An IBM P550 with 4 x 1.9 GHz Power 5 processors does 7881 on SPECweb 2005 and costs arround $50,000 (IBM's web pricing system does not cover the config they tested)
The Dell 2850 with 2 x dual core CPU's does 4850 on SPECweb 2005 and costs $22,447
The IBM x346 with 2 x 3.8 GHx CPU's does 4348 on SPECweb and costs $27,874
The Sun T2000 with 1 Chip and 8 cores does 14001 on SPECweb 2005 and costs
$26,995
On this basis the T2000 looks pretty compelling its between 1.9x and 3.2x faster than the most current competitors and is cheaper than all of them except the Dell.
Sadly your view on the free availability of RedHat ES/AS from other sources than RedHat is not shared by any of the commercial software companies that support their products on RedHat.
Unless it is RedHat from RedHat then you will not get support for their products on it. So if you are a SW developer who doesn't need OS platform support but does need support for the DB/Middleware/Appserver layer then your example is of little interest. On the other hand they will be fine with Solaris 10. No of course they could all be using MySQL and JBoss but this is the real world so they probably arn't.
In addition if you have any issues with the OS/HW combination you will have rather serious difficulties getting any of the major platform vendors to honour their commitments to support RedHat/SuSE on their platforms. Of course you could be using a no name vendor but again this is the real world so you probably are not.
You need workloads which exhibit a high degree of thread level parallelism. These are generally workloads such as Web and File servers, J2EE app servers, mail authentication and directory servers, DBMS's and of course servers where there are largish number of independant processes.
When Sun aquired Affara (the initiators of the T1 design) they had a breakdown of the kind of apps that would benefit from multi-threaded CPU's.
Other people in this thread have commented on the suitability of the T1 as a notebook CPU. Apart from the fact the the T1 is a bit to hot for a notebook it will also suffer because desktop apps tend to be mostly single threaded with the focus of the operator on the performance of one app. They may have a large number running but at any one point in time they are only interested in the performance of one of them. Since the T1's single threaded performance is nothing to write home about it would not be a good choice for a desktop/notebook.
Rock which is the Sun's next major processor release after the T1 and its second revision will have much better single threaded performance. It is also unlikely to turn up as a desktop because it is targeted at the large server market.
You seem to have missed the point of the T1. On a conventional system a cache miss generates a stall and a wait while main memory fetches the page(s) while this happens the whole of your 100+ watt CPU is doing absolutely nothing for however many cycles it takes to get the page(s).
Not so the T1 which has another 31 threads which arn't stalled hence the T1 needs a wide but not particularly low latency memory subsystem.
It is not at any risk from SCO, Sun bought out their UNIX license from Unix Software Labs AKA Novell before SCO were in the picture.
No one really cares about the CDDL vs GPL issue except people who like the idea of flaming Sun. The reality is that there is very limitted scope for cutting and pasting Solaris kernel code into Linux and the other way arround and because of that CDDL vs GPL incompatibility is close to irrelevant. You can of course mix the user space stuff.
You may not be able to get OpenSolaris support but you can get Solaris x86 support roughly equivalent to the relationship between RHEL and Fedora.
I run Solaris 10 x86 on a couple of x86 boxes the normal things that trouble a Linux distribution troubled x86 such as Graphics adaptor, Wifi and Centrino drivers, apart from that it seems very stable and has some rather interesting features such as dtrace.
Hum, so you discount the rather large contribution that Sun make and has made to Gnome. Or for that matter the huge investment that Sun has put into OpenOffice/StarOffice one of the prime candidates to run in that desktop environment.
I rather doubt that Linux has more commercial software available for it than Solaris. The reality is that much more commercial software is 1st tier on Solaris than Linux.
It all rather depends on how many patents end up in the combined pot and what value these patents have in the real world.
The reality of IBM's last foray into patent donation to the OpenSource community was much much less impressive than the publicity it generated. Most of the patents were either irrelevant to the OpenSource community or about to expire or both.
I think you will find that Clustra is alive as part of Sun's J2EE App server, it provides state replication which allows for N+1 clustered nodes running the app server.
The Fat Duck at Bray, which is according to the survey the worlds best eatery serves food which would defy being pidgeonholed into a national stereotype. Try snail porridge as an example.
The Green project which spawned Java at Sun origionally targeted set top boxes and consumer devices. Oak which morphed into Java was intended to support the kind of consumer device which blu-ray will be.
The slowdown in interest in set top boxes coincided with the rise in interest in the Internet and Java was born. We are have now come full circle with most mobile phones supporting java.
Now that Global warming has been accepted as being a fact the insurance industry has now started trying to quantify how much it may cost them in terms of payouts due to changes in weather conditions.
The latest assesement is that the industry should expect normal payouts to increase by 11 billion dollars a year with extreme weather events costing up to 80 billion dollars as a one off with these increasing in frequency.
In the UK alone property worth 380 billion dollars is now at risk from river or coastal flooding. Property owners in these zones can expect to see their property values fall by as much as 40%.
In the light of this 500 billion dollars seems like a bargain.
Personally I think that OpenSolaris and the free availability of Solaris 10 on x86 spells the end of Linux as an OS for commercial server use.
My prediction is that Solaris x86 will replace most of the Linux systems that are currently being used as servers leaving Linux to fight it out with Apple for the non windows desktop category.
The only area I can see where Linux is likely to survive in commercial use is PDA's and small applicances.
Solaris 10 is free to use, you only pay if you want commercial support. Put another way this is rather like Fedora and RedHat ES/AS actually being the same byte for byte release which of course they arn't.
Your point question/point neatly illustrates how far the perception of Sun's support for OpenSource differs from the reality.
Sun is the single largest commercial donator to the OpenSource community and the second largest donator behind FSF. Huge chunks of this have not been headline grabbing stuff but instead relate to the mundane but impossible to do without basic plumbing for OS's. XFN, PAM etc etc.
However there has also been some headline grabbing stuff as well
OpenOffice
NetBeans
Grid
Big chunks of Gnome
etc etc
Sorry but this isn't supported by any of the admittedly limited tests that have been done comparing Solaris 10x86 with Linux/BSD on the same hardware. BR>
Solaris always appears to scale better than any of the Linux BSD variants and even more worrying if you are a Linux advocate the Solaris 10 x86 single processor throughput is within a gnats whisker of the best that Linux can do except on TCP dominated workloads where Solaris creams Linux.
I have been very happy with Solaris 9 x86 running on my Dell Laptop. Very stable and without the vices associated with RedHat Linux like having to recompile the kernel at a drop of a hat and RPM hell.
I would seriously think again when looking at Solaris 10.
Sun created a benchmark called libmicro which was essentially someting that measured the performance of all the system calls that Solaris did poorly on when compared with Linux for small CPU systems.
They gave the benchmark to the Solaris kernel engineers fed them Pizza and Jolt Cola and penned them into their cubicles until Solaris 10 was within 5% of the performance of RedHat ES on the same hardware.
So for single threaded applications expect Solaris 10 on x86 to deliver very similar performance to Linux on the same hardware. (Some public tests have been run showing this with MySQL).
Expect Solaris to absolutely trash Linux on anything that is TCP intensive and in the next release expect this to be replicated for UDP.
Also expect Solaris to be a much much much much nicer platform to deliver badly behaved apps on, the availability of dtrace makes it the premier platform for diagnosing aberant application behaviour on.
The performance changes made to Solaris 10 essentially push the Linux performance advantage window down to 1 CPU systems for edge case applications and for ahything that is network intensive there is no window at all.
I would agree I have used Skype to to call the West Coast of the US from the UK as well as Australia from the UK. I have had very few problems except with navigating touchtone voicemail and conferencing systems but even this seems to have been fixed in the latest release. These are SkypeOut calls, Skype to Skype is generally better.
The quality of your microphone and headset are also contributing factors.
Does anyone actually think it will? It looks like a fine upgrade for shops that are already heavily invested in Solaris, but I highly doubt that Solaris 10 (or 11 or 12 or 25 for that matter) will ever really be a 'Linux alternative'. Why would anyone using Linux go for a closed, proprietary Unix flavor? They cattle are stampeding in the other direction and will continue to do so.
Of course it will or rather of course it is. The reality of Linux adoption is that most of the commercial users of Linux adopted it not out of any conviction that OpenSource/Linux was better ( in fact many are still concerned that it is worse} but because it gave them access to low cost commodity cycles while providing a UNIX like environment.
Well guess what Solaris 9/10 x86 gives them access to exactly the same cycles is as fast or faster than Linux and is a commercial grade OS. When you couple that with a much more lenient support regime than say RedHat and lower prices and you have a pretty compelling reason to take a long hard look at your Linux deployment plans.
This is why the article missed the point so spetacularly, Solaris 10's hardware support is good enough for most commercial customers and they really don't care about the proprietary point unless it means them paying more money and since Solaris is a lot cheaper than RedHat it makes no difference.
Slashdot Mirror
An IBM P550 with 4 x 1.9 GHz Power 5 processors does 7881 on SPECweb 2005 and costs arround $50,000 (IBM's web pricing system does not cover the config they tested)
The Dell 2850 with 2 x dual core CPU's does 4850 on SPECweb 2005 and costs $22,447
The IBM x346 with 2 x 3.8 GHx CPU's does 4348 on SPECweb and costs $27,874
The Sun T2000 with 1 Chip and 8 cores does 14001 on SPECweb 2005 and costs $26,995
On this basis the T2000 looks pretty compelling its between 1.9x and 3.2x faster than the most current competitors and is cheaper than all of them except the Dell.
Sadly your view on the free availability of RedHat ES/AS from other sources than RedHat is not shared by any of the commercial software companies that support their products on RedHat.
Unless it is RedHat from RedHat then you will not get support for their products on it. So if you are a SW developer who doesn't need OS platform support but does need support for the DB/Middleware/Appserver layer then your example is of little interest. On the other hand they will be fine with Solaris 10. No of course they could all be using MySQL and JBoss but this is the real world so they probably arn't.
In addition if you have any issues with the OS/HW combination you will have rather serious difficulties getting any of the major platform vendors to honour their commitments to support RedHat/SuSE on their platforms. Of course you could be using a no name vendor but again this is the real world so you probably are not.
You need workloads which exhibit a high degree of thread level parallelism. These are generally workloads such as Web and File servers, J2EE app servers, mail authentication and directory servers, DBMS's and of course servers where there are largish number of independant processes.
When Sun aquired Affara (the initiators of the T1 design) they had a breakdown of the kind of apps that would benefit from multi-threaded CPU's.
Other people in this thread have commented on the suitability of the T1 as a notebook CPU. Apart from the fact the the T1 is a bit to hot for a notebook it will also suffer because desktop apps tend to be mostly single threaded with the focus of the operator on the performance of one app. They may have a large number running but at any one point in time they are only interested in the performance of one of them. Since the T1's single threaded performance is nothing to write home about it would not be a good choice for a desktop/notebook.
Rock which is the Sun's next major processor release after the T1 and its second revision will have much better single threaded performance. It is also unlikely to turn up as a desktop because it is targeted at the large server market.
You seem to have missed the point of the T1. On a conventional system a cache miss generates a stall and a wait while main memory fetches the page(s) while this happens the whole of your 100+ watt CPU is doing absolutely nothing for however many cycles it takes to get the page(s).
Not so the T1 which has another 31 threads which arn't stalled hence the T1 needs a wide but not particularly low latency memory subsystem.
It has good enough driver support
It is not at any risk from SCO, Sun bought out their UNIX license from Unix Software Labs AKA Novell before SCO were in the picture.
No one really cares about the CDDL vs GPL issue except people who like the idea of flaming Sun. The reality is that there is very limitted scope for cutting and pasting Solaris kernel code into Linux and the other way arround and because of that CDDL vs GPL incompatibility is close to irrelevant. You can of course mix the user space stuff.
You may not be able to get OpenSolaris support but you can get Solaris x86 support roughly equivalent to the relationship between RHEL and Fedora.
I run Solaris 10 x86 on a couple of x86 boxes the normal things that trouble a Linux distribution troubled x86 such as Graphics adaptor, Wifi and Centrino drivers, apart from that it seems very stable and has some rather interesting features such as dtrace.
Hum, so you discount the rather large contribution that Sun make and has made to Gnome. Or for that matter the huge investment that Sun has put into OpenOffice/StarOffice one of the prime candidates to run in that desktop environment.
I rather doubt that Linux has more commercial software available for it than Solaris. The reality is that much more commercial software is 1st tier on Solaris than Linux.
It all rather depends on how many patents end up in the combined pot and what value these patents have in the real world.
The reality of IBM's last foray into patent donation to the OpenSource community was much much less impressive than the publicity it generated. Most of the patents were either irrelevant to the OpenSource community or about to expire or both.
I think you will find that Clustra is alive as part of Sun's J2EE App server, it provides state replication which allows for N+1 clustered nodes running the app server.
The Fat Duck at Bray, which is according to the survey the worlds best eatery serves food which would defy being pidgeonholed into a national stereotype. Try snail porridge as an example.
Actually AMD is better at both FP and INT.
For SPECint the fastest AMD64 result is 1854
For SPECint the fastest Intel result is 1815
For SPECfp the fastest AMD64 result is 2261
For SPECfp the fastest Intel result is 2016
When it comes to power the Intel P4 EE is 115 watts, the AMD Athlon 64 FX is 72.
AMD's lead widens for anything that involves multiple CPU's because of its superior system architecture, Hypertransport vs Frontside.
If you compare any 5-6 year old system with the latest system you can buy now you will always find that the newer system is much faster and cheaper.
How about comparing the Opteron powered HP DL with a Sun V20z or V40z.
The Green project which spawned Java at Sun origionally targeted set top boxes and consumer devices. Oak which morphed into Java was intended to support the kind of consumer device which blu-ray will be.
The slowdown in interest in set top boxes coincided with the rise in interest in the Internet and Java was born. We are have now come full circle with most mobile phones supporting java.
Now that Global warming has been accepted as being a fact the insurance industry has now started trying to quantify how much it may cost them in terms of payouts due to changes in weather conditions.
The latest assesement is that the industry should expect normal payouts to increase by 11 billion dollars a year with extreme weather events costing up to 80 billion dollars as a one off with these increasing in frequency.
In the UK alone property worth 380 billion dollars is now at risk from river or coastal flooding. Property owners in these zones can expect to see their property values fall by as much as 40%.
In the light of this 500 billion dollars seems like a bargain.
How will having AC help you if the Gulf Stream stops ? Don't assume that all the consequences of global warming involve you actually getting warmer.
Personally I think that OpenSolaris and the free availability of Solaris 10 on x86 spells the end of Linux as an OS for commercial server use.
My prediction is that Solaris x86 will replace most of the Linux systems that are currently being used as servers leaving Linux to fight it out with Apple for the non windows desktop category.
The only area I can see where Linux is likely to survive in commercial use is PDA's and small applicances.
We should see in a couple of years.
Solaris 10 is free to use, you only pay if you want commercial support. Put another way this is rather like Fedora and RedHat ES/AS actually being the same byte for byte release which of course they arn't.
I guess this also explains why Linux is doomed
Your point question/point neatly illustrates how far the perception of Sun's support for OpenSource differs from the reality.
Sun is the single largest commercial donator to the OpenSource community and the second largest donator behind FSF. Huge chunks of this have not been headline grabbing stuff but instead relate to the mundane but impossible to do without basic plumbing for OS's. XFN, PAM etc etc.
However there has also been some headline grabbing stuff as well
OpenOffice
NetBeans
Grid
Big chunks of Gnome
etc etc
Sorry but this isn't supported by any of the admittedly limited tests that have been done comparing Solaris 10x86 with Linux/BSD on the same hardware. BR>
Solaris always appears to scale better than any of the Linux BSD variants and even more worrying if you are a Linux advocate the Solaris 10 x86 single processor throughput is within a gnats whisker of the best that Linux can do except on TCP dominated workloads where Solaris creams Linux.
I have been very happy with Solaris 9 x86 running on my Dell Laptop. Very stable and without the vices associated with RedHat Linux like having to recompile the kernel at a drop of a hat and RPM hell.
Dtrace and no its not an idea thats easily borrowed unless you envisage re-writing the OpenBSD kernel as an easy option.
The Solaris 10 IP stack ditto.
Solaris Service Manager ditto.
I would seriously think again when looking at Solaris 10.
Sun created a benchmark called libmicro which was essentially someting that measured the performance of all the system calls that Solaris did poorly on when compared with Linux for small CPU systems.
They gave the benchmark to the Solaris kernel engineers fed them Pizza and Jolt Cola and penned them into their cubicles until Solaris 10 was within 5% of the performance of RedHat ES on the same hardware.
So for single threaded applications expect Solaris 10 on x86 to deliver very similar performance to Linux on the same hardware. (Some public tests have been run showing this with MySQL).
Expect Solaris to absolutely trash Linux on anything that is TCP intensive and in the next release expect this to be replicated for UDP.
Also expect Solaris to be a much much much much nicer platform to deliver badly behaved apps on, the availability of dtrace makes it the premier platform for diagnosing aberant application behaviour on.
The performance changes made to Solaris 10 essentially push the Linux performance advantage window down to 1 CPU systems for edge case applications and for ahything that is network intensive there is no window at all.
I would agree I have used Skype to to call the West Coast of the US from the UK as well as Australia from the UK. I have had very few problems except with navigating touchtone voicemail and conferencing systems but even this seems to have been fixed in the latest release. These are SkypeOut calls, Skype to Skype is generally better.
The quality of your microphone and headset are also contributing factors.
Does anyone actually think it will? It looks like a fine upgrade for shops that are already heavily invested in Solaris, but I highly doubt that Solaris 10 (or 11 or 12 or 25 for that matter) will ever really be a 'Linux alternative'. Why would anyone using Linux go for a closed, proprietary Unix flavor? They cattle are stampeding in the other direction and will continue to do so.
Of course it will or rather of course it is. The reality of Linux adoption is that most of the commercial users of Linux adopted it not out of any conviction that OpenSource/Linux was better ( in fact many are still concerned that it is worse} but because it gave them access to low cost commodity cycles while providing a UNIX like environment.
Well guess what Solaris 9/10 x86 gives them access to exactly the same cycles is as fast or faster than Linux and is a commercial grade OS. When you couple that with a much more lenient support regime than say RedHat and lower prices and you have a pretty compelling reason to take a long hard look at your Linux deployment plans.
This is why the article missed the point so spetacularly, Solaris 10's hardware support is good enough for most commercial customers and they really don't care about the proprietary point unless it means them paying more money and since Solaris is a lot cheaper than RedHat it makes no difference.