parked outside. Next we added a power strip to the end of the extension and plugged in the modem, server and monitor. On powering up the fuse for the cigarette adapter blew. We clamped up directly to the battery then. Powered up the monitor, modem, then PC. Everything worked for about 3 seconds until the BIOS splash screen turned on. Then it all went dead. The 300W inverter was not enough to power on both the server and the old CRT. We had the bright idea to charge a UPS for 30 minutes. With the monitor plugged into the UPS, we had just enough juice to see that the server has hanging on a bad filesystem. Then it died.
This is where it got fun.
I unplugged the monitor. As the system booted, I replayed in my head the steps I needed to bring the filesystem back. I knew that needed to login to maintenance mode first. I knew this by entering the root password then typing (blindly) "touch /tmp/foo; find /tmp -name foo". When I saw the hard drive light flicker when I pressed enter I knew I was at the shell.
I had to check the filesystems... I didn't remember what partition it was on, so on a piece of paper I wrote out an awk script that would peek through /etc/fstab, grab the relevant filesystems and the appropriate /dev entry, then pass that to stdout. I piped that output to a file then used that file to run fsck. All of this was done without seeing my commands or the output from those commands.
When the remote user was able to connect via mail then I knew it was working..
It wasn't particularly ingenious, but the circumstances made it memorable. Missing pieces of the room, navigating around downed trees to get to the site, complete darkness except for a door propped open on the other side of the room (server room was the farthest room in the office and had no windows or doors to the outside), hot hot hot hot hot (Florida weather), and users calling every five minutes trying to connect... Power came up later that day, but what an experience.
Yeah, I worked at UPS. It wasn't only the employees.
In our Hialeah sort facility we had a device called a BullFrog which sorted packages along a conveyor belt. In its initial stages it didn't work quite correctly and some packages made it through to the end and just *fell* about ten feet into a bin for reprocessing.
During the peak season (i.e., right now), thousands of packages would travel through chutes meant for hundreds of packages. During regular days there would be a jam every few minutes as oversize packages clogged the chutes. In peak it was much worse. Packages would crush up against each other and an employee would need to walk up the conveyor belt and clear the jams. Many packages were damaged at these times.
There are very explicit rules against damaging packages and penalties can include termination. However, the volume of packages during peak is enormous and loaders fall behind, package cars get overloaded and need to be "assisted" to add extra packages, and sometimes yes, there are some bad employees.
Is it better at other companies? Not necessarily. I received a package via FedEx that was accordianed. It looked like the box was compressed from about 12" to about 5". And they still delivered it in that condition.
I loved school. I took classes because I was really interested in them. Though I ended up with computer science/mathematics, on the way I took art classes, history, literature, philosophy, chemistry... most of them not required for my degree. And throughout elementary school, middle school, high school and college, I never cheated. Not even once.
Twenty years later, the teachers who invested the time in their classrooms are still influencing me today. Whether watching Lost and recalling Robinson Crusoe from English Lit, or hearing Lady Gaga and thinking about Warhol, or reading about Buffett and remembering an economics course, those few hours spent learning makes my life so much richer.
I actually feel an incredible sadness when I see kids rushing to get through college. Those years are so infinitely amazing that cheating your way through it is an affront to life itself. There will rarely be another time when you will be surrounded with so many vibrant minds.
And to those who make the excuse that they cannot afford to take courses outside their degree, know that I worked my way through school. It was hard. Damned hard. Going to work at 2AM, getting out at 8:30AM, then rushing to class.. Studying until 8PM, sleeping for a few hours... On weekends scraping together a few dollars to go out with friends (because that's part of it too). But it was all worth it. Given the chance again, I'd do it in a heartbeat.
OK, seriously, I've done a couple dozen of these 10 to 50 user installations. Half the time is spent at the beginning to determine what the customer needs and wants, and what the budgeting will be. Things invariably cost a lot more than the customer anticipated so your goal is to manage expectations. If you don't do that, your life will either become a living hell (if you will be providing long-term support) or you will leave behind an unhappy customer.
Some of the basic things that were not considered when customers brought me on:
Are there remote employees? Will they need VPN access? What platforms are they using to connect? Can you verify that the endpoints are secure?
What is the anticipated volume of mail? In this day, it's often much cheaper to outsource to Google for smaller installations, but in some cases it makes a lot of sense to keep in-house.
When hosting your own web server how much downtime is acceptable? Do you need 24/7 uptime or will you have maintenance windows? What if your primary site burns to the ground? Do you have the floor space and adequate cooling? How much traffic is anticipated at the beginning of the project? How much do you expect to grow?
What applications do you need in-house? Accounting packages? Company intranet? Database? How will you separate your LAN for security purposes? Do you take credit cards as part of business?
What infrastructure applications do you need? Can you afford downtime on these? How many ports/switches do you need? Wireless? Separate backup LAN? OOB management for your servers?
Before you even start pricing hardware, find out what your customer needs and wants and willing to pay for.
The "Cloud", as referenced here, is nothing more than the delegation of responsibilities...specifically those of infrastructure. That's it. It's not some mystical cure all. In fact, it's nothing more than a glorified way to outsource applications.
Well, no. The cloud they referenced was an "abstracted data-center infrastructure" and not necessarily a means of outsourcing applications. Yes, the downside/upside is that it eases moving workloads from internal to external clouds, but that's the point.
Now there are specific technologies which lend themselves to this concept ( those of virtualization, certainly ), but the overall goal is the same; the business doesn't want to worry about the infrastructure behind their app. They simply want it to work.
Is that a bad thing not to want to worry about the infrastructure? Traditionally servers are designed around the concept of a physical server. We used to name servers by rack number or some other geographic location. Virtual machines were often named according to what physical server they resided within. Cloud technology, once the marketing speak is burned away and the APIs get to a mature and standard state (i.e., an in-house or an outside hosted cloud looks the same to an application), would allow other ways of managing the hundreds of thousands of machines in large data centers.
For example, capacity planning is a big deal. One of the responsibilities of a system engineer is to ensure that workloads can run properly on the servers. When there is a planned outage on one server or an increased load due to seasonal or scheduled work, the admins have to juggle the resources of the servers. In a planned outage we may use VMWare VMotion or Workload Migration and swing the workloads across. But then we often have to worry about IP changes, hostnames, virtual host software levels, etc.. With a properly configured internal cloud, this is a non-issue. I can literally click a button and remove a physical server from the cluster and it's completely transparent to end-users. Need to add capacity? I SAN-boot a cloned disk and the new server is automatically part of the cloud and ready to take on work.
We used to build our environments around managing discrete servers. Even if we had streamlined the process, it was still very much centered around the physical box. For example, we can stand up a box in a manner of minutes using RHEL kickstart, but if we wanted to add high availability this often meant configuring heartbeat IPs, swing SAN disks, /etc/hosts files for private IP ranges, etc.. HA on a cloud is almost too trivial to detail.
Of course it's not there yet, but it's where the more recent virtualization technologies was 5 years ago (and yeah, virtualilzation has been out for decades, but it has only within the past decade really surged).
:)
I am, as I type, building a RHEL6 with the latest kernel. So, based on previous compile times, I'd say about 2 more hours to go..
I know exactly what the problem is. Try passing different elevator options to the kernel at boot time. It helps with clock skew on virtual environments.
They will get cheaper. I picked up a 60G SSD from Newegg for under $100 after rebate. In a few months I expect that to be the normal price.
Is it worth it? Hell, yes. For systems where you need a lot of space or battery life isn't an issue, then they're probably not ideal. However, in a netbook they are amazing. I have a Samsung N120 with a 1.6GhZ Atom. With a standard HD, it was boggy. Resuming from suspend would take a minute. Launching apps would take 15 to 30 seconds. After installing the SSD it's like a new machine. Resume takes a few seconds. App launch times is a second or three. Browsing the web is snappier. I.e., anything that does multiple reads from the drive is much faster. If you replaced your standard laptop drive you may not notice it, but replacing a relatively slow HD in a netbook makes a huge difference. On top of it, my battery times climbed to at least 4 hours of constant use.
BTW, the SSDs run great with bcache/Linux. I'm putting together some benchmarks, but even before I run the numbers I can tell you that CentOS and Ubuntu on an Atom-based machine (a mini-pc form factor) runs incredibly.
I know that there are folks who enjoy wrestling as much as I enjoy sci-fi/fantasy but never made sense to me that they'd stick them both on the same channel. I don't know if they just needed *any* channel to air it and it was purely economical but I can imagine some exec thinking, "Well, the demographic for sci-fi/fantasy is mainly male. Therefore, they will like wrestling." I recently canceled all but the most basic cable and Internet after realizing that every time I wanted to watch something on sci-fi I found either some idiotic ghost busters reality show or wrestling.
"Never attribute to malice what can be explained by incompetence..."
The problem is likely some poor interface design. I've seen it used deliberately on some installers in order to sneakily add other products. It may follow a series of "Next" buttons that asks "Also install McAfee agent" or "Install Yahoo Toolbar"... In this case, the checkbox for the candidate may happen to be on the "Next" button of the previous screen.
It's a boring sentence trapped in a boring, verbose memo, so I found it a new home in a Philip K Dick story:
The movie was based on a Dick story, but that line was only in the screenplay (and modified by Rutger Hauer during filming)...
As to the corporate speak.. yeah, that's part of corporate culture. You have to remember that the people in the meetings are not always very well edu-macated. I remember attending meetings at UPS. Most of the managers were promoted through the ranks and started as truck drivers. Many had only a high school education. Many still spoke like dock workers (which is where they may have entered the company).
I was representing the TSG (Technology Support Group) and still remember a wonderful "discussion" that went something like this:
"The pre-load crew wants to know what time we're starting on Monday" (Monday being the first day of the Peak season, which was typically the day after Thanksgiving break).
"Fuck 'em. They'll come in we tell them."
"Well, yeah, but we need to let them know when to come in."
"Fuck you. You don't have a fucking clue."
And this was just a normal meeting.
Corporate speak reminds me of listening to some Jamaicans (and yes, I'm Jamaican and maybe as guilty). In Jamaica much of the population did not have formal education beyond the equivalent to 8th grade. But you know, the less you know, the more you think you know. It was always entertaining to hear some of the less educated ones using and mis-using words.
"We're going to creativate new jobs by learning the children about technological progress in the classroom. We are going to learn them about the computer and the CRT and the hard drive and the programming. We are going to elevate their understanding and invigorate the learning. So Mr. Accountant with his calculator and his writing implements and accounting books, I as a man and as a computer operator -- yes, computer operator -- demand that these funds and moneys go to the Kingston school district."
No, really, this was really what I heard... Corporate speak sounds the same to me....
There are some advantages to running KVM. There's a lot of flexibility in storage configuration and pretty much any host supported storage device will work. You also get some management and reporting capability. But yeah, I understand your point. At work I run a lot of PowerVM systems, actually 99% of our AIX environment is an LPAR, but even the VIO servers and HMCs run an OS. If you don't go the OS route then the hypervisor tends to be limited on what it can support. As to recoverability, using an OS means you can use all the OS tools. E.g., dd or rsync or LVM snapshots can be used to backup an OS image. With a pure hypervisor you tend to rely on exporting a guest VM.
Xen and paravirt is great. You can certainly cram a lot of instances into a Xen system. The downside is that your client OS must support that virtualization. Now Xen has made some strides with Windows and other non-Linux workloads, but many still need to run in full virt.
I'm running multiple VM systems including VMWare Server, VMWare ESX, VMWare Workstation, xen, KVM and VirtualBox.
VMWare Server is going away and sort of a pain to manage. However, it was free and worked decently. I have since replaced it with VMWare Workstation on my desktop and laptop systems. I use VirtualBox on my Mac laptop because it's free and was the easiest/cheapest to get going.
On my servers I am running VMWare ESX, xen and KVM on AMD systems (mostly dual core, but a couple quad core systems in the mix).
VMWare ESX was the most finicky as to installation but has been pretty simple to manage. The remote console options are simple. The VSphere management client is Windows only though. There is support for command line administration, but it's somewhat of a bear. You can script around it though and many people have done so and provide scripts online. Check out the VMWare community pages. Support is so so..
Xen was my workhorse for the longest time, but since my primary OS is RedHat/CentOS and RH is moving towards KVM, I've also been moving to KVM. The GUI management tools work fine, but are not as polished as VMWare ESX. However, it very much makes up for it in being able to do just about everything from the command line. I can deploy an image with a single command and this works wonderfully for testing. Performance is awesome with both xen and KVM. Well, the caveat is that some network intensive stuff seems to be bottlenecking somewhere, but it only has a single gigabit NIC across 8 VMs. I'll be adding another NIC in the next couple weeks and either bonding the adapter or just splitting them up.
Be aware that client/guest images generally do not have video acceleration so many games will fail to load. If you're running VMWare Workstation on a laptop, or the more recent KVMs then there is some measure of acceleration, but not 100%. Also, sound can be finicky especially across the network.
I know... 3D right now is a gimmick. It was a gimmick in the 50s too...
But they also said color was gimmickry. So were moving pictures. And sound.
The whole cloud concept is still being defined.. It's, um, cloudy....
That said, there are many reasons why some companies could benefit from the idea. For example, say that you want to build a global enterprise. Up until recently, this would take lots of money in order to have a physical "presence" in many different countries. With a global internet it's possible and it doesn't matter where the physical server is located as you can make the best use of local resources. Extend this concept a bit further and you realize that it doesn't matter where your physical server is located. Companies began to co-locate servers so they didn't need to have offices in Singapore to have a geographically ideal location for business, political, financial and even technical reasons.
The same challenges faced with local, company hosted resources are there with the cloud model. What happens when your company hosted server shuts down? what happens when your company network connections go down and your customers can't access your servers? What happens when an internal employee gets fired and sends out your customer list?
What the cloud concept can offer is reduced expense of hosting a large data center. For example, in the US land prices in areas where talent is located is generally high. It goes hand in hand. If a city is popular with technical folks then land prices will be high. The solution may be to host the personnel offices in one location and the servers somewhere in the cloud. Remember, building a data center is not a trivial thing. You need air conditioning, water, ducts, conduits in the floor, power, backup batteries and generators, fire suppression systems, etc..
No, clouds don't magically make standard IT problems disappear, but they can give a lot of flexibility in how you approach them.
I wonder what the laws are if you happen to forget the password? I use one-time passwords all the time. Some are 10 characters or more. I count on my ability to either reset the password or re-create the data. Politicians do it all the time. "It slipped my mind" or "It was ten years ago" or "I get so many papers that it's hard to remember what I signed".