The Itanium 1 was a concept chip, the second one was meant to go into general use.. it has failed, so people label it a concept chip to try and save face..
Actually i had a lot of console problems on more modern machines.. Especially when switching between X and the console. Some machines wouldn`t do svga textmodes, and forced me to use framebuffer.. other machines would either corrupt the console, or crash X (and drop me to a corrupted console) when trying to switch between X11 and the console.
Also the appeal of using an old machine is that you dont spend any money, when the hd or the psu dies... you dont spend $40-60 replacing it, you get a free replacement from a dumpster.
But it is a hardware raid5 solution, with custom hardware to perform the checksumming and such, so as far as the computer is concerned it`s just got a big fast SCSI disk attached to it.
Not to mention the fact that SUN ditched their BSD-based OS (SunOS 4.0 and earlier) to create a SysV based OS (Solaris, AKA SunOS 5.x) that would be better suited to their newer multiprocessor systems. As for journaling and UFS... Linux has had UFS support for years, but it`s not so widely used.. if it was as brilliant as you say, people wouldnt have bothered making journaled filesystems for linux.
So, he wanted you to work for him for free adding features that may or may not have actual interest for you... Notice how he asked you, rather than making the changes himself, as the gpl allows, and then submitting patches for you to make your own decision about.
Exactly, static content, plenty of cache.. no problem. It`s dynamic content that causes the load, and worse.. the excessive use people make of dynamic content and databases.. for instance i have seen many sites where things which should be static, such as images, downloadable files, etc... are served from a database, this just causes totally unnecessary load on the server.. It`s harder to cache, consumes more ram when doing so, and requires more processor time to process the request. We had a similar pornsite, one stats server running various cgi scripts designed to record hits and referrers so we knew which of our affiliates was sending us the most traffic etc, and setting cookies on the visitors so that our affiliates can see when were sending content to them. The other simply served static content, the main index pages, the porn images themselves. The static server was a P2/400 and spent most of its time idle, the dynamic server was a dual P3/800 and was usually pretty heavily loaded, despite generating about 0.5% of the network traffic the static server did.
No, that graph looks pretty indicative of a slashdotting... an initial burst when the story is first posted, and then a steady stream of hits while the story is still on the front page.. which dies down once it`s been relegated to "yesterday`s news"
35 seconds to load on your dialup? 35 seconds to render on your 486? 35 seconds to transfer from the other side of the world over a congested backbone? there could be many reasons why it takes so long to load... in my case the content is downloaded much quicker than the machine (a humble 250mhz) can render it.
Infact, the SOFTWARE emulation of x86 that`s used by the alpha could probably beat the x86 emulation abilities of the itanium, and i`d like to see how the virtualpc programs for macs (and i think solaris/sparc?) stack up. A 64bit arch that offers compatibility with x86 through emulation, but which beats the itanium`s performance, could prove successfull.
Well, your point is very true.. intel has no competition at the low end (read: x86) market, their chips have much higher clock rates than AMD offerings, and are edging ahead in speed now, but the high clock is most effective in selling them to the masses. Where they need to develop and compete, is at the high end market, where they have a rather lackluster product of their own, the itanium... which is being completely blown away by alpha in the raw performance stakes, i think sparc and power4 might be nudging ahead of it too.. But when you consider the poor compiler and application support for itanium right now, they REALLY fall behind the others... And as has been stated before, itanium should never have existed... hp should be concentrating on the alpha, which already has the software support, performance and reputation that itanium is still striving for.
But an improparly designed app should not take down the os of the machine it`s running on, let alone other machines networked to it.. that IS poor os design, exposed by equally poor app design.
The NT kernel was a very nice design to start with... when it was ripped/stolen from DEC that is, since microsoft got their hands on it.. it has gone downhill, from a clean microkernel architecture to a so-called microkernel architecture thats bigger than most monolithic kernels, with legacy code bolted on for backwards compatibility, things shoehorned into the kernel that shouldnt be there, or atleast there should be the choice.. like display drivers, and even chunks of code designed to speed up a web browser..
Ofcourse we can really trust the ability of the openbsd team, despite their line by line auditing MANY bugs have gotten through, think ftpd, flaws in the kernel itself, openssl, openssh especially and nodoubt other things i dont remember. So despite all their claims, which are no better than microsoft-style fud/hype (yet people champion them because its opensource???) they STILL let vulns through. Infact openssh is probably among the top 5 for numbers of security vulns discovered in a single application, especially recently. All this false advertising and unfounded claims, ala microsoft, are what make groups like "gobbles" and nodoubt others hate them so much.
And it should be even more important for a thief to erase the stolen disks he`s selling, incase the buyer recovers enough info to find who really owned the disks, and tells him where he bought the disk from.
Actually ripping old computers apart, especially servers with lots of cards and scsi disks.. is often more profitable than selling them as complete machines.
Which is nodoubt why he asked your consulting firm to dispose of them, rather than just disposing of them with all the other garbage.. he was assuming you would do the data destruction for him.
Well, not to mention the amiga with 4096 colors in the same resolution as original VCD`s/mpeg.. looks ok from a distance on a tv set.. and i`m sure MIT had some high end machines which could do 1280*1024 in 24bit... i`m sure SGI made such hardware in the 80`s
Low level formats are a standard feature of SCSI drives tho, theyre far less standardized and reliable for IDE drives. Infact, the Amiga user manual used to warn you against low level formatting ide drives (the ide support in later amiga`s was presented to the OS via scsi emulation, so that existing scsi tools would work with it.. so the partitioning/formatting tool couldnt tell them apart)
Now that is a nice idea.. i have a few old scsi drives, only 7200 rpm.. theyre dead but they do spin up (fucked controller i guess) If i can get them running from a +12V car battery, i will give it a shot
Slashdot Mirror
The Itanium 1 was a concept chip, the second one was meant to go into general use.. it has failed, so people label it a concept chip to try and save face..
I had a similar hack for amigaos, you could mount an ftp site as if it was a regular volume,same as a cd or a floppy etc.
Actually i had a lot of console problems on more modern machines.. Especially when switching between X and the console. Some machines wouldn`t do svga textmodes, and forced me to use framebuffer.. other machines would either corrupt the console, or crash X (and drop me to a corrupted console) when trying to switch between X11 and the console. Also the appeal of using an old machine is that you dont spend any money, when the hd or the psu dies... you dont spend $40-60 replacing it, you get a free replacement from a dumpster.
But it is a hardware raid5 solution, with custom hardware to perform the checksumming and such, so as far as the computer is concerned it`s just got a big fast SCSI disk attached to it.
Not to mention the fact that SUN ditched their BSD-based OS (SunOS 4.0 and earlier) to create a SysV based OS (Solaris, AKA SunOS 5.x) that would be better suited to their newer multiprocessor systems.
As for journaling and UFS... Linux has had UFS support for years, but it`s not so widely used.. if it was as brilliant as you say, people wouldnt have bothered making journaled filesystems for linux.
Sure i did, and last i checked windows 95 didnt have very comprehensive support for raid hardware and no software-raid atall.
So, he wanted you to work for him for free adding features that may or may not have actual interest for you... Notice how he asked you, rather than making the changes himself, as the gpl allows, and then submitting patches for you to make your own decision about.
But what would the drives after number 24 (Z) be called?
Exactly, static content, plenty of cache.. no problem.
It`s dynamic content that causes the load, and worse.. the excessive use people make of dynamic content and databases.. for instance i have seen many sites where things which should be static, such as images, downloadable files, etc... are served from a database, this just causes totally unnecessary load on the server.. It`s harder to cache, consumes more ram when doing so, and requires more processor time to process the request.
We had a similar pornsite, one stats server running various cgi scripts designed to record hits and referrers so we knew which of our affiliates was sending us the most traffic etc, and setting cookies on the visitors so that our affiliates can see when were sending content to them. The other simply served static content, the main index pages, the porn images themselves. The static server was a P2/400 and spent most of its time idle, the dynamic server was a dual P3/800 and was usually pretty heavily loaded, despite generating about 0.5% of the network traffic the static server did.
No, that graph looks pretty indicative of a slashdotting... an initial burst when the story is first posted, and then a steady stream of hits while the story is still on the front page.. which dies down once it`s been relegated to "yesterday`s news"
35 seconds to load on your dialup? 35 seconds to render on your 486? 35 seconds to transfer from the other side of the world over a congested backbone? there could be many reasons why it takes so long to load... in my case the content is downloaded much quicker than the machine (a humble 250mhz) can render it.
Infact, the SOFTWARE emulation of x86 that`s used by the alpha could probably beat the x86 emulation abilities of the itanium, and i`d like to see how the virtualpc programs for macs (and i think solaris/sparc?) stack up. A 64bit arch that offers compatibility with x86 through emulation, but which beats the itanium`s performance, could prove successfull.
Well, your point is very true.. intel has no competition at the low end (read: x86) market, their chips have much higher clock rates than AMD offerings, and are edging ahead in speed now, but the high clock is most effective in selling them to the masses.
Where they need to develop and compete, is at the high end market, where they have a rather lackluster product of their own, the itanium... which is being completely blown away by alpha in the raw performance stakes, i think sparc and power4 might be nudging ahead of it too.. But when you consider the poor compiler and application support for itanium right now, they REALLY fall behind the others...
And as has been stated before, itanium should never have existed... hp should be concentrating on the alpha, which already has the software support, performance and reputation that itanium is still striving for.
But an improparly designed app should not take down the os of the machine it`s running on, let alone other machines networked to it.. that IS poor os design, exposed by equally poor app design.
The NT kernel was a very nice design to start with... when it was ripped/stolen from DEC that is, since microsoft got their hands on it.. it has gone downhill, from a clean microkernel architecture to a so-called microkernel architecture thats bigger than most monolithic kernels, with legacy code bolted on for backwards compatibility, things shoehorned into the kernel that shouldnt be there, or atleast there should be the choice.. like display drivers, and even chunks of code designed to speed up a web browser..
Ofcourse we can really trust the ability of the openbsd team, despite their line by line auditing MANY bugs have gotten through, think ftpd, flaws in the kernel itself, openssl, openssh especially and nodoubt other things i dont remember. So despite all their claims, which are no better than microsoft-style fud/hype (yet people champion them because its opensource???) they STILL let vulns through.
Infact openssh is probably among the top 5 for numbers of security vulns discovered in a single application, especially recently.
All this false advertising and unfounded claims, ala microsoft, are what make groups like "gobbles" and nodoubt others hate them so much.
And it should be even more important for a thief to erase the stolen disks he`s selling, incase the buyer recovers enough info to find who really owned the disks, and tells him where he bought the disk from.
Actually ripping old computers apart, especially servers with lots of cards and scsi disks.. is often more profitable than selling them as complete machines.
Large numbers of used hard drives purchased from ebay with fraudulent credit cards!
Which is nodoubt why he asked your consulting firm to dispose of them, rather than just disposing of them with all the other garbage.. he was assuming you would do the data destruction for him.
Except you LOSE the potential money you could have gained from selling the drive on ebay, so its just a win win, assuming a lose cancels out a win
Well, not to mention the amiga with 4096 colors in the same resolution as original VCD`s/mpeg.. looks ok from a distance on a tv set.. and i`m sure MIT had some high end machines which could do 1280*1024 in 24bit... i`m sure SGI made such hardware in the 80`s
Low level formats are a standard feature of SCSI drives tho, theyre far less standardized and reliable for IDE drives.
Infact, the Amiga user manual used to warn you against low level formatting ide drives (the ide support in later amiga`s was presented to the OS via scsi emulation, so that existing scsi tools would work with it.. so the partitioning/formatting tool couldnt tell them apart)
Not to mention the risk of overwriting your deleted data, either with your swapfile, or with norton utilities themselves when you install it.
Now that is a nice idea.. i have a few old scsi drives, only 7200 rpm.. theyre dead but they do spin up (fucked controller i guess)
If i can get them running from a +12V car battery, i will give it a shot