Meanwhile, I've been happily using my 750 GB Momentus XT at home (I have 256GB SSD at work), quite happily as a trade off between capacity and performance. To get the 750 GB i want in my laptop would cost me about 6x the price and would certainly not give me 6x the performance. It's also 1/4 of a mortgage payment back on my mortgage.
Because I want 750 GB of space, and that's quarter of a mortgage payment. I can get 750 GB which performs "well enough" to not piss me off like a totally mechanical drive for ~15% of that.
You say that, but I guarantee you they will do a deal with Dell, Lenovo, HP or another supplier and will sell millions of these drives. What some nerds on slashdot think will have pretty much zero bearing on sales. Even Winmodems were huge, and they actually had major performance problems and were hated by the nerd community.
Yeah, write cache is turned off in Windows by default for good reason: if you had say 20 GB worth of writes in RAM which hadn't yet hit the disk, guess what happens when the power drops or your OS crashes? RAM is fine for READ cache, but caching writes, dubious. Which is why enterprise level storage gear uses NVRAM, battery backup, etc. And even if you turn on write caching, if an application forces the filesystem to sync and verify the data was written, it will be flushed anyhow.
Eventually, you need to write to permanent storage. More RAM can help performance but it can't safely do much for writes.
Eventually, you need to write to disk. And yes, RAM is faster than SSD, and yes, it will continue to be used as cache. Also - you need to GET THE DATA INTO RAM. This won't just be operating system startup, but will also be application startup. Also, this can be used to cache writes (turned OFF by default in Windows, for good reason) - which, if you do so in RAM is leaving you open to significant data loss if you were to have a power outage, operating system crash, etc.
I just use mount points which are even available under xp (i think) and later windows installs to map out the bigger files.
Heaps of people seem to think this is the same thing... but it is an extremely inefficient way of utilizing your space - if you have a 2GB file with only say, 100 MB of which is "hot" then you are moving the entire 2GB file.
A controller / driver / intelligent operating system can determine which BLOCKS in the file need to be cached, so in the above example, instead of consuming 2GB of SSD, the caching would consume 100 MB for that file - leaving the other 1.9 GB free to cache something else.
Also, you are not able to re-shuffle what is cached, on the fly while using the system as performance requirements change. A controller/driver or intelligent OS can do this. So, say you're in a game that wouldn't fit on your SSD due to you putting other stuff there, and it needs to frequently read/write from part of a data file as you traverse a level. The controller/OS can cache that hot part of the file while you're in the game, automatically.
If you were shuffling the data around yourself, manually you just couldn't get that optimisation.
Yup. Precisely why I have a hybrid 750 in my laptop. I dual boot, and have the drive split 500/250 for OS X and Windows (gaming). I can spin up virtual machines for anything else (Linux, BSD, etc.) on the OS X side. The drive cost me about 130 bucks. Currently, a 500gb SSD would cost me near on 450-500 Australian, and I still wouldn't really have enough space to do what I am doing today.
Depends on the life expectancy of the SSD part. Hybrid drives are generally using SLC NAND which has a far higher life expectancy than MLC consumer grade NAND used in regular SSDs. If the NAND fails on the Seagate hybrid drives, it supposedly reverts back to regular hard drive performance.
Exactly. And in terms of "knowing" what to put in the cache, the drive knows just as well as anything else what should be there - because it knows what sectors have been hot. The filesystem on top is pretty irrelevant.
I've been pretty happy with my Seagate Momentus XT 750, which works this way. I'm not sure if they do write caching or not yet, but that would clearly help, as would some more NAND. Early days yet, but if seagate can get say, 32GB of NAND and write caching on a new 2TB drive, sign me up.
netsh is not a powershell cmdlet. write-output is explicitly for converting objects into a screen-readable format, and is used on the end of a pipe. any decent sized organisation will already have a CA infrastructure in place for stuff like 802.11x and WIFI. lack of convenience in Linux for implementing signing is pretty irrelevant... that is a Linux flaw, not a powershell problem.
As to WINE, whether it works or not is irrelevant. When you're paying for a support contract for your mission critical app, you run it on the platform the vendor supports. If that is Linux, fine run it on Linux. If it is Windows, you run it on Windows.. NONE of our business apps will officially support WINE.
If you want to fuck about trying to diagnose a problem caused by a needed application update breaking compatibility with your unsupported platform while the vendor says "works on our supported platform", be my guest.
In the real world, IT do not make the purchasing decisions on things like engineering software. The person actually doing the engineering job chooses the tool and it is IT's job to make it work.
Whether or not alternatives to Minecad, Mine24d, Pitram, etc exist on Linux is irrelevant. They have not been chosen by the business as the best tool for the job.
As I said, we run multiple platforms in house, and run Linux where it is appropriate. It is not appropriate, or most cost effective for ALL jobs, and for some of the applications the business has decided to run, credible Linux alternatives DO NOT EXIST.
Comparing uptimes is pretty irrelevant when the Linux machine can not run the application required.
Having a working knowledge of a variety of platforms is a good idea. Given that the vulnerability seems to be common to all these servers when run on Linux, now would be a prime time to flick services over to FreeBSD (or something else) until the root cause has been determined.
That no one knows the actual method of infection yet though is a fair cause for concern; it means you can't effectively defend against it when running that platform.
All pipelines in powershell are objects, and something will only become text if it has become serialised somewhere along the way (e.g., read/write from disk).
Returning multiple objects from a query is a feature, not a bug. Just because it works differently, does not mean it is "wrong". It is what it is, and if you're expecting it to behave like other scripting languages I can see why you would have problems.
Yes, it has some warts, it's clearly a v1.x product. And no, a cmd script can NOT turn code-signing off if you have UAC turned on (as you should) or the user doesn't have administrative access (as they shouldn't).
If you have an enterprise CA, and an AD integrated subordinate, signing scripts is trivial. My workflow with writing powershell involves saving the file in notepad++, running "./add-signature foo.ps1" (add-signature is a 1-2 line script I've got which retrieves my code-signing cert and signs the argument with it) and then running the script. That's it.
And yes, it does provide a level of protection - because I can leave code-signing turned on, push the authorised programmer certs to the enterprise and not have to worry about people running malicious powershell scripts. If someone in the enterprise pushes a dodgy script (either accidentally, or maliciously), I can easily disable it from running, wherever the copies may exist by simply revoking the cert.
But, there are things you can do in Powershell in a few lines of code that you just can't do with a shell script without a massive time investment.
Getting hardware and software info out of the machine is trivial. For example. Much of the power of powershell lies in its ability to easily query and interact with WMI objects, which VB could do but was a complete and utter pain in the arse.
Netapp management/reporting tools for my FAS - require a Windows server
Run our banking software
Run our safety management system
Run our enterprise business intelligence system
Run our Mine production system
Manage our vSphere environment - though this is getting better, many of the management tools run on Windows server, including vCenter
Distribute updates to our windows clients (there are applications we need that are Windows only - mostly specialist mining and financial applications)
Run above mentioned Windows specific mining and financial applications
Can, but nowhere near as easily - run Solarwinds for enterprise network monitoring. Yes, I've used zabbix, nagios, etc. They are WAY, WAY more work
Reliably log into a secure WIFI network
Just for starters...
And don't even consider mentioning running the apps in Wine... it's nowhere near there yet - and if i was to call say, the vendor of Mine24d and say "yeah, there's some wierd graphical bug, i'm running it in Wine under Ubuntu. We need to get this 100 million dollar mine design project out the door this evening", I'd get precisely zero support.
In the real world, getting real production done, Windows is often a requirement. Where it is not needed, or not efficient, I don't run it. But there are places where it is FAR less work keeping the Windows machines I need maintained than trying to hack the app to run on something else.
If you've got major problems keeping a Windows network running smoothly in 2013, the problem is the admin, not the platform (though it does have its issues no doubt, like everything else).
Learn what Powershell can do and get back to us. It's an entirely different, object-based shell and is far superior in some ways to anything available in Unix land (including Linux and OS X). For a start, you can code-sign your scripts to ensure they haven't been modified and come from trusted sources. You can manipulate any wmi object directly from within the shell. There's a lot that's crap in the Windows world, but powershell is interesting and "good". Learn it, and it will make you a more employable person.
Besides, in the real world, you run the appropriate platform for the applications the business requires. We have a mixed environment with FreeBSD, Linux and Windows machines doing various jobs. Linux can not do everything.
Server 2012 uses Metro. If you don't want to deal with slow RDP, use Powershell (lol. yes I know, but that's the official strategy. powershell itself isn't that bad, but there are heaps of server applications which don't support no-gui mode).
Failure by what metric? Apple is one of the most profitable companies in the world, Mac sales growth is oustripping Windows/PC in percentage terms and have been for about 5+ years now.
Not everybody drives a Ferrari, yet they aren't failing either. Apple don't have to outsell the PC to "succeed".
Windows 7 is supported until 2020. Enterprises have plenty of time up their sleeves to wait this hair-brained fuckwittery out and/or move to another platform.
Slashdot Mirror
Meanwhile, I've been happily using my 750 GB Momentus XT at home (I have 256GB SSD at work), quite happily as a trade off between capacity and performance. To get the 750 GB i want in my laptop would cost me about 6x the price and would certainly not give me 6x the performance. It's also 1/4 of a mortgage payment back on my mortgage.
But then they get blamed for bugs. Fuck that, that's microsoft's job.
Because I want 750 GB of space, and that's quarter of a mortgage payment. I can get 750 GB which performs "well enough" to not piss me off like a totally mechanical drive for ~15% of that.
You say that, but I guarantee you they will do a deal with Dell, Lenovo, HP or another supplier and will sell millions of these drives. What some nerds on slashdot think will have pretty much zero bearing on sales. Even Winmodems were huge, and they actually had major performance problems and were hated by the nerd community.
Yeah, write cache is turned off in Windows by default for good reason: if you had say 20 GB worth of writes in RAM which hadn't yet hit the disk, guess what happens when the power drops or your OS crashes? RAM is fine for READ cache, but caching writes, dubious. Which is why enterprise level storage gear uses NVRAM, battery backup, etc. And even if you turn on write caching, if an application forces the filesystem to sync and verify the data was written, it will be flushed anyhow.
Eventually, you need to write to permanent storage. More RAM can help performance but it can't safely do much for writes.
Eventually, you need to write to disk. And yes, RAM is faster than SSD, and yes, it will continue to be used as cache. Also - you need to GET THE DATA INTO RAM. This won't just be operating system startup, but will also be application startup. Also, this can be used to cache writes (turned OFF by default in Windows, for good reason) - which, if you do so in RAM is leaving you open to significant data loss if you were to have a power outage, operating system crash, etc.
Heaps of people seem to think this is the same thing... but it is an extremely inefficient way of utilizing your space - if you have a 2GB file with only say, 100 MB of which is "hot" then you are moving the entire 2GB file.
A controller / driver / intelligent operating system can determine which BLOCKS in the file need to be cached, so in the above example, instead of consuming 2GB of SSD, the caching would consume 100 MB for that file - leaving the other 1.9 GB free to cache something else.
Also, you are not able to re-shuffle what is cached, on the fly while using the system as performance requirements change. A controller/driver or intelligent OS can do this. So, say you're in a game that wouldn't fit on your SSD due to you putting other stuff there, and it needs to frequently read/write from part of a data file as you traverse a level. The controller/OS can cache that hot part of the file while you're in the game, automatically.
If you were shuffling the data around yourself, manually you just couldn't get that optimisation.
Yup. Fusion is essentially a JBOD with the operating system "aware" of the different performance characteristics of drives in the JBOD.
No, it's not actually.
Yup. Precisely why I have a hybrid 750 in my laptop. I dual boot, and have the drive split 500/250 for OS X and Windows (gaming). I can spin up virtual machines for anything else (Linux, BSD, etc.) on the OS X side. The drive cost me about 130 bucks. Currently, a 500gb SSD would cost me near on 450-500 Australian, and I still wouldn't really have enough space to do what I am doing today.
Depends on the life expectancy of the SSD part. Hybrid drives are generally using SLC NAND which has a far higher life expectancy than MLC consumer grade NAND used in regular SSDs. If the NAND fails on the Seagate hybrid drives, it supposedly reverts back to regular hard drive performance.
Exactly. And in terms of "knowing" what to put in the cache, the drive knows just as well as anything else what should be there - because it knows what sectors have been hot. The filesystem on top is pretty irrelevant.
I've been pretty happy with my Seagate Momentus XT 750, which works this way. I'm not sure if they do write caching or not yet, but that would clearly help, as would some more NAND. Early days yet, but if seagate can get say, 32GB of NAND and write caching on a new 2TB drive, sign me up.
netsh is not a powershell cmdlet. write-output is explicitly for converting objects into a screen-readable format, and is used on the end of a pipe. any decent sized organisation will already have a CA infrastructure in place for stuff like 802.11x and WIFI. lack of convenience in Linux for implementing signing is pretty irrelevant... that is a Linux flaw, not a powershell problem.
As to WINE, whether it works or not is irrelevant. When you're paying for a support contract for your mission critical app, you run it on the platform the vendor supports. If that is Linux, fine run it on Linux. If it is Windows, you run it on Windows.. NONE of our business apps will officially support WINE.
If you want to fuck about trying to diagnose a problem caused by a needed application update breaking compatibility with your unsupported platform while the vendor says "works on our supported platform", be my guest.
In the real world, IT do not make the purchasing decisions on things like engineering software. The person actually doing the engineering job chooses the tool and it is IT's job to make it work.
Whether or not alternatives to Minecad, Mine24d, Pitram, etc exist on Linux is irrelevant. They have not been chosen by the business as the best tool for the job.
As I said, we run multiple platforms in house, and run Linux where it is appropriate. It is not appropriate, or most cost effective for ALL jobs, and for some of the applications the business has decided to run, credible Linux alternatives DO NOT EXIST.
Comparing uptimes is pretty irrelevant when the Linux machine can not run the application required.
Having a working knowledge of a variety of platforms is a good idea. Given that the vulnerability seems to be common to all these servers when run on Linux, now would be a prime time to flick services over to FreeBSD (or something else) until the root cause has been determined.
That no one knows the actual method of infection yet though is a fair cause for concern; it means you can't effectively defend against it when running that platform.
All pipelines in powershell are objects, and something will only become text if it has become serialised somewhere along the way (e.g., read/write from disk).
Returning multiple objects from a query is a feature, not a bug. Just because it works differently, does not mean it is "wrong". It is what it is, and if you're expecting it to behave like other scripting languages I can see why you would have problems.
Yes, it has some warts, it's clearly a v1.x product. And no, a cmd script can NOT turn code-signing off if you have UAC turned on (as you should) or the user doesn't have administrative access (as they shouldn't).
If you have an enterprise CA, and an AD integrated subordinate, signing scripts is trivial. My workflow with writing powershell involves saving the file in notepad++, running "./add-signature foo.ps1" (add-signature is a 1-2 line script I've got which retrieves my code-signing cert and signs the argument with it) and then running the script. That's it.
And yes, it does provide a level of protection - because I can leave code-signing turned on, push the authorised programmer certs to the enterprise and not have to worry about people running malicious powershell scripts. If someone in the enterprise pushes a dodgy script (either accidentally, or maliciously), I can easily disable it from running, wherever the copies may exist by simply revoking the cert.
But, there are things you can do in Powershell in a few lines of code that you just can't do with a shell script without a massive time investment.
Getting hardware and software info out of the machine is trivial. For example. Much of the power of powershell lies in its ability to easily query and interact with WMI objects, which VB could do but was a complete and utter pain in the arse.
Run things like
Just for starters...
And don't even consider mentioning running the apps in Wine... it's nowhere near there yet - and if i was to call say, the vendor of Mine24d and say "yeah, there's some wierd graphical bug, i'm running it in Wine under Ubuntu. We need to get this 100 million dollar mine design project out the door this evening", I'd get precisely zero support.
In the real world, getting real production done, Windows is often a requirement. Where it is not needed, or not efficient, I don't run it. But there are places where it is FAR less work keeping the Windows machines I need maintained than trying to hack the app to run on something else.
If you've got major problems keeping a Windows network running smoothly in 2013, the problem is the admin, not the platform (though it does have its issues no doubt, like everything else).
Learn what Powershell can do and get back to us. It's an entirely different, object-based shell and is far superior in some ways to anything available in Unix land (including Linux and OS X). For a start, you can code-sign your scripts to ensure they haven't been modified and come from trusted sources. You can manipulate any wmi object directly from within the shell. There's a lot that's crap in the Windows world, but powershell is interesting and "good". Learn it, and it will make you a more employable person.
Besides, in the real world, you run the appropriate platform for the applications the business requires. We have a mixed environment with FreeBSD, Linux and Windows machines doing various jobs. Linux can not do everything.
Actually they're around 10%. which is significantly more than Linux.
Yeah and I'm sure their "response" will be "Upgrades from Windows XP or Windows 7 are not supported".
Server 2012 uses Metro. If you don't want to deal with slow RDP, use Powershell (lol. yes I know, but that's the official strategy. powershell itself isn't that bad, but there are heaps of server applications which don't support no-gui mode).
Yet their mobile web browsing share (and thus, potential as a mobile web application platform) is far in excess of any other competitor.
Failure by what metric? Apple is one of the most profitable companies in the world, Mac sales growth is oustripping Windows/PC in percentage terms and have been for about 5+ years now.
Not everybody drives a Ferrari, yet they aren't failing either. Apple don't have to outsell the PC to "succeed".
Windows 7 is supported until 2020. Enterprises have plenty of time up their sleeves to wait this hair-brained fuckwittery out and/or move to another platform.