Also, there are 'five nines' server companies here in the UK who would love to argue with your downtime argument regarding Windows servers. Five nines? So they never patch them?
Hierarchies are a horrible way to manage data, because no one "category" is always a subset of another. Pick the more general term here:
Pictures 2006 Christmas Trip
Believe it or not, hierarchical file system designers are actually aware of this, considered alternatives, and decided that hierarchical file systems (with a few workarounds) still are the way to go.
File systems are strange, complex beasts that basically cannot fail, cannot be slow, cannot leak memory, so on, so forth. Additional complexity in critical code such as this may lead to disaster. Just look at what Microsoft tried to achieve with WinFS. And how it failed.
Of course 'no one "category" is always a subset of another'. That's why real hierarchical file systems have workarounds to allow the notion of "multiple parents". And they come in two flavors.
If you want to express the idea that a file has several parents in the same file system, and this file is only to be removed when the last reference to it is removed, use hard links.
If you want to express the idea that a file is a reference (alias) to another file, use symbolic (soft) links. Symlinks become invalid when their target is removed, but they can span across file systems.
Disclaimer: For this to work, you have to use a real file system. Shortcuts (.lnk files) under Windows are not symlinks. They're not hard links either (you must use a dedicated API to resolve them). Vista shortcuts may or may not natively be able to emulate symlinks to folders, but definitely not for files.
Someone's competitor plans to launch a product with a 2% advantage over the product you can already get, mere years after something with a 100% advantage was demonstrated, and within only 8 months of something with 200% advantage!
Sorry, Sony. Those 2% are the proverbial straw that broke the camel's back. Come back next gen.
The previously capacity-challenged HD-DVD grows larger than its Blue-Ray rival, therefore eliminating the last remaining advantage or BR and more or less killing it in the short-to-medium term... Along with the PS3.
This just after HD-DVD encryption was broken? I have to get my tinfoil hat.
What kind of language would have resulted from passing variables of type int by value and objects by reference?
C#. And everything targetting the.NET platform.
Sorry, but I don't think I would want to program in that.
It's actually not that bad once you grasp the strange concepts of "value" and "reference" types. I.e. an int is a value type. A string is a value type. A class instance is a reference type, but a struct instance is a value type. Are we sufficiently confused yet ?
There's a common misunderstanding about stimulants like Adderall and Ritalin. They don't make you smarter or faster. They make you able to focus, and they make typically miserable tasks interesting. (Wiring database fields to GUI forms all day is boring, soul-crushing work, but well-paying, challenging jobs don't grow on trees.) They make you feel productive while performing the most menial tasks.
I am genuinely intrigued by this. You appear to be saying that programming is a menial task (or is it only VB programming? You know, you can wire DB fields to GUI forms all day long in C# too;)
First reaction: Are you sure you took the right job? Solving problems by writing some kind of obscure code that mere machines can understand should be a least a little entertaining even if you're forced to use VB *shrug*. Is it isn't, how do you manage deadlines and PHBs and retarded co-workers spitting out code like "If i = 0 Or i = 7 Or i = 14 Or i = 21 Or i = You_get_my_drift..."?
Second reaction: Oh, that's what the drug is for. Silly me.
More seriously, IMHO if you have to take drugs (okay, maybe except recreational ones, and even that I'm not so sure) in order to accomplish your job, you should change jobs. Maybe, somewhere on the way, you passed something that would have been exciting and entertaining to you, as well as make some money... Time to get back and find your True Function In Life (TM).
Disclaimer: I might not know what I'm talking about.
What if the parent process forks a gazillion child processes which never write, and the the parent process writes to a memory location? Do you then have to allocate memory for all gazillion child processes at once? That could suck.
I do not think you would do that, because it's not very logical. You would just duplicate the page used by the parent and let the gazillion children share the "original".
How many child processes never write to memory?
Don't know. Depends on you talking about heap memory or stack memory to begin with.
What exactly is it doing if not writing to memory?
You can do many things without ever writing to heap memory. Also note that we're talking about copying pages of memory on demand, so only the pages affected by a write are copied, and pages weighted 4K on x86 last time I checked. If you're only writing to stack memory then there are only a few memory pages impacted and COW should still win.
Does Copy on Write only save you memory if you fork a bunch of child processes that do nothing?
Yes. If all your child processes do nothing (say sleep(1000) in a tight loop), they only touch their stack (if the function call is not inlined), and only one memory page is copied. If they do nothing at all and exit immediately, exactly 0 pages are copied. If they exec() something, all the pages are dropped, but you didn't copy anything before.
Without COW, fork() implies duplicating all memory pages of the parent process - even swapped-out heap pages that the _parent_ itself never used for days...
Or is it a matter of saving the allocation of memory until the parent process is done, thus speeding up the parent process at the time of the fork?
It is a matter of duplicating memory only when one client of the shared pool writes to it, and duplicating for this client only.
Why not then copy on parent-write or child-execution, whichever comes first?
Copy on parent-write is a subset of COW. If you can copy onwrite, you can copy on parent-write. Copy on child-exec should not happen because it does not make sense to copy the memory of the parent right before discarding it to map another process.
So yes, COW is a good technique with fork()s. What is debated here is that Linus thinks it sucks for optimized I/O and said it on a bad day.
Also waiting for OO.o to build (to compare, several hundred Gentoo installs have been completed in the same time.
I use gentoo, I have been in the process of ugrading glibc on a mere 3 machines for 2 hours now, and I do not think you know what you're talking about:)
This is a dupe of almost the same story from the same source.
If you had read TFA, you would know they use Slashdot feeds as an entropy source for their one-time pads. They do report problems though, since during a recent test run they noticed 42% of their one-time pads were effectively equal...
Zimbra $70000 > Exchange $10600 -- Now as a CEO/CFO or company controller, which would you go with?
$10,000?
$70,000??
For email?!?
Please, pretty please, tell me you're just kidding. There are many FS/OSS alternatives. email is not supposed to cost $10,000 a year, and anyone who chokes $70,000 a year for email is... well...
At home, my wife and myself already use a wiki to keep track of things to do, current events, the progress of our baby, etc.
A wiki coupled with a spreadsheet would bring us: - An easier way to share tabular content or to format content in tabular form (file listings, tasks, events), - An easier way to balance accounts. Well, easier than GNUCash anyway:) - A collaborative framework to implement useful computations. If you consider formulas as functions, you have an easy-to-use language to create, use, or extend functions, coupled with a transparent revision control system.
Needless to say, since Visicalc was The First One, I'm more than interested in this idea, and I think that, properly handled, it has quite a potential.
1) Sell out to SCO, 2) Have Oracle buy the most feature-full database implementation we managed to get our hands on, 3) Piss off Open Source users, 4) Kill off PHP (since it's the only thing that still gets us going...) 5)... 6) DEFICIT!
Slashdot Mirror
Priceless. Thank you.
Believe it or not, hierarchical file system designers are actually aware of this, considered alternatives, and decided that hierarchical file systems (with a few workarounds) still are the way to go.
File systems are strange, complex beasts that basically cannot fail, cannot be slow, cannot leak memory, so on, so forth. Additional complexity in critical code such as this may lead to disaster. Just look at what Microsoft tried to achieve with WinFS. And how it failed.
Of course 'no one "category" is always a subset of another'. That's why real hierarchical file systems have workarounds to allow the notion of "multiple parents". And they come in two flavors.
If you want to express the idea that a file has several parents in the same file system, and this file is only to be removed when the last reference to it is removed, use hard links.
If you want to express the idea that a file is a reference (alias) to another file, use symbolic (soft) links. Symlinks become invalid when their target is removed, but they can span across file systems.
Disclaimer: For this to work, you have to use a real file system. Shortcuts (.lnk files) under Windows are not symlinks. They're not hard links either (you must use a dedicated API to resolve them). Vista shortcuts may or may not natively be able to emulate symlinks to folders, but definitely not for files.
Sorry, Sony. Those 2% are the proverbial straw that broke the camel's back. Come back next gen.
Are you kidding me?
The previously capacity-challenged HD-DVD grows larger than its Blue-Ray rival, therefore eliminating the last remaining advantage or BR and more or less killing it in the short-to-medium term... Along with the PS3.
This just after HD-DVD encryption was broken? I have to get my tinfoil hat.
Of course, there are always IBM's counterclaims, but it's unlikely there will be anything left after Novell is done.
10. Novell recently engaged in a partnership with Microsoft.
Go figure.
In your bathroom? Do you use Sony batteries or something?
It's actually not that bad once you grasp the strange concepts of "value" and "reference" types. I.e. an int is a value type. A string is a value type. A class instance is a reference type, but a struct instance is a value type. Are we sufficiently confused yet ?
You keep using that operator. I don't think it does what you think it does.
I think the GP refers to the new "secure" string functions in the VS2005 CRT API, such as strncpy_s (or _tcsncpy_s if you're into unicode).
Isn't Opera free already ?
they split into either overly opinionated engineers who fight there point of view to the death regardless of facts
Their. Not there. Please. Have mercy. Your UID is < 200000. You should be over 25 years old. Their. Please.
"I don't remember... I think I'm the third one"
.
I am genuinely intrigued by this. You appear to be saying that programming is a menial task (or is it only VB programming? You know, you can wire DB fields to GUI forms all day long in C# too
First reaction: Are you sure you took the right job? Solving problems by writing some kind of obscure code that mere machines can understand should be a least a little entertaining even if you're forced to use VB *shrug*. Is it isn't, how do you manage deadlines and PHBs and retarded co-workers spitting out code like "If i = 0 Or i = 7 Or i = 14 Or i = 21 Or i = You_get_my_drift..."?
Second reaction: Oh, that's what the drug is for. Silly me.
More seriously, IMHO if you have to take drugs (okay, maybe except recreational ones, and even that I'm not so sure) in order to accomplish your job, you should change jobs. Maybe, somewhere on the way, you passed something that would have been exciting and entertaining to you, as well as make some money... Time to get back and find your True Function In Life (TM).
God, my english is awful tonight.
90% of current programmers probably do not use those drugs, since they're overkill for Visual Basic coding...
Actually, I can't even think of any french word that wasn't borrowed from english that has a W in it.
"Wagon" does not come from english AFAIK and is pronounced "vagon" nonetheless.
Disclaimer: I might not know what I'm talking about.
What if the parent process forks a gazillion child processes which never write, and the the parent process writes to a memory location? Do you then have to allocate memory for all gazillion child processes at once? That could suck.
I do not think you would do that, because it's not very logical. You would just duplicate the page used by the parent and let the gazillion children share the "original".
How many child processes never write to memory?
Don't know. Depends on you talking about heap memory or stack memory to begin with.
What exactly is it doing if not writing to memory?
You can do many things without ever writing to heap memory. Also note that we're talking about copying pages of memory on demand, so only the pages affected by a write are copied, and pages weighted 4K on x86 last time I checked. If you're only writing to stack memory then there are only a few memory pages impacted and COW should still win.
Does Copy on Write only save you memory if you fork a bunch of child processes that do nothing?
Yes. If all your child processes do nothing (say sleep(1000) in a tight loop), they only touch their stack (if the function call is not inlined), and only one memory page is copied. If they do nothing at all and exit immediately, exactly 0 pages are copied. If they exec() something, all the pages are dropped, but you didn't copy anything before.
Without COW, fork() implies duplicating all memory pages of the parent process - even swapped-out heap pages that the _parent_ itself never used for days...
Or is it a matter of saving the allocation of memory until the parent process is done, thus speeding up the parent process at the time of the fork?
It is a matter of duplicating memory only when one client of the shared pool writes to it, and duplicating for this client only.
Why not then copy on parent-write or child-execution, whichever comes first?
Copy on parent-write is a subset of COW. If you can copy onwrite, you can copy on parent-write. Copy on child-exec should not happen because it does not make sense to copy the memory of the parent right before discarding it to map another process.
So yes, COW is a good technique with fork()s. What is debated here is that Linus thinks it sucks for optimized I/O and said it on a bad day.
Also waiting for OO.o to build (to compare, several hundred Gentoo installs have been completed in the same time.
:)
I use gentoo, I have been in the process of ugrading glibc on a mere 3 machines for 2 hours now, and I do not think you know what you're talking about
This is a dupe of almost the same story from the same source.
If you had read TFA, you would know they use Slashdot feeds as an entropy source for their one-time pads. They do report problems though, since during a recent test run they noticed 42% of their one-time pads were effectively equal...
Zimbra $70000 > Exchange $10600 -- Now as a CEO/CFO or company controller, which would you go with?
$10,000?
$70,000??
For email?!?
Please, pretty please, tell me you're just kidding. There are many FS/OSS alternatives. email is not supposed to cost $10,000 a year, and anyone who chokes $70,000 a year for email is... well...
You want a use case? I'll give you a use case.
:)
At home, my wife and myself already use a wiki to keep track of things to do, current events, the progress of our baby, etc.
A wiki coupled with a spreadsheet would bring us:
- An easier way to share tabular content or to format content in tabular form (file listings, tasks, events),
- An easier way to balance accounts. Well, easier than GNUCash anyway
- A collaborative framework to implement useful computations. If you consider formulas as functions, you have an easy-to-use language to create, use, or extend functions, coupled with a transparent revision control system.
Needless to say, since Visicalc was The First One, I'm more than interested in this idea, and I think that, properly handled, it has quite a potential.
1) Sell out to SCO, ...
2) Have Oracle buy the most feature-full database implementation we managed to get our hands on,
3) Piss off Open Source users,
4) Kill off PHP (since it's the only thing that still gets us going...)
5)
6) DEFICIT!
So we got two problems, not only do the protocols not scale well, but the prices as well...