Esxi - is vmware's Type 1 bare metal hypervisor vSphere - is the windows centric management system that allows you to centerally manage a whole pile of esxi boxes. vSphere client - is the client that can connect either to a vsphere server or directly to an esxi box and let you manage everything.
vmware player / fusion / workstation / server are vmware's type 2 hypervisor products. I know player and server are free. Workstation and Fusion require a license.
As for vmware being a falling giant, I don't think so. More like a giant walking through very thick mud. As a product goes, esxi is great. However, vsphere and the client are a bloated mess. Their licensing model is at best confusing and at worst intentionally confusing. I also find it boggling that they're constantly renaming everything. However, like a lot of things, in a lot of organizations it's gained glacial momentum and shifting to something else will be hard.
I'm hearing you say "clusterfs" but what I'm reading from your post is "remotely recoverable filesystem". A cluster filesystem makes lots and lots of sense if you've got a 100 nodes that need high speed access to a single piece of storage.... this doesn't sound like this application...
What you should setup is a ZFS box (I'm a fan of raid10, but pick the raid of your choice) we'll call this machine A. Now go build an indentical box named "B".
Now, what you'll want to do is setup an rsync process that does the following... 1. send a zfs command from A to B creating a snapshot of the appropriate file systems. 2. rsync said filesystem from A to B 3. Sleep some amount of time, goto step 1. Now at some point you'll want to clean up all the snapshots on B, but that's an excersie I'll leave to the reader.
Another option is to take the snapshot on A and then use zfs send to send the snapshots as well.
Your fears are misplaced. Firstly I won't get into the "big sky" theory. Secondly, in most cases the FAA and pertinent government agencies are notified.
If you have a rogue trader who games the system, you can look at UBS and say "geez, I guess you'll be investing in a better risk management system!"
But if you have a good risk management system that throws alarms and nobody looks at them, or follows up on them, then it's all on their heads.
They only had to look over one of their borders into France to see what a rogue trader could do. This isn't a novel problem, rogue traders taking positions, then losing money and then taking crazier positions to get back what they lost isn't a new problem.
People have stopped thinking about the spirit of a rule and applying it appropriately, but just following the letter of the rule blindly with no sense towards common sense.
This poster had the word "kill" in it, therefore it must go. If the other tenured professors had any balls, they'd put this poster on every door in campus and tell the police to actually do their job and catch real criminals.
Well in Germany, they have speed detection cameras that read your license plate and just mail you a ticket... It's already your idea only without the danger of the ticket accidentally flying out the window...
I'm not saying that these benchmarks are bunk, I'd just like to see these same benchmarks with Gentoo/Fedora/etc. It's been my experience that while the user experience of Ubuntu is generally good, they turn on every bell/whistle.
Think about the huge carbon costs in moving large volumes of paper from one part of the country to another part of the country. In this day of high speed networks, this is foolish. Kill the existing system and start over. The post office needs to reinvent itself as a leaner, more focused organization. People have to get it into their heads there is a real cost to moving crap all over the place.
Here's what they need to do.
1. Go down to 2 or 3 day delivery. 2. Ditch the bulk mail crap. Nobody wants credit card offers and all the other tripe. 3. Close the small town post offices, roll them into an existing establishment, or replace them with a drop box. For the Holidays you could even roll up a more elaborate van that would have someone in it that could handle larger packages.
The currently system doesn't make any sense in the digital age.
Correlation does not imply causation. There have now been studies conducted using the medical records of 100k+ children and there is no correlation between getting the MMR vaccine and onset of autism. The percentage of children presenting autistic spectum disorders were statistically identical between those who were vaccinated and those who were not. In short there is no scientific correlation, so not getting your child vaccinated because of the fear of "giving" them autism is bunk. Note that the original study by Andrew Wakefield was so flawed that the British government revoked his medical license.
Yes, there are a small number of children who have severe side effects because of the vaccinations. However, we are not in a federalized healthcare system. I find it very hard to believe that you couldn't cold call labs until you found a lab that would take your blood sample and run some specific tests on it.
Now, contrast that with the possible side effects of getting mumps, measles and rubella. These include such things as infertility, deafness, blindness, damage to organs, etc. Yes, those side effects are in the single digit percentages, but then again, so are the percentages for allergic reaction.
What they should have done is take the coupon out and replace it with one of their own. GameStop relies on the goodwill of their customers... this damages this good will.
I was making an off color joke about the instability of the vSphere client... though now that I've had to explain the joke, it's not nearly as funny...
I usually can only destroy 10 or so vm's before my vsphere client runs out of memory / handles or just segfaults for the fun of it. Needless to say, my displeasure with that vpshere client has caused me to become somewhat of a vsphere command line ninja.
Firstly, it appears this guy was treated poorly and not only is he a nitwit, it would appear that most of his coworkers/management were as well.
Secondly, it's acts of sabotage like this that make it hard for the rest of us to do our jobs.
Thirdly, on a not so serious note... wi-fi from McDonalds? vSphere console? How did he think he was NOT going to get caught? Did he even try to wipe the logs off the vsphere server? Had this guy two brain cells in his head, he could have obliterated their infrastructure and not left a trace of evidence.
This is why, while I own a kindle... I'm yet to actually buy a book from amazon. I've bought plenty of other books from reputable sources in epub format and converted them to mobi. In a nutshell, when I buy a book I want to actually *own* it.
Also, you can do cool things like we did, which was to systematically replace our 1TB drives for 2TB and ZFS magically saw double the space without having to reboot!
I'll start by saying this most assuredly was a government job. Either done by the US, Israel or Russia.
1. There's obviously a spy somewhere. Iran isn't going to make public the intimate details of their reprocessing plants, let alone the exact configuration of their control terminals / PLC controllers and centrifuges. You need hard data for that. Who helped Iran build these plants? Who designed this particular cascade process?
2. People who have a seriously intimate knowledge of this type of hardware had to be involved. It's one thing to say "If there's a motor attached, double it's frequency" and then let the thing burn out. It's a whole other thing to say "up the motor by 20mhz for 50 minutes" knowing it would introduce subtle failures that would be argued away as poor components, overuse, etc. Also, what does that do to the quality of the uranium coming out of the process? Maybe the plan was to not only break the plant but corrupt the output as well. I can't imagine this type of knowledge is wide spread...
3. What was/is the end game? Iran (while it'll never say it) wants the bomb. They want parity with Israel or at least the argument of MAD. I think possibly stuxnet might have had some end game, but barring that, it was a delaying tactic.
We had a real situation where a person in our group actually had a cerebral hemorrhage and tragically passed away. I should point out as not to sully this guy, since he a great fellow that he wasn't hording information intentionally the company merely had a poor system for documentation. Unfortunately, this person never fully documented their procedures and it took us some ugly trial and error to figure what he were doing. Out of that tragically, we established a very through and cross verified documentation system. Part of the system we called "musical chairs" where we'd switch functions for a week (usually during the winter) and we made it a point not todo hands on cross training but to rely entirely on the documentation.
Needless to say, the first couple of rounds discovered some HUGE holes in our documentation, but we fixed it and it worked. On a side note, when the company a couple years later decided to become ISO9001 compliant, at least the IS department passed with flying colors.
IF only Lucas had decided to make a bunch of offshoot movies based around this... oh well. They're a great read. Aaron Allston and Timothy Zhan are great writers.
Actually there are naive in kernel ZFS modules. There was a bunch of work done by the Lawrence Livermore National Laboratory to do a clean port. Currently the vdev layer is totally solid. Their using that as back end block storage for a lustre filesystem. However the guy who did the port has also released the posix layer as well.
So, I can't speak to the licensing issues, but there is native ZFS support. Granted it's got a ways to go to be completely stable (from the posix layer level) but the vdev stuff works great.
Actually btrfs merely uses lvm and md devices on the underside. You can still run the lvscan / lvdisplay / mdadm, however what you'll see will look rather obtuse.
Firstly, this wasn't the primary, but one of several redundant backup systems. Granted any redundant system not fully tested is not to be considered tested.
Secondly, the NRC has a long and storied history of letting nuclear plants run with known issues based on the promises that they'd be fixed. Now that they're in the spotlight because of Fukishima they're doing this shocking thing and actually calling plants on issues that have been long standing.
Thirdly, as a country we need to take a honest look at our existing nuclear plants. They're old. We've made HUGE advancements in nuclear power (just look at any reactor on a navy vessel) What we need to do is use that knowledge to either reengineer our existing reactors or look to replace them in place with better reactors.
Fourthly, we need to take an honest look at our nuclear fuel cycle, which is retarded. We need to start reprocessing fuel, not just storing it in dry casks. There is a huge amount of wasted energy not being extracted from those rods.
Same here... we've benched out i7's vs. non i7 (prior generation, i.e Q6600) and for the same clock rate, we're seeing a constant 20% performance increase.
Slashdot Mirror
Esxi - is vmware's Type 1 bare metal hypervisor
vSphere - is the windows centric management system that allows you to centerally manage a whole pile of esxi boxes.
vSphere client - is the client that can connect either to a vsphere server or directly to an esxi box and let you manage everything.
vmware player / fusion / workstation / server are vmware's type 2 hypervisor products. I know player and server are free. Workstation and Fusion require a license.
As for vmware being a falling giant, I don't think so. More like a giant walking through very thick mud. As a product goes, esxi is great. However, vsphere and the client are a bloated mess. Their licensing model is at best confusing and at worst intentionally confusing. I also find it boggling that they're constantly renaming everything. However, like a lot of things, in a lot of organizations it's gained glacial momentum and shifting to something else will be hard.
I'm hearing you say "clusterfs" but what I'm reading from your post is "remotely recoverable filesystem". A cluster filesystem makes lots and lots of sense if you've got a 100 nodes that need high speed access to a single piece of storage.... this doesn't sound like this application...
What you should setup is a ZFS box (I'm a fan of raid10, but pick the raid of your choice) we'll call this machine A.
Now go build an indentical box named "B".
Now, what you'll want to do is setup an rsync process that does the following...
1. send a zfs command from A to B creating a snapshot of the appropriate file systems.
2. rsync said filesystem from A to B
3. Sleep some amount of time, goto step 1.
Now at some point you'll want to clean up all the snapshots on B, but that's an excersie I'll leave to the reader.
Another option is to take the snapshot on A and then use zfs send to send the snapshots as well.
Your fears are misplaced. Firstly I won't get into the "big sky" theory. Secondly, in most cases the FAA and pertinent government agencies are notified.
If you have a rogue trader who games the system, you can look at UBS and say "geez, I guess you'll be investing in a better risk management system!"
But if you have a good risk management system that throws alarms and nobody looks at them, or follows up on them, then it's all on their heads.
They only had to look over one of their borders into France to see what a rogue trader could do. This isn't a novel problem, rogue traders taking positions, then losing money and then taking crazier positions to get back what they lost isn't a new problem.
People have stopped thinking about the spirit of a rule and applying it appropriately, but just following the letter of the rule blindly with no sense towards common sense.
This poster had the word "kill" in it, therefore it must go. If the other tenured professors had any balls, they'd put this poster on every door in campus and tell the police to actually do their job and catch real criminals.
Well in Germany, they have speed detection cameras that read your license plate and just mail you a ticket... It's already your idea only without the danger of the ticket accidentally flying out the window...
I'm not saying that these benchmarks are bunk, I'd just like to see these same benchmarks with Gentoo/Fedora/etc.
It's been my experience that while the user experience of Ubuntu is generally good, they turn on every bell/whistle.
Give me a cut down box running xfce any day.
Think about the huge carbon costs in moving large volumes of paper from one part of the country to another part of the country. In this day of high speed networks, this is foolish. Kill the existing system and start over. The post office needs to reinvent itself as a leaner, more focused organization. People have to get it into their heads there is a real cost to moving crap all over the place.
Here's what they need to do.
1. Go down to 2 or 3 day delivery.
2. Ditch the bulk mail crap. Nobody wants credit card offers and all the other tripe.
3. Close the small town post offices, roll them into an existing establishment, or replace them with a drop box. For the Holidays you could even roll up a more elaborate van that would have someone in it that could handle larger packages.
The currently system doesn't make any sense in the digital age.
Correlation does not imply causation. There have now been studies conducted using the medical records of 100k+ children and there is no correlation between getting the MMR vaccine and onset of autism. The percentage of children presenting autistic spectum disorders were statistically identical between those who were vaccinated and those who were not. In short there is no scientific correlation, so not getting your child vaccinated because of the fear of "giving" them autism is bunk. Note that the original study by Andrew Wakefield was so flawed that the British government revoked his medical license.
Yes, there are a small number of children who have severe side effects because of the vaccinations. However, we are not in a federalized healthcare system. I find it very hard to believe that you couldn't cold call labs until you found a lab that would take your blood sample and run some specific tests on it.
Now, contrast that with the possible side effects of getting mumps, measles and rubella. These include such things as infertility, deafness, blindness, damage to organs, etc. Yes, those side effects are in the single digit percentages, but then again, so are the percentages for allergic reaction.
Okay, I learned about TRESOR, that's cool. Also, running firefox as a different user is an old trick I've been using for a long time.
However, I live by a basic rule that's served me well. Laptops are fundamentally weak places to keep data.
What they should have done is take the coupon out and replace it with one of their own. GameStop relies on the goodwill of their customers... this damages this good will.
Don't forget that Boeing is still making the 737... As of June 2001, 2,109 orders are yet to be fulfilled... I'd say their doing a-okay.
I was making an off color joke about the instability of the vSphere client... though now that I've had to explain the joke, it's not nearly as funny...
I usually can only destroy 10 or so vm's before my vsphere client runs out of memory / handles or just segfaults for the fun of it. Needless to say, my displeasure with that vpshere client has caused me to become somewhat of a vsphere command line ninja.
Firstly, it appears this guy was treated poorly and not only is he a nitwit, it would appear that most of his coworkers/management were as well.
Secondly, it's acts of sabotage like this that make it hard for the rest of us to do our jobs.
Thirdly, on a not so serious note... wi-fi from McDonalds? vSphere console? How did he think he was NOT going to get caught? Did he even try to wipe the logs off the vsphere server? Had this guy two brain cells in his head, he could have obliterated their infrastructure and not left a trace of evidence.
This is why, while I own a kindle... I'm yet to actually buy a book from amazon. I've bought plenty of other books from reputable sources in epub format and converted them to mobi. In a nutshell, when I buy a book I want to actually *own* it.
Also, you can do cool things like we did, which was to systematically replace our 1TB drives for 2TB and ZFS magically saw double the space without having to reboot!
Thanks! It's not often that I get actual reasoned responses around here. That's interesting about the Bushehr plant...
Hopefully what will come out of all of this is that people will get paranoid about their SCADA network... I know, crazy talk, but who knows.
I'll start by saying this most assuredly was a government job. Either done by the US, Israel or Russia.
1. There's obviously a spy somewhere. Iran isn't going to make public the intimate details of their reprocessing plants, let alone the exact configuration of their control terminals / PLC controllers and centrifuges. You need hard data for that. Who helped Iran build these plants? Who designed this particular cascade process?
2. People who have a seriously intimate knowledge of this type of hardware had to be involved. It's one thing to say "If there's a motor attached, double it's frequency" and then let the thing burn out. It's a whole other thing to say "up the motor by 20mhz for 50 minutes" knowing it would introduce subtle failures that would be argued away as poor components, overuse, etc. Also, what does that do to the quality of the uranium coming out of the process? Maybe the plan was to not only break the plant but corrupt the output as well. I can't imagine this type of knowledge is wide spread...
3. What was/is the end game? Iran (while it'll never say it) wants the bomb. They want parity with Israel or at least the argument of MAD. I think possibly stuxnet might have had some end game, but barring that, it was a delaying tactic.
We had a real situation where a person in our group actually had a cerebral hemorrhage and tragically passed away. I should point out as not to sully this guy, since he a great fellow that he wasn't hording information intentionally the company merely had a poor system for documentation. Unfortunately, this person never fully documented their procedures and it took us some ugly trial and error to figure what he were doing. Out of that tragically, we established a very through and cross verified documentation system. Part of the system we called "musical chairs" where we'd switch functions for a week (usually during the winter) and we made it a point not todo hands on cross training but to rely entirely on the documentation.
Needless to say, the first couple of rounds discovered some HUGE holes in our documentation, but we fixed it and it worked. On a side note, when the company a couple years later decided to become ISO9001 compliant, at least the IS department passed with flying colors.
Michael Stackpole has also written a whole bunch of great X-wing books as well.
IF only Lucas had decided to make a bunch of offshoot movies based around this... oh well. They're a great read. Aaron Allston and Timothy Zhan are great writers.
Actually there are naive in kernel ZFS modules. There was a bunch of work done by the Lawrence Livermore National Laboratory to do a clean port. Currently the vdev layer is totally solid. Their using that as back end block storage for a lustre filesystem. However the guy who did the port has also released the posix layer as well.
So, I can't speak to the licensing issues, but there is native ZFS support. Granted it's got a ways to go to be completely stable (from the posix layer level) but the vdev stuff works great.
http://zfsonlinux.org/
Actually btrfs merely uses lvm and md devices on the underside. You can still run the lvscan / lvdisplay / mdadm, however what you'll see will look rather obtuse.
Firstly, this wasn't the primary, but one of several redundant backup systems. Granted any redundant system not fully tested is not to be considered tested.
Secondly, the NRC has a long and storied history of letting nuclear plants run with known issues based on the promises that they'd be fixed. Now that they're in the spotlight because of Fukishima they're doing this shocking thing and actually calling plants on issues that have been long standing.
Thirdly, as a country we need to take a honest look at our existing nuclear plants. They're old. We've made HUGE advancements in nuclear power (just look at any reactor on a navy vessel) What we need to do is use that knowledge to either reengineer our existing reactors or look to replace them in place with better reactors.
Fourthly, we need to take an honest look at our nuclear fuel cycle, which is retarded. We need to start reprocessing fuel, not just storing it in dry casks. There is a huge amount of wasted energy not being extracted from those rods.
Same here... we've benched out i7's vs. non i7 (prior generation, i.e Q6600) and for the same clock rate, we're seeing a constant 20% performance increase.