Being a clustering product manager at a Linux company, I've been cornered into coming up with a definition of clustering that seems to fit most technologies that people consider clustering. It may have been abstracted to the point of being overly vague, but the following definition seems to fit most forms of clustering...
"A cluster is a group of systems, bound together into a common resource pool. A given task, whether that task is a web server, mathematical calculation, or robotic cooking widget is able to be properly and arbitrarily executed on any of the member nodes within the cluster. (This does not imply that it can be run concurrently or in parallel on multiple nodes)
To the 'outside world', or the entity using the cluster, the cluster appears as a single object. That is to say, the cluster has the image of a single system. (single state image)"
I sometimes think an accurate and useful definition of 'cluster' is one of the holy grails of this industry. It's about as overloaded a term as the word 'stuff'.:) If there are comments that can make the above definition better, please feel free to comment.
Aaron McKee Clustering Products Manager TurboLinux, Inc.
The Linux industry and community have whole-heartadly put their support behind the LPI certification. (http://www.lpi.org) This is an industry standard certification board, supported by Caldera, SuSE, TurboLinux, LinuxCare, IBM, SGI, VA Linux, and others that will prepare the certification candidate to be prepared on Linux... not just Red Hat. Individuals wishing to be prepared for Linux would be well served by visiting the above link.
Noteably absent from the above list of companies is Red Hat. As a member of the Linux community, Red Hat should be working with others to help define standards for the entire industry, and not just itself. Adopting a mindset of "We're the standard, so you play by our rules." is not what the community wants to see and is not the way to accomplish this goal. Whether you're a Debian, SuSE, TurboLinux, or Caldera user... standards are critical to us all. They help make each of our lives significantly easier. Staying cohesive as a group also gives us additional strength. It seems difficult to expect that one rogue straggler could ever succeed. As with Linux itself, our success will be based on the efforts of the larger Community more than just the efforts of one company.
Also, as anyone who has worked with a vendor's own certification program will know, most of those non-standard programs are used more for marketing than independant analysis of knowledge. ("You barked the right marketing bullets, here's a bone. Good boy!") This is certainly true for the MCSE certification process (of which I am certified and have seen firsthand:( ) and has been reported as true for the RHCE certification, of which I've had numerous friends attend.
Anyway, the whole argument of whether RHCE is better than MCSE seems more religious and inflammatory than what can possibly be answered in this setting. Both certainly will have their positive and negative points. However, if we choose to support something vendor-neutral such as the LPI we can, as a community, address the weaknesses.
(My comments are my own and don't necessarily reflect the opinions of TurboLinux.) Aaron McKee Clustering Products Manager TurboLinux, Inc.
As previously mentioned, TurboLinux also announced similar partnerships with IBM regarding the Java technology. LinuxToday picked up all the various press releases on the topic.
On the surface, it appears that the Red Hat claim to being the "first to license" is incorrect. My preferred News Authority (LinuxToday:) actually presented some other similar announcements prior to RH's own press release. Go figure.
The important thing to note about these announcements is that it brings more choice to those of us who program in Java under Linux. I think we're all in agreement that it's not in any of our best interests to have one corporation with exclusive access to such a critical technology, especially considering a demonstrated tendancy towards extreme territorial behavior. This helps level the playing field, and demonstrates IBM's continued support of the Linux platform.
Aaron McKee Clustering Products Manager TurboLinux, Inc.
I don't want to appear hostile, as I very warmly welcome the criticism and feedback from the community, but I am a bit defensive. Let me first preface these comments with the fact that I was not responsible for the press release. I am, however, one of the engineers here. Please don't construe this as an argument.
I don't believe the title was wrong because the product "Microsoft Windows 98" is a distinctly different product than the "Microsoft Windows 98: Upgrade" product, which is distinctly different than the academic versions. Functionally, they are similar, but as far as statistics and sales tracking goes they are about as different as OS/2 from NT. Each product has a different SKU, sales channel, and price point. Each is marketed in a slightly different fashion. The news agency that collected the information, Business Computer News, decided to separate out the figures in that fashion because that was their policy. Within this scope, the title was correct: we did sell more copies of TurboLinux than Microsoft sold copies of Windows 98.
TurboLinux wasn't trying to be deceptive, and I don't think we deliberately said any mistruths, but we did want to boast a bit. I expect this is natural. I think the only thing we can really be considered guilty of is a little spin doctoring. I used Microsoft as an extreme example of what is considered acceptable spinning, although I personally feel that most of what comes out of their marketing department tends to be of marginal value. Most other companies, including Red Hat, try to post as much favorable press as they can. This is what I was referring to when I said 'completely standard business practice', and I don't think we did anything more than that.
I wish you luck with trying out the distribution. When you do get a chance to take a look at it, please let me know what you think. We're always receptive to ways of improving the product, and the community almost always provides excellent feedback.
Best Regards, Aaron McKee Sr. Technical Marketing Engineer TurboLinux, Inc.
Although we were expecting the numbers to go down, as well, we were pleasantly surprised to see that TurboLinux held on for another week. For the second week of it's release, 23% of all retail OS sales were TurboLinux.
It's expected that sales are usually larger when a product is first released, and that's when most business analysts correlate and publish their statistics. It's fairly standard business practice. Microsoft even announces pre-release statistics for orders that are in a queue. The numbers will certainly go down, but a lot of individuals are only interested in peak values.
I think a lot of people are also missing the real point of this. Linux is appearing as a viable OS platform to a very large and important segment of the world's computing population. Whether you use TurboLinux or not, this is certainly a "Good Thing(tm)" for almost every Linux user and is what a lot of people have been working towards for a long time. To a lot of people who don't currently use Linux, this is one indication that Linux is 'growing up' and turning into a mainstream operating sytem.
Best Regards, Aaron McKee Sr. Technical Marketing Engineer TurboLinux, Inc.
Although we were expecting the numbers to go down, as well, we were pleasantly surprised to see that TurboLinux held on for another week. For the second week of it's release, 23% of all retail OS sales were TurboLinux.
It's expected that sales are usually larger when a product is first released, and that's when most business analysts correlate and publish their statistics. It's not a lie, it's just standard business practice. Microsoft even announces pre-release statistics for orders that are in a queue.
I think a lot of people are missing the real point of this. Linux is appearing as a viable OS platform to a very large and important segment of the world's computing population. Whether you use TurboLinux or not, this is certainly a "Good Thing(tm)" and is what a lot of people have been working towards for a very long time. One can quibble over how the numbers were calculated, but in the end TurboLinux still accounted for 24% of the sales for operating systems. To a lot of people who don't currently use Linux, this is one indication that Linux is 'growing up' and turning into a mainstream operating sytem.
Best Regards,
Aaron McKee Sr. Technical Marketing Engineer TurboLinux, Inc.
These statistics were based on retail sales of software, not hardware. Although virtually every new PC comes with Windows 98, there are a significant number of individuals who actually purchase Windows 98 to upgrade their Windows 3.x or Windows 95 platforms, as the 'Windows 98 Upgrade' number indicates.
The numbers don't pretend to indicate market share, and TurboLinux was not responsible for collecting the numbers. These values came straight out of Business Computer News. If BCN chose not to include OS/HW bundling in their sales statistics, it was their decision.
I think you'll also find that there were no lies within the press announcement. The report explicitly stated that TurboLinux outsold every other individual OS. Because each of the variants of Windows 98 has been SKU'd differently and has a different price point, Business Computer News likely chose to separate out the values themselves. Again, that was their policy and not ours.
I think the only thing we can be said to be guilty of is a little spinning. This is a completely standard business practice, and I think it was justified. We're very proud of Linux, and our particular distribution, and we want the world to realize that Linux is a viable platform. If the article helped convince even one business manager that Linux is a grown-up operating system, and not just a fringe-player, then I think the press release did the community a service.
Whether you use TurboLinux or not, I think everyone who loves Linux should feel proud to see a distribution achieve this level of market penetration.
Best Regards, Aaron McKee Sr. Technical Marketing Engineer TurboLinux, Inc.
Contrary to your assertation, Sun has been in the x86 world for quite some time. The Intel Platform Edition of Solaris 7 is one of the most popular commerical UNIX operating systems for Intel based servers. They've been supporting IA for years, and years, and the OS is quite popular.
As to no workstation users choosing to have an Intel chip, I'd have to disagree. Most workstatipons *are* based on the Intel architecture, such as those by Integraph and SGI.
Aaron McKee Sr. Technical Marketing Engineer TurboLinux, Inc.
Check out www.turbolinux.com. We're currently hiring junior and senior Linux programmers (kernel and application), technical support engineers, test engineers, and a whole host of other individuals. We're located about 5 miles south of the City, and the commute time from downtown is typically less than 15 minutes. Our offices overlook the bay, the mountains, and 3com park.
Although fairly new to the United States, TurboLinux is the largest distributor of Linux in the Asian region. Earlier this month, our TurboLinux 4.0J release outsold even Microsoft's operating systems. We're focused on driving Linux into the enterprise and are looking for people that can help make that a reality.
Best Regards, Aaron McKee Sr. Technical Marketing Engineer TurboLinux, Inc.
Slashdot Mirror
Being a clustering product manager at a Linux company, I've been cornered into coming up with a definition of clustering that seems to fit most technologies that people consider clustering. It may have been abstracted to the point of being overly vague, but the following definition seems to fit most forms of clustering...
:) If there are comments that can make the above definition better, please feel free to comment.
"A cluster is a group of systems, bound together into a common resource pool. A given task, whether that task is a web server, mathematical calculation, or robotic cooking widget is able to be properly and arbitrarily executed on any of the member nodes within the cluster. (This does not imply that it can be run concurrently or in parallel on multiple nodes)
To the 'outside world', or the entity using the cluster, the cluster appears as a single object. That is to say, the cluster has the image of a single system. (single state image)"
I sometimes think an accurate and useful definition of 'cluster' is one of the holy grails of this industry. It's about as overloaded a term as the word 'stuff'.
Aaron McKee
Clustering Products Manager
TurboLinux, Inc.
The Linux industry and community have whole-heartadly put their support behind the LPI certification. (http://www.lpi.org) This is an industry standard certification board, supported by Caldera, SuSE, TurboLinux, LinuxCare, IBM, SGI, VA Linux, and others that will prepare the certification candidate to be prepared on Linux... not just Red Hat. Individuals wishing to be prepared for Linux would be well served by visiting the above link.
:( ) and has been reported as true for the RHCE certification, of which I've had numerous friends attend.
Noteably absent from the above list of companies is Red Hat. As a member of the Linux community, Red Hat should be working with others to help define standards for the entire industry, and not just itself. Adopting a mindset of "We're the standard, so you play by our rules." is not what the community wants to see and is not the way to accomplish this goal. Whether you're a Debian, SuSE, TurboLinux, or Caldera user... standards are critical to us all. They help make each of our lives significantly easier. Staying cohesive as a group also gives us additional strength. It seems difficult to expect that one rogue straggler could ever succeed. As with Linux itself, our success will be based on the efforts of the larger Community more than just the efforts of one company.
Also, as anyone who has worked with a vendor's own certification program will know, most of those non-standard programs are used more for marketing than independant analysis of knowledge. ("You barked the right marketing bullets, here's a bone. Good boy!") This is certainly true for the MCSE certification process (of which I am certified and have seen firsthand
Anyway, the whole argument of whether RHCE is better than MCSE seems more religious and inflammatory than what can possibly be answered in this setting. Both certainly will have their positive and negative points. However, if we choose to support something vendor-neutral such as the LPI we can, as a community, address the weaknesses.
(My comments are my own and don't necessarily reflect the opinions of TurboLinux.)
Aaron McKee
Clustering Products Manager
TurboLinux, Inc.
As previously mentioned, TurboLinux also announced similar partnerships with IBM regarding the Java technology. LinuxToday picked up all the various press releases on the topic.
:) actually presented some other similar announcements prior to RH's own press release. Go figure.
Our specific information can be found at http://linuxtoday.com/stories/15563.html
On the surface, it appears that the Red Hat claim to being the "first to license" is incorrect. My preferred News Authority (LinuxToday
The important thing to note about these announcements is that it brings more choice to those of us who program in Java under Linux. I think we're all in agreement that it's not in any of our best interests to have one corporation with exclusive access to such a critical technology, especially considering a demonstrated tendancy towards extreme territorial behavior. This helps level the playing field, and demonstrates IBM's continued support of the Linux platform.
Aaron McKee
Clustering Products Manager
TurboLinux, Inc.
Simon,
I don't want to appear hostile, as I very warmly welcome the criticism and feedback from the community, but I am a bit defensive. Let me first preface these comments with the fact that I was not responsible for the press release. I am, however, one of the engineers here. Please don't construe this as an argument.
I don't believe the title was wrong because the product "Microsoft Windows 98" is a distinctly different product than the "Microsoft Windows 98: Upgrade" product, which is distinctly different than the academic versions. Functionally, they are similar, but as far as statistics and sales tracking goes they are about as different as OS/2 from NT. Each product has a different SKU, sales channel, and price point. Each is marketed in a slightly different fashion. The news agency that collected the information, Business Computer News, decided to separate out the figures in that fashion because that was their policy. Within this scope, the title was correct: we did sell more copies of TurboLinux than Microsoft sold copies of Windows 98.
TurboLinux wasn't trying to be deceptive, and I don't think we deliberately said any mistruths, but we did want to boast a bit. I expect this is natural. I think the only thing we can really be considered guilty of is a little spin doctoring. I used Microsoft as an extreme example of what is considered acceptable spinning, although I personally feel that most of what comes out of their marketing department tends to be of marginal value. Most other companies, including Red Hat, try to post as much favorable press as they can. This is what I was referring to when I said 'completely standard business practice', and I don't think we did anything more than that.
I wish you luck with trying out the distribution. When you do get a chance to take a look at it, please let me know what you think. We're always receptive to ways of improving the product, and the community almost always provides excellent feedback.
Best Regards,
Aaron McKee
Sr. Technical Marketing Engineer
TurboLinux, Inc.
Although we were expecting the numbers to go down, as well, we were pleasantly surprised to see that TurboLinux held on for another week. For the second week of it's release, 23% of all retail OS sales were TurboLinux.
It's expected that sales are usually larger when a product is first released, and that's when most business analysts correlate and publish their statistics. It's fairly standard business practice. Microsoft even announces pre-release statistics for orders that are in a queue. The numbers will certainly go down, but a lot of individuals are only interested in peak values.
I think a lot of people are also missing the real point of this. Linux is appearing as a viable OS platform to a very large and important segment of the world's computing population. Whether you use TurboLinux or not, this is certainly a "Good Thing(tm)" for almost every Linux user and is what a lot of people have been working towards for a long time. To a lot of people who don't currently use Linux, this is one indication that Linux is 'growing up' and turning into a mainstream operating sytem.
Best Regards,
Aaron McKee
Sr. Technical Marketing Engineer
TurboLinux, Inc.
Although we were expecting the numbers to go down, as well, we were pleasantly surprised to see that TurboLinux held on for another week. For the second week of it's release, 23% of all retail OS sales were TurboLinux.
It's expected that sales are usually larger when a product is first released, and that's when most business analysts correlate and publish their statistics. It's not a lie, it's just standard business practice. Microsoft even announces pre-release statistics for orders that are in a queue.
I think a lot of people are missing the real point of this. Linux is appearing as a viable OS platform to a very large and important segment of the world's computing population. Whether you use TurboLinux or not, this is certainly a "Good Thing(tm)" and is what a lot of people have been working towards for a very long time. One can quibble over how the numbers were calculated, but in the end TurboLinux still accounted for 24% of the sales for operating systems. To a lot of people who don't currently use Linux, this is one indication that Linux is 'growing up' and turning into a mainstream operating sytem.
Best Regards,
Aaron McKee
Sr. Technical Marketing Engineer
TurboLinux, Inc.
These statistics were based on retail sales of software, not hardware. Although virtually every new PC comes with Windows 98, there are a significant number of individuals who actually purchase Windows 98 to upgrade their Windows 3.x or Windows 95 platforms, as the 'Windows 98 Upgrade' number indicates.
The numbers don't pretend to indicate market share, and TurboLinux was not responsible for collecting the numbers. These values came straight out of Business Computer News. If BCN chose not to include OS/HW bundling in their sales statistics, it was their decision.
I think you'll also find that there were no lies within the press announcement. The report explicitly stated that TurboLinux outsold every other individual OS. Because each of the variants of Windows 98 has been SKU'd differently and has a different price point, Business Computer News likely chose to separate out the values themselves. Again, that was their policy and not ours.
I think the only thing we can be said to be guilty of is a little spinning. This is a completely standard business practice, and I think it was justified. We're very proud of Linux, and our particular distribution, and we want the world to realize that Linux is a viable platform. If the article helped convince even one business manager that Linux is a grown-up operating system, and not just a fringe-player, then I think the press release did the community a service.
Whether you use TurboLinux or not, I think everyone who loves Linux should feel proud to see a distribution achieve this level of market penetration.
Best Regards,
Aaron McKee
Sr. Technical Marketing Engineer
TurboLinux, Inc.
Contrary to your assertation, Sun has been in the x86 world for quite some time. The Intel Platform Edition of Solaris 7 is one of the most popular commerical UNIX operating systems for Intel based servers. They've been supporting IA for years, and years, and the OS is quite popular.
As to no workstation users choosing to have an Intel chip, I'd have to disagree. Most workstatipons *are* based on the Intel architecture, such as those by Integraph and SGI.
Aaron McKee
Sr. Technical Marketing Engineer
TurboLinux, Inc.
Check out www.turbolinux.com. We're currently hiring junior and senior Linux programmers (kernel and application), technical support engineers, test engineers, and a whole host of other individuals. We're located about 5 miles south of the City, and the commute time from downtown is typically less than 15 minutes. Our offices overlook the bay, the mountains, and 3com park.
Although fairly new to the United States, TurboLinux is the largest distributor of Linux in the Asian region. Earlier this month, our TurboLinux 4.0J release outsold even Microsoft's operating systems. We're focused on driving Linux into the enterprise and are looking for people that can help make that a reality.
Best Regards,
Aaron McKee
Sr. Technical Marketing Engineer
TurboLinux, Inc.