Xen is indeed the platform on which Amazon ec2 is based, but I would say that decision might be made differently today if given the chance. First off, at the time when it was being created, Xen was the only real option. KVM simply didn't exist until after ec2's launch in 2006, and the only real alternative was VMware. Second, ec2 initially used Paravirtualization exclusively, meaning hardware-assited virtualization functionality wasn't required or used, and performance was extremely good, but compatibility was limited to Xen-aware Linux systems.
Fast forward to today and KVM-enabled kernels are performance-comparable to Paravirtualized Xen instances, good Windows drivers exist for KVM virtualized hardware, and maybe most importantly, KVM is part of the kernel of all major Linux distributions.
Using virt-manager to configure a Host's network interfaces to bridging is pretty straightforward. I've got a 5 node KVM cluster at home home office, running Ubuntu 12.04 LTS, managing them via virt-manager on Ubuntu 12.10. They are all operating in bridge mode.
Now, a feature that doesn't quite work as advertised, doing a live migration (without shared storage) requires an extra step that isn't really made clear, you have to create the destination disk image (same size in bytes) before it will migrate.
If you're leaving it as ChromeOS, you'll regret buying the Acer. Its a lot heavier, requires active cooling, and the battery life is ass. Basically, its a cheap x86 laptop, with the normal issues of a cheap x86 laptop. The little bit of speed you pick up will pale in comparison to how much nicer the Samsung is to use in the real world. She won't notice that Facebook loads.001 seconds quicker, but she will notice her lap getting warm, the fan running constantly, and having to take her power cord with her every time she brings it somewhere.
You do need to expose your shares as HTTP, but you probably don't need to wrap them in a flash player. The built-in media player supports several formats, h264 included. I use FreeNAS as my file server and a few minutes with that got me a nice web accessible, directory-indexed media share. It isn't perfect but it does the job with minimal fuss.
And what Chromebooks can do that tablets can't do: Replace traditional laptops. You can actually do real work with these things.
Sure, the high res screen would be nice. If I could ask for a single upgrade for my Samsung Chromebook, that would be it.. But the last few Chromebooks released have been disappointing to me, and I'm skeptical that this would be any better.
I'd say Samsung got it right with the Exynos CPU, and ChromeOS runs very, very well on ARM. Between my home and business we've replaced 3 laptops with Chromebooks, but it seems like the Samsung model is the only one that I would even consider. If you want a cheap feeling, heavy, loud, low battery life laptop, there's plenty of those to choose from. If you want something that runs cool, runs for long time, doesn't annoy you with fans, and doesn't burn your lap, then the Samsung is your only pick.
Google, if you want to release a "powerhouse" chromebook, try the Exynos quad core, and throw in some more memory. Exynos supports 40 bit memory addressing, so the 4G barrier doesn't really exist in a practical sense (32 bit address space exposed to each processes, chrome is multi-process).
Try the +1 front sprocket. Its like adding a gear, and was a big win for mine MPG-wise. It took me a while to addjust to it though, I spent months just riding 10-15mph faster.
My friend, you are mistaken. You will not find 1 Gbit much less 10 or 100Gbit unless the "right place" to live is a datacenter. Average internet speeds from July last year are still in the single-digit Megabit range. Local storage is several orders of magnitude faster than Internet for the foreseeable future.
MySQL is just one example.
This comparison gives you Linux, Windows XP, and OSX on the same Core Duo machine. These benchmarks do almost no I/O, so your filesystem complaints are moot.
They also did one using the G5. Same story.
Not to say that the filesystem in freebsd doesn't hurt it, or that threading in freebsd doesn't hurt it, but you can't blame it for everything.
No.
Mach is a microkernel. OSX implements a BSD kernel as one server running under this Mach kernel.
It has the worst parts of both worlds! This is one of the reasons why server benchmarks on OSX are so bad, compared to linux on the same box.
The empty can rattles the most.. The people with the least understanding of this topic seem to be the ones with the most to say. 95% of them don't even understand the title. Maybe its some crazy brain-fuck where "undervolting" looks like "underclocking".
I see 3 posters, including the parent, that understand the story. For the rest of you:
#1. Had you read the damn article, you wouldn't make yourselves look like such fools. #2. Had you read the title, you wouldn't make yourselves look like such fools. #3. Had you read the Slashdot summary, you wouldn't make yourselves look like such fools.
Unless, of course, you are a retard, in which case you should go back to humping something inanimate.
Underclocking is what you do to make your CPU run slower. Sometimes this means it runs cooler. SpeedStep, PowerNow! LongRun, etc are all forms of dynamic underclocking. Your windows control panel, BIOS power saving settings, etc are all forms of dynamic underclocking. This is NOT what this story is about. If you mentioned any of these things in your post, someone really should mod you OFF TOPIC.
Undervolting is what you do to make your CPU use less power at a given speed. You can usually get a good percentage drop before you hit instability, as long as you aren't overlocking. This will cause your system to run cooler, and your battery (on a laptop) to last longer,l without giving up performance!/me prepares to be modded down or pointing out the obvious
The UltraSparc processor was introduced to the world in November of 1995. This is around the same time as the Pentium 133, and just befor ethe Pentium Pro 150.
You're going to try to tell me that a CPU that is closing on its 10th anniversery isn't "wasted" running linux, because it has real work to do?
Or will you tell me that the original UltraSparc isn't a real UltraSparc?
What real business are you likely to conduct on a Sun Ultra1 with a 167mhz processor and 64 megs of blistering 60ns memory? Holding down papers? Holding open doors? You certainly won't be running Solaris 10 on it, as it is not an officially supported system.
Back to reality.... Current generation UltraSparc processors are not competitive with Intel/AMD processors until you have more than a handful of said processors in a single image system. If you need to scale your system beyond about 8 cpus, you'll want to look at UltraSparc.
Besides, Solaris is expressly written for UltraSparc about as accurately as Linux is written expressly for UltraSparc. It started off on Motorola 68ks, then 32 bit sparcs, then x86, then 64 bit sparcs, then x86 again, then x86_64.
This guy designed a fairly basic system for fuel injection around a 68HC908 processor. Most people using it have converted pre-fi cars to fi using it, but a fair amount have replaced other FI systems with it.
Where are my mod points when I need them? Finally, someone who gets it!
Theft can only apply to physical things. If I steal your CD, I have stolen your CD. If I copy your CD, I have (potentially) violated copyright law, but that is not Theft.
Come on Slashdot, you're supposed to be the "smart", "enlightened" crowd. Stop embracing ignorance!
They interviewed me after promising the client that they had a candidate, and presented me as though I had a long association with them. Not the most honest relationship, but my employer took the right-to-hire option and I'm happy with the turnout.
Head hunters are opportunists, and in my opinion are as straightforward as your average marketing department.
Unfortunately, you can't look at it that way in the real world. An ISP's responsibility to provide connectivity is its highest priority. They don't care about Blaster Joe or Typhoid Annie, and shouldn't be expected to put up any safeguards against them spreading thier joy. However, in the case of the "good" worm, you're pretty much stuck with having to block certain traffic in a big ISP. The wonderful pings that it sends, coming from hundreds or even thousands of users on the same termination router, basicly turns into a massive ARP storm. This is enough to bring a device (which usually runs at 30-50% capacity) to the rev limiter, hindering the subscribers' ability to pass legitimate traffic, and creates a level of instability on the units (Hoorray for malloc errors in IOS!)
An ISP shouldn't keep you from visiting tubgirl, goatse, or nambla, or doing whatever you want to do.. They should be prepared, and expected, to block a DOS attack, even if it is unwittingly coming from thier own users.
Having worked for both, I agre. Privately held companies are driven by one thing: Generating Profit. Publicly traded companies are driven by something else: Stock Price. Publicly traded companies change thier focus from development of newer/better/cheaper product or service, to generating press releases showing partnerships with other publicly traded companies that may or may not actually have a worthwhile product/service. Publicly traded companies are just fine with skipping a partnership and announcing layoffs to achieve the same goal.
This seems to be a relatively new function for business, but one that has been true in my experiences. Hats off to Google for fighting the urge to IPO.
I work at a cable company, and I stress this exact point. It is a silly way to cap bandwidth in the first place, and it was crazy to ever be adopted. Maybe this is why Docsis is only an ad-hoc standard? The bad part is, the method of enforcing speeds employed by most (I stress MOST, you bet your ass that my methods aren't so easily fooled) cable operators has the same problem. They want to get your speed by SNMP query to your cable modem. Which again puts the trust in the client. While I haven't seen any SNMP faker hacks, I'm sure that they aren't too far behind. Another silly note is that most of those guys are comparing your speed to a list of approved speeds, not to a list of what customers bought what. This includes thier Business lines, which run over the same gear. You won't be able to sneak through with a 2meg/2meg pipe, but a 1.5meg/768k is a service they probobly sell, and would get right through thier checks.
The control method that you will see soon is called "shared secret", and is an encrypted passphrase-type method. Basicly, your cable modem gets a config file that has a key in it, which is basicly a signature of the bin file. It then generates a new passkey based on those two items, and send it to the CMTS. The CMTS verifies that it got a correct passkey, and then lets you connect. The encryption they used is junk, though, and there are efforts underway to break it. This is yet another dumb method that will only work for a short while!
What I will say is that there is a better way, and it is 100% effective. Your cable modem doesn't just "make up" a speed and magicly work, it has to register its rates with the CMTS. This is where the speed is truly controlled. While it isn't likely that Cisco will have a good method for capping individual users at the CMTS level, they are nice enough to tell you what speed someone is registered at. This is the method that I am using, and I *am* comparing speeds against what customers are paying for... So if you live in a town where you can get Imo's pizza, the square beyond compare, this is your warning!:)
On the flip side, once an abuser is identified, the info gets sent to marketing, and who knows what happens from there. We don't just pull the plug on abusers (yet).
The moron at wininformant added all exploits for all linux distributions together. Image the obvious scenerio, where bind8.x.x has a root compromise. This would only count as a single exploit, however the article counts it once for eash distribution that acknowledged it.
If you look at the charts yourself, you see that Win2k had 42 exploits in 2001. In comparison, SuSe had 21. Redhat had 54. OpenBSD had 14. The figures also are not focused on a particular release. I would expect that the numbers would be substantially lower if it only look into account the current releases. Suprise, SuSE still publishes security announcements for 6.x in addition to 7.x, and those are counted.
THe author of the atricle need to look up Aggregate and try writing an article again.
Slashdot Mirror
I wonder what percentage of Slashdot users will even get that joke...
Xen is indeed the platform on which Amazon ec2 is based, but I would say that decision might be made differently today if given the chance. First off, at the time when it was being created, Xen was the only real option. KVM simply didn't exist until after ec2's launch in 2006, and the only real alternative was VMware. Second, ec2 initially used Paravirtualization exclusively, meaning hardware-assited virtualization functionality wasn't required or used, and performance was extremely good, but compatibility was limited to Xen-aware Linux systems. Fast forward to today and KVM-enabled kernels are performance-comparable to Paravirtualized Xen instances, good Windows drivers exist for KVM virtualized hardware, and maybe most importantly, KVM is part of the kernel of all major Linux distributions.
Using virt-manager to configure a Host's network interfaces to bridging is pretty straightforward. I've got a 5 node KVM cluster at home home office, running Ubuntu 12.04 LTS, managing them via virt-manager on Ubuntu 12.10. They are all operating in bridge mode. Now, a feature that doesn't quite work as advertised, doing a live migration (without shared storage) requires an extra step that isn't really made clear, you have to create the destination disk image (same size in bytes) before it will migrate.
If you're leaving it as ChromeOS, you'll regret buying the Acer. Its a lot heavier, requires active cooling, and the battery life is ass. Basically, its a cheap x86 laptop, with the normal issues of a cheap x86 laptop. The little bit of speed you pick up will pale in comparison to how much nicer the Samsung is to use in the real world. She won't notice that Facebook loads .001 seconds quicker, but she will notice her lap getting warm, the fan running constantly, and having to take her power cord with her every time she brings it somewhere.
the Chromebook has been Amazon's best selling laptop for a while now. That doesn't mean its outselling all Windows 8 PCs, but at least on Amazon its outselling any particular Windows 8 PC. Take that as you will, since the Apple TV is also on that page. http://www.amazon.com/Best-Sellers-Electronics/zgbs/electronics/ref=zg_bs_electronics_home_all
You do need to expose your shares as HTTP, but you probably don't need to wrap them in a flash player. The built-in media player supports several formats, h264 included. I use FreeNAS as my file server and a few minutes with that got me a nice web accessible, directory-indexed media share. It isn't perfect but it does the job with minimal fuss.
And what Chromebooks can do that tablets can't do: Replace traditional laptops. You can actually do real work with these things.
Sure, the high res screen would be nice. If I could ask for a single upgrade for my Samsung Chromebook, that would be it.. But the last few Chromebooks released have been disappointing to me, and I'm skeptical that this would be any better.
I'd say Samsung got it right with the Exynos CPU, and ChromeOS runs very, very well on ARM. Between my home and business we've replaced 3 laptops with Chromebooks, but it seems like the Samsung model is the only one that I would even consider. If you want a cheap feeling, heavy, loud, low battery life laptop, there's plenty of those to choose from. If you want something that runs cool, runs for long time, doesn't annoy you with fans, and doesn't burn your lap, then the Samsung is your only pick.
Google, if you want to release a "powerhouse" chromebook, try the Exynos quad core, and throw in some more memory. Exynos supports 40 bit memory addressing, so the 4G barrier doesn't really exist in a practical sense (32 bit address space exposed to each processes, chrome is multi-process).
Try the +1 front sprocket. Its like adding a gear, and was a big win for mine MPG-wise. It took me a while to addjust to it though, I spent months just riding 10-15mph faster.
Advertised latency of SSDs are Much Lower than 5ms, in the order of 70-ish MICROseconds, not MILLIseconds.
My friend, you are mistaken. You will not find 1 Gbit much less 10 or 100Gbit unless the "right place" to live is a datacenter. Average internet speeds from July last year are still in the single-digit Megabit range. Local storage is several orders of magnitude faster than Internet for the foreseeable future.
MySQL is just one example. This comparison gives you Linux, Windows XP, and OSX on the same Core Duo machine. These benchmarks do almost no I/O, so your filesystem complaints are moot. They also did one using the G5. Same story. Not to say that the filesystem in freebsd doesn't hurt it, or that threading in freebsd doesn't hurt it, but you can't blame it for everything.
No. Mach is a microkernel. OSX implements a BSD kernel as one server running under this Mach kernel. It has the worst parts of both worlds! This is one of the reasons why server benchmarks on OSX are so bad, compared to linux on the same box.
The empty can rattles the most.. The people with the least understanding of this topic seem to be the ones with the most to say. 95% of them don't even understand the title. Maybe its some crazy brain-fuck where "undervolting" looks like "underclocking".
/me prepares to be modded down or pointing out the obvious
I see 3 posters, including the parent, that understand the story. For the rest of you:
#1. Had you read the damn article, you wouldn't make yourselves look like such fools.
#2. Had you read the title, you wouldn't make yourselves look like such fools.
#3. Had you read the Slashdot summary, you wouldn't make yourselves look like such fools.
Unless, of course, you are a retard, in which case you should go back to humping something inanimate.
Underclocking is what you do to make your CPU run slower. Sometimes this means it runs cooler. SpeedStep, PowerNow! LongRun, etc are all forms of dynamic underclocking. Your windows control panel, BIOS power saving settings, etc are all forms of dynamic underclocking. This is NOT what this story is about. If you mentioned any of these things in your post, someone really should mod you OFF TOPIC.
Undervolting is what you do to make your CPU use less power at a given speed. You can usually get a good percentage drop before you hit instability, as long as you aren't overlocking. This will cause your system to run cooler, and your battery (on a laptop) to last longer,l without giving up performance!
Please...
The UltraSparc processor was introduced to the world in November of 1995. This is around the same time as the Pentium 133, and just befor ethe Pentium Pro 150.
You're going to try to tell me that a CPU that is closing on its 10th anniversery isn't "wasted" running linux, because it has real work to do?
Or will you tell me that the original UltraSparc isn't a real UltraSparc?
What real business are you likely to conduct on a Sun Ultra1 with a 167mhz processor and 64 megs of blistering 60ns memory? Holding down papers? Holding open doors? You certainly won't be running Solaris 10 on it, as it is not an officially supported system.
Back to reality.... Current generation UltraSparc processors are not competitive with Intel/AMD processors until you have more than a handful of said processors in a single image system. If you need to scale your system beyond about 8 cpus, you'll want to look at UltraSparc.
Besides, Solaris is expressly written for UltraSparc about as accurately as Linux is written expressly for UltraSparc. It started off on Motorola 68ks, then 32 bit sparcs, then x86, then 64 bit sparcs, then x86 again, then x86_64.
I call bullshit. E450 supports a maximum of 4G ram.
But don't take my word for it.
Check out MegaSquirt http://www.bgsoflex.com/megasquirt.html
This guy designed a fairly basic system for fuel injection around a 68HC908 processor. Most people using it have converted pre-fi cars to fi using it, but a fair amount have replaced other FI systems with it.
Where are my mod points when I need them? Finally, someone who gets it!
Theft can only apply to physical things. If I steal your CD, I have stolen your CD. If I copy your CD, I have (potentially) violated copyright law, but that is not Theft.
Come on Slashdot, you're supposed to be the "smart", "enlightened" crowd. Stop embracing ignorance!
They interviewed me after promising the client that they had a candidate, and presented me as though I had a long association with them. Not the most honest relationship, but my employer took the right-to-hire option and I'm happy with the turnout.
Head hunters are opportunists, and in my opinion are as straightforward as your average marketing department.
Unfortunately, you can't look at it that way in the real world. An ISP's responsibility to provide connectivity is its highest priority. They don't care about Blaster Joe or Typhoid Annie, and shouldn't be expected to put up any safeguards against them spreading thier joy. However, in the case of the "good" worm, you're pretty much stuck with having to block certain traffic in a big ISP. The wonderful pings that it sends, coming from hundreds or even thousands of users on the same termination router, basicly turns into a massive ARP storm. This is enough to bring a device (which usually runs at 30-50% capacity) to the rev limiter, hindering the subscribers' ability to pass legitimate traffic, and creates a level of instability on the units (Hoorray for malloc errors in IOS!)
An ISP shouldn't keep you from visiting tubgirl, goatse, or nambla, or doing whatever you want to do.. They should be prepared, and expected, to block a DOS attack, even if it is unwittingly coming from thier own users.
Having worked for both, I agre. Privately held companies are driven by one thing: Generating Profit. Publicly traded companies are driven by something else: Stock Price.
Publicly traded companies change thier focus from development of newer/better/cheaper product or service, to generating press releases showing partnerships with other publicly traded companies that may or may not actually have a worthwhile product/service. Publicly traded companies are just fine with skipping a partnership and announcing layoffs to achieve the same goal.
This seems to be a relatively new function for business, but one that has been true in my experiences. Hats off to Google for fighting the urge to IPO.
Squash
Hardly.
Were you around when the AMD K7 was release? If so, you would remember, and I quote:
August 18, 1999 As a matter of fact, there's only a minority of motherboard manufacturers that are actually offering Athlon-platforms right now, the minority is either quiet, or even announcing that they will not produce Athlon-motherboards for rather cheesy reasons.
Or,
August 23, 1999
Now we should hope that finally more motherboard makers will supply Athlon-boards soon, so that those powerful but lonely AMD-chips can finally find a new home in computer systems.
Or,
November 10, 1999
A person must look extremely hard on most of these manufacturers website's to find any literature or promotion for their Athlon supported motherboards.
And then,
December 8, 1999
I believe we will start seeing Athlon's motherboard support grow.
And finally, the first chipset NOT based on the AMD 750 Reference Chipset,
Februrary 4, 2000
AMD's Athlon is currently still damned to run on platforms using the one-and-only Athlon chipset 'Irongate', also manufactured by AMD. This chipset was initially still good enough to compete against all Intel processors on any platform, but it was already pretty outdated when it came to market.
AMD got burned last time, and is (rightfully so) doing what it can to keep it from happening again.
With thanks to Tom's Hardware...
I hate PDF files, so I converted the paper to html, and posted it Here.
Is there a real valid reason for this type of document to be in PDF form? Not to mention it is 122k vs 44k for HTML.
I work at a cable company, and I stress this exact point. It is a silly way to cap bandwidth in the first place, and it was crazy to ever be adopted. Maybe this is why Docsis is only an ad-hoc standard?
:)
The bad part is, the method of enforcing speeds employed by most (I stress MOST, you bet your ass that my methods aren't so easily fooled) cable operators has the same problem. They want to get your speed by SNMP query to your cable modem. Which again puts the trust in the client. While I haven't seen any SNMP faker hacks, I'm sure that they aren't too far behind. Another silly note is that most of those guys are comparing your speed to a list of approved speeds, not to a list of what customers bought what. This includes thier Business lines, which run over the same gear. You won't be able to sneak through with a 2meg/2meg pipe, but a 1.5meg/768k is a service they probobly sell, and would get right through thier checks.
The control method that you will see soon is called "shared secret", and is an encrypted passphrase-type method. Basicly, your cable modem gets a config file that has a key in it, which is basicly a signature of the bin file. It then generates a new passkey based on those two items, and send it to the CMTS. The CMTS verifies that it got a correct passkey, and then lets you connect. The encryption they used is junk, though, and there are efforts underway to break it. This is yet another dumb method that will only work for a short while!
What I will say is that there is a better way, and it is 100% effective. Your cable modem doesn't just "make up" a speed and magicly work, it has to register its rates with the CMTS. This is where the speed is truly controlled. While it isn't likely that Cisco will have a good method for capping individual users at the CMTS level, they are nice enough to tell you what speed someone is registered at. This is the method that I am using, and I *am* comparing speeds against what customers are paying for... So if you live in a town where you can get Imo's pizza, the square beyond compare, this is your warning!
On the flip side, once an abuser is identified, the info gets sent to marketing, and who knows what happens from there. We don't just pull the plug on abusers (yet).
The moron at wininformant added all exploits for all linux distributions together. Image the obvious scenerio, where bind8.x.x has a root compromise. This would only count as a single exploit, however the article counts it once for eash distribution that acknowledged it.
If you look at the charts yourself, you see that Win2k had 42 exploits in 2001. In comparison, SuSe had 21. Redhat had 54. OpenBSD had 14. The figures also are not focused on a particular release. I would expect that the numbers would be substantially lower if it only look into account the current releases. Suprise, SuSE still publishes security announcements for 6.x in addition to 7.x, and those are counted.
THe author of the atricle need to look up Aggregate and try writing an article again.
A quick check on CompUSA shows the USR model is available, and In Stock at my local store for $249.99.