but could it be that Apple couldn't sell enough/build enough to make this profitable and that is why this line is being discontinued? I am sure that for those that committed to this product this announcement is a real pain. I see some thinking that this means that Apple will stop developing for OS X server but I honestly don't see that. Would Apple have come out with their other server products if this was the case?
Who was most of Apple's server business with anyway and is it possible that most of them will be better served with Apple's present line of servers?
A dual-quad core with 12gb of ram, plus disk drives, raid card & apple care was like $8,000 at the time. I bought a Linux server with the same specs from Silicon Mechanics for $4,000 instead.
Why would I buy an Xserve when I know Apple is NOT and NEVER HAS BEEN serious about business support anyway? You knew this would be discontinued when it came out.
Apple is great on the consumer front, but completely clueless on the business side.
I'm not an IT guy, but I'm bummed that Apple seems to be abandoning the enterprise market. IF Apple is trying to make itself into a "consumer electronics" company, I think that's a terrible idea. They have tons of cash on hand, and certainly enough to really put some effort?brain and marketing?into making their mark in enterprise. The reason Apple has failed in the enterprise is that they've put as much effort into positioning themselves there as they have entering the Indian market (consumer/IT/Pro/anything!). That is to say, virtually none.
Someone in another forum speculated that perhaps they're planing a "one more thing" event to spring a new Mac Pro design on the world, one that would have the versatility to work as a desktop machine or a robust server. The current Mac Pro design, excellent as it is, is getting rather long in the tooth. It's been around for nearly a decade, which is an eternity in the computerverse.
In my dream world, Apple releases a new Mac Pro in a case design that could work either as a tower, for home/small business, or with a more robust motherboard/internal hardware configuration, a rackmount 3U(4U?) server: Externally accessible hard drives, hot swappable, redundant power supplies, etc.
The nightmare scenario is that Apple completely abandons the enterprise/IT market, the fallout being a return to the early/mid 90s, where the Mac was seen as novelty, but not a serious computer. iOS is great, and it's put Apple at the top of the smartphone/tablet?a.k.a. "appliance" market, but others will catch up after a while, and Apple needs to stay diversified.
I think in addition to a total lack of marketing, the Xserve was doomed because it was underpowered and overpriced, reinforcing the notion that Apple products are "luxury" items, but not to be taken seriously. Hopefully Apple is realizing this, and will build a better machine.
The Xserve was not underpowered or overpriced and Virginia Tech has proven.
As for hinting that Apple abandoning the enterprise market, is FUD. The primary reason Apple is perceived as not-enterprise friendly comes from the comparatively relative lack of interest from third-party independent professional consultants.
Not that there is not interest, it is just a simple fact that there is less need of their expertise to use a Mac, and internal IT personnel has always had a history to shun the Mac for fear of losing their jobs. Keep in mind, that most internal server systems are running in most part on legacy systems which are gradually modified or updated on a needs basis. And most of the needs fulfillment is based on having the necessary monies to do so.
As some have demonstrated, e.g., Virginia Tech, the casinos in Vegas, etc., the Mac makes a pretty damn good and cost-effective server using Mac Minis and Mac Pros, with or without Xserve. A strategy that is building significantly around the world.
No one was buying them because hardly anyone knew about them.
What was the Enterprise Group's advertising budget? $5?
I think you'd find plenty of people knew about them. It was the support that was non-existent, along with releasing server after server version with no long term support of the existing servers. IT will balk if there isn't any real hardware/software longevity, and Apple's veil of secrecy makes them a no-go in the enterprise.
The one vantage point I have not seen shown in the midst of all of this is what if Apple is again seeing low sales, and longer term, they are looking at the future, and seeing a lot less need for such servers to be needed.
Most servers support PCs (or Macs, duh so far) but as more cloud services are used, and mobile devices become more and more prevalent and powerful, it is not a far-fetched view of the future to see the desktop PC for most office workers to be replaced by their own personal mobile device.
Let the commodity hardware rule in the data centers where the clouds are being hosted, but the apps and data no longer need to updated and controlled through servers, and no AV necessary, and patch management, and all that mess.
I know I'm simplifying a lot here, but I'm trying to make the point that just because the world continues to see all of this infrastructure necessary because it has been for the last 20 years doesn't mean in 5 years that will still be the case.
Especially for the smaller companies where the Xserve was typically more targeted towards, those will be among the first to move more towards this model, where support almost entirely disappears.
I know in my office of 85 people, there is little our servers do but offer file access, we have a private cloud CRM and GoogleApps for our 2 most critical services, so handle some cloud file-sharing and all my users could nearly use mobile devices exclusively now (with a keyboard and monitor functionality of course). So to see that as a definite future step in 3-5 years is a no-brainer from my vantage point. I maintain access to the various cloud services, but SSO is not critical for that, a nicety, but not a deal-breaker.
Actually a Mac Mini Server would make an excellent server, actually. The specs of a mac mini are about the same as the specs we use to build our servers... I don't see where you would think that Mac Minis would make a poor server infrastructure.
So when you BTO your servers, you select "Shit" from the drop down menu?
I'm sure Apple is going to use the stock of XServers to populate their new datacenter. Once they get done with that, they'll refresh the line and come out with a new design. I see in the future a design that would enable a MAC Pro to be mounted in a standard rack.
If that was the case, announcing the end of Xserve without announcing it's replacement would be about the stupidest thing Apple could do. MS is famous for going off in a direction, building partnerships, and the screwing their partners by changing their direction. What you are suggesting Apple might be doing would be the same type of behavior that would prevent any potential partner/client from ever doing business with them because they are unstable and unreliable. MS has been wishy-washy in the audio player market, which is [one reason] why they are unsuccesful. Same for Apple in the enterprise market.
Far more than in the consumer market, the enterprise market wants/needs stability and reliability, not from a technical standpoint but from a business relationship standpoint. And I think that's one reason Apple didn't do well in the enterprise. They didn't show the commitment business wants to see before spending 10s-100s of thousands of dollars.
I'm sure Apple is going to use the stock of XServers to populate their new datacenter. Once they get done with that, they'll refresh the line and come out with a new design. I see in the future a design that would enable a MAC Pro to be mounted in a standard rack.
How are you "sure" of this? Apple's own job postings for the new data center don't seem to point to this being a fact.
One reason IT people were not that interested in Xserve is because it is a completely foreign OS for them. I would speculate that almost every Xserve was maintained by the person who bought it and installed it. If that person goes away, nobody knows how to work on it. Unlike Windows and Linux where there are millions of regular techs that can jump in and manage it at a moments notice.
I suppose an apple forum is the wrong place to ask this question, but:
- How many of you have rack-mounted servers at your company?
- For those of you who do, what OS?
We have Exchange servers for email/calendaring, and UNIX servers (I have no idea what flavor; it's not my specialty) for everything else.
Call me stupid ("okay, stupid") but I think the only people who care what flavor the rackmount is are the IT people working on it. Everyone else just cares if it works. Unix works great as a server platform with or without the mac veneer on top, and end users are none the wiser. IT geeks should know their way around a unix terminal without needing some shiny veneer. Hence, no need for a mac rack-mount server.
For the less-intelligent folk (again, me) there is a need for mac veneers, but our needs run to things like mac mini servers and mac pro servers. Once we get into rack-mounting things we know we're unqualified.
WE are running 7 of them with RAID 5 for CGI rendering, and they are fantastic, we now have to look and sideways Pro's if we add now... bummer
I suppose an apple forum is the wrong place to ask this question, but:
- How many of you have rack-mounted servers at your company?
- For those of you who do, what OS?
We are a small company, ~25 employees. While our first two servers (after we outgrew our NAS device) were towers, everything else is rackmount. (We have an Asterisk box for our phone system, a primary server that hosts the file server and some virtual machines for our internal wiki and other stuff, and a backup server that snapshots all of the other systems.) All of our servers are 2U because of the need for drive space or interface cards for the phone system. Everything runs Linux. (Our firewall box was replaced with an appliance recently.)
I wanted to buy an X-Serve; it was reasonably priced for an enterprise-class machine, and it looks pretty. It just wasn't a functional solution for us, as we would be forced into an external SAN box for both the primary and backup server. I wasn't confident that Apple would support the product at their retail stores (less than a mile away) either.
OSX Server really didn't offer anything for us that Linux didn't already have. Too many comments about the GUI getting in the way.
What's to keep people from building Hackintosh servers if they need rack mount? Pick up a 4U case, throw in a decent PS, mobo, Core i7, lots of RAM, a Silicon Image RAID controller and four hard drives. Install OS X Server and you should be good to go for less than $1000. Any motherboard that supports Snow Leopard well, like Gigbabyte's P55 series, should run SL server. Better yet, all the little problems that might afflict a Hackintosh system wouldn't matter as much where a server is concerned, if at all. No Quartz Extreme? Who cares? No audio? Who needs it? Can't sleep the system? Nobody puts servers to sleep anyway.
What's to keep people from building Hackintosh servers if they need rack mount? Pick up a 4U case, throw in a decent PS, mobo, Core i7, lots of RAM, a Silicon Image RAID controller and four hard drives. Install OS X Server and you should be good to go for less than $1000. Any motherboard that supports Snow Leopard well, like Gigbabyte's P55 series, should run SL server. Better yet, all the little problems that might afflict a Hackintosh system wouldn't matter as much where a server is concerned, if at all. No Quartz Extreme? Who cares? No audio? Who needs it? Can't sleep the system? Nobody puts servers to sleep anyway.
Yep. Another route is VirtualBox. Although not approved by Oracle or Apple, it works.
The Xserve was not underpowered or overpriced and Virginia Tech has proven.
I thought the whole VATech setup was pretty exciting, and hoped that it would spread to other universities/companies.
Quote:
As for hinting that Apple abandoning the enterprise market, is FUD. The primary reason Apple is perceived as not-enterprise friendly comes from the comparatively relative lack of interest from third-party independent professional consultants.
I freely admit to having FUD when it comes to Apple's relationship to enterprise/IT. I just remember when the 1U Xserves were introduced, they were priced at $2999 and, people were saying that was rather high.
Quote:
Not that there is not interest, it is just a simple fact that there is less need of their expertise to use a Mac, and internal IT personnel has always had a history to shun the Mac for fear of losing their jobs. Keep in mind, that most internal server systems are running in most part on legacy systems which are gradually modified or updated on a needs basis. And most of the needs fulfillment is based on having the necessary monies to do so.
I think you're totally right about IT managers' resistance to the OS X Server environment. IT personnel like their "status". Because of the complexity of straight-up Linux/Unix/Windows Server environment, only certified personnel with extensive training can function easily. In other words, in the traditional sphere, server environments are inaccessible to ordinary laypersons. Mac OS X Server, marketed for its ease of use as well as its robustness, has the potential to change that, which as you suggested is a threat to the IT "priesthood".
Quote:
As some have demonstrated, e.g., Virginia Tech, the casinos in Vegas, etc., the Mac makes a pretty damn good and cost-effective server using Mac Minis and Mac Pros, with or without Xserve. A strategy that is building significantly around the world.
I hope so, and I hope that Apple continues to grow and build a stronger presence in the IT world.
Especially for the smaller companies where the Xserve was typically more targeted towards, those will be among the first to move more towards this model, where support almost entirely disappears.
I know in my office of 85 people, there is little our servers do but offer file access, we have a private cloud CRM and GoogleApps for our 2 most critical services, so handle some cloud file-sharing and all my users could nearly use mobile devices exclusively now (with a keyboard and monitor functionality of course). So to see that as a definite future step in 3-5 years is a no-brainer from my vantage point. I maintain access to the various cloud services, but SSO is not critical for that, a nicety, but not a deal-breaker.
The X-Serve product never really had a great business driver to use; render farms were always better off with blades, and SMBs that didn't need a SAN were better off with 2U servers that can hold more SAS. The only real motivations for buying them seem to fall down to: looks/loyalty, need to mount in server or AV rack, or the perceived simplicity of administration. Apparently there are some grid computing applications where they are better than blades, but that is outside of my expertise.
But, SMB server needs aren't going away. Smart companies are having e-mail hosted; that is clearly cost-effective. Salesforce and the ilk are cost effective for the first few people, but it becomes expensive when the whole company needs it, so I doubt that is really going to be a driver. Things like Amazon S3 sounded good for remote data backup... but they are too expensive as well relative to placing a backup server in a colo.
Apple likely didn't think they had a way to compete without a virtualization asset, and I would say they were right.
The Apple G5 Xserve did have some advantages compared to typical Intel systems at the time due to the Altivec units, so you could get more flops per buck and watt than typical PC based clusters. However, now that Apple uses the same chips as everyone else there is no unique advantage to the Apple systems in hardware, and Mac OS is not specially optimized for cluster use, so no advantage there. A lot more variety in hardware from many companies to choose from. Apple Server software may have some advantages to casual server operators, but they seldom need rack mount systems.
Heck the XServes were innovative, striking, and influential, but they didn't really leverage Apple's existing advantages enough, and the best parts were simply copied in a more economical way by competitors. As innovative as the X Serves were, they were rather pricy. Apple has an overarching strategy that of creating synergy through recursive simplification and reuse; product lines, components, etc. Hardware—CPUs, GPUs, Memory, cameras, displays; OS—OS X, OS X Sever, i OS; Physical parts—iPhone, iPod Touch; Productions techniques, etc. Even their philosophy of coding using C, Objective-C, Cocoa, etc. follows this idea.
They didn't get much of that at all with the XServes.
Perhaps they can put together a system that will use a number of unmodified (or only slightly modified) subsystems common across several lines. This would produce an advantage in product, performance, cost, and price offering to customers. Say a blade system that uses Mac Pro or even Mac Mini boards in combination with some other Apple subsystems, all combined in an efficient, innovative, and compact way? The could get the overall cost down and figure out how to spend a bit less for the exterior (it's for racks after all!) while still maintaining the signature Apple look of quality.
Or how about this? An Apple blade system with 20 quad core CPUs per 2U using low power, high performance, souped up AppleTV boards in them. First stop? A deal with intel to license the right to make custom Apple SoCs with intel cores (but not intel graphics.) Could be extremely viable and would be much more in sync with the more typical Apple strategic plan.
Jobs allegedly responded on his iPhone: "Hardly anyone was buying them."
Not surprising since hardly anyone marketed them.
They were great blade servers (significantly better than the competition when they were running PowerPC chips) and I feel bad for the few IT people who stuck their neck out for them. This is another hard lesson that you really cant trust a company as chimerical as Apple with infrastructure technologies -- no matter how insanely great.
Does anyone remember OpenDoc (as it really was not as is described now by the powers that want the concept to remain dead).
Comments
The Subject line is not just empty but missing from the message header. Why is that?
However I would believe fewer Xserves have been sold than anything else ever offered on apple.com, except maybe the white iPhone 4.
Who was most of Apple's server business with anyway and is it possible that most of them will be better served with Apple's present line of servers?
Just asking the question.
Neal
Why would I buy an Xserve when I know Apple is NOT and NEVER HAS BEEN serious about business support anyway? You knew this would be discontinued when it came out.
Apple is great on the consumer front, but completely clueless on the business side.
I'm not an IT guy, but I'm bummed that Apple seems to be abandoning the enterprise market. IF Apple is trying to make itself into a "consumer electronics" company, I think that's a terrible idea. They have tons of cash on hand, and certainly enough to really put some effort?brain and marketing?into making their mark in enterprise. The reason Apple has failed in the enterprise is that they've put as much effort into positioning themselves there as they have entering the Indian market (consumer/IT/Pro/anything!). That is to say, virtually none.
Someone in another forum speculated that perhaps they're planing a "one more thing" event to spring a new Mac Pro design on the world, one that would have the versatility to work as a desktop machine or a robust server. The current Mac Pro design, excellent as it is, is getting rather long in the tooth. It's been around for nearly a decade, which is an eternity in the computerverse.
In my dream world, Apple releases a new Mac Pro in a case design that could work either as a tower, for home/small business, or with a more robust motherboard/internal hardware configuration, a rackmount 3U(4U?) server: Externally accessible hard drives, hot swappable, redundant power supplies, etc.
The nightmare scenario is that Apple completely abandons the enterprise/IT market, the fallout being a return to the early/mid 90s, where the Mac was seen as novelty, but not a serious computer. iOS is great, and it's put Apple at the top of the smartphone/tablet?a.k.a. "appliance" market, but others will catch up after a while, and Apple needs to stay diversified.
I think in addition to a total lack of marketing, the Xserve was doomed because it was underpowered and overpriced, reinforcing the notion that Apple products are "luxury" items, but not to be taken seriously. Hopefully Apple is realizing this, and will build a better machine.
The Xserve was not underpowered or overpriced and Virginia Tech has proven.
As for hinting that Apple abandoning the enterprise market, is FUD. The primary reason Apple is perceived as not-enterprise friendly comes from the comparatively relative lack of interest from third-party independent professional consultants.
Not that there is not interest, it is just a simple fact that there is less need of their expertise to use a Mac, and internal IT personnel has always had a history to shun the Mac for fear of losing their jobs. Keep in mind, that most internal server systems are running in most part on legacy systems which are gradually modified or updated on a needs basis. And most of the needs fulfillment is based on having the necessary monies to do so.
As some have demonstrated, e.g., Virginia Tech, the casinos in Vegas, etc., the Mac makes a pretty damn good and cost-effective server using Mac Minis and Mac Pros, with or without Xserve. A strategy that is building significantly around the world.
No one was buying them because hardly anyone knew about them.
What was the Enterprise Group's advertising budget? $5?
I think you'd find plenty of people knew about them. It was the support that was non-existent, along with releasing server after server version with no long term support of the existing servers. IT will balk if there isn't any real hardware/software longevity, and Apple's veil of secrecy makes them a no-go in the enterprise.
Most servers support PCs (or Macs, duh so far) but as more cloud services are used, and mobile devices become more and more prevalent and powerful, it is not a far-fetched view of the future to see the desktop PC for most office workers to be replaced by their own personal mobile device.
Let the commodity hardware rule in the data centers where the clouds are being hosted, but the apps and data no longer need to updated and controlled through servers, and no AV necessary, and patch management, and all that mess.
I know I'm simplifying a lot here, but I'm trying to make the point that just because the world continues to see all of this infrastructure necessary because it has been for the last 20 years doesn't mean in 5 years that will still be the case.
Especially for the smaller companies where the Xserve was typically more targeted towards, those will be among the first to move more towards this model, where support almost entirely disappears.
I know in my office of 85 people, there is little our servers do but offer file access, we have a private cloud CRM and GoogleApps for our 2 most critical services, so handle some cloud file-sharing and all my users could nearly use mobile devices exclusively now (with a keyboard and monitor functionality of course). So to see that as a definite future step in 3-5 years is a no-brainer from my vantage point. I maintain access to the various cloud services, but SSO is not critical for that, a nicety, but not a deal-breaker.
Actually a Mac Mini Server would make an excellent server, actually. The specs of a mac mini are about the same as the specs we use to build our servers... I don't see where you would think that Mac Minis would make a poor server infrastructure.
So when you BTO your servers, you select "Shit" from the drop down menu?
I'm sure Apple is going to use the stock of XServers to populate their new datacenter. Once they get done with that, they'll refresh the line and come out with a new design. I see in the future a design that would enable a MAC Pro to be mounted in a standard rack.
If that was the case, announcing the end of Xserve without announcing it's replacement would be about the stupidest thing Apple could do. MS is famous for going off in a direction, building partnerships, and the screwing their partners by changing their direction. What you are suggesting Apple might be doing would be the same type of behavior that would prevent any potential partner/client from ever doing business with them because they are unstable and unreliable. MS has been wishy-washy in the audio player market, which is [one reason] why they are unsuccesful. Same for Apple in the enterprise market.
Far more than in the consumer market, the enterprise market wants/needs stability and reliability, not from a technical standpoint but from a business relationship standpoint. And I think that's one reason Apple didn't do well in the enterprise. They didn't show the commitment business wants to see before spending 10s-100s of thousands of dollars.
Yoda: "Do, or do not. There is no try."
I'm sure Apple is going to use the stock of XServers to populate their new datacenter. Once they get done with that, they'll refresh the line and come out with a new design. I see in the future a design that would enable a MAC Pro to be mounted in a standard rack.
How are you "sure" of this? Apple's own job postings for the new data center don't seem to point to this being a fact.
http://jobs.apple.com/index.ajs?BID=...&CurrentPage=8
http://jobs.apple.com/index.ajs?BID=...CurrentPage=11
I suppose an apple forum is the wrong place to ask this question, but:
- How many of you have rack-mounted servers at your company?
- For those of you who do, what OS?
We have Exchange servers for email/calendaring, and UNIX servers (I have no idea what flavor; it's not my specialty) for everything else.
Call me stupid ("okay, stupid") but I think the only people who care what flavor the rackmount is are the IT people working on it. Everyone else just cares if it works. Unix works great as a server platform with or without the mac veneer on top, and end users are none the wiser. IT geeks should know their way around a unix terminal without needing some shiny veneer. Hence, no need for a mac rack-mount server.
For the less-intelligent folk (again, me) there is a need for mac veneers, but our needs run to things like mac mini servers and mac pro servers. Once we get into rack-mounting things we know we're unqualified.
WE are running 7 of them with RAID 5 for CGI rendering, and they are fantastic, we now have to look and sideways Pro's if we add now... bummer
I suppose an apple forum is the wrong place to ask this question, but:
- How many of you have rack-mounted servers at your company?
- For those of you who do, what OS?
We are a small company, ~25 employees. While our first two servers (after we outgrew our NAS device) were towers, everything else is rackmount. (We have an Asterisk box for our phone system, a primary server that hosts the file server and some virtual machines for our internal wiki and other stuff, and a backup server that snapshots all of the other systems.) All of our servers are 2U because of the need for drive space or interface cards for the phone system. Everything runs Linux. (Our firewall box was replaced with an appliance recently.)
I wanted to buy an X-Serve; it was reasonably priced for an enterprise-class machine, and it looks pretty. It just wasn't a functional solution for us, as we would be forced into an external SAN box for both the primary and backup server. I wasn't confident that Apple would support the product at their retail stores (less than a mile away) either.
OSX Server really didn't offer anything for us that Linux didn't already have. Too many comments about the GUI getting in the way.
What's to keep people from building Hackintosh servers if they need rack mount? Pick up a 4U case, throw in a decent PS, mobo, Core i7, lots of RAM, a Silicon Image RAID controller and four hard drives. Install OS X Server and you should be good to go for less than $1000. Any motherboard that supports Snow Leopard well, like Gigbabyte's P55 series, should run SL server. Better yet, all the little problems that might afflict a Hackintosh system wouldn't matter as much where a server is concerned, if at all. No Quartz Extreme? Who cares? No audio? Who needs it? Can't sleep the system? Nobody puts servers to sleep anyway.
Yep. Another route is VirtualBox. Although not approved by Oracle or Apple, it works.
The Xserve was not underpowered or overpriced and Virginia Tech has proven.
I thought the whole VATech setup was pretty exciting, and hoped that it would spread to other universities/companies.
As for hinting that Apple abandoning the enterprise market, is FUD. The primary reason Apple is perceived as not-enterprise friendly comes from the comparatively relative lack of interest from third-party independent professional consultants.
I freely admit to having FUD when it comes to Apple's relationship to enterprise/IT. I just remember when the 1U Xserves were introduced, they were priced at $2999 and, people were saying that was rather high.
Not that there is not interest, it is just a simple fact that there is less need of their expertise to use a Mac, and internal IT personnel has always had a history to shun the Mac for fear of losing their jobs. Keep in mind, that most internal server systems are running in most part on legacy systems which are gradually modified or updated on a needs basis. And most of the needs fulfillment is based on having the necessary monies to do so.
I think you're totally right about IT managers' resistance to the OS X Server environment. IT personnel like their "status". Because of the complexity of straight-up Linux/Unix/Windows Server environment, only certified personnel with extensive training can function easily. In other words, in the traditional sphere, server environments are inaccessible to ordinary laypersons. Mac OS X Server, marketed for its ease of use as well as its robustness, has the potential to change that, which as you suggested is a threat to the IT "priesthood".
As some have demonstrated, e.g., Virginia Tech, the casinos in Vegas, etc., the Mac makes a pretty damn good and cost-effective server using Mac Minis and Mac Pros, with or without Xserve. A strategy that is building significantly around the world.
I hope so, and I hope that Apple continues to grow and build a stronger presence in the IT world.
Especially for the smaller companies where the Xserve was typically more targeted towards, those will be among the first to move more towards this model, where support almost entirely disappears.
I know in my office of 85 people, there is little our servers do but offer file access, we have a private cloud CRM and GoogleApps for our 2 most critical services, so handle some cloud file-sharing and all my users could nearly use mobile devices exclusively now (with a keyboard and monitor functionality of course). So to see that as a definite future step in 3-5 years is a no-brainer from my vantage point. I maintain access to the various cloud services, but SSO is not critical for that, a nicety, but not a deal-breaker.
The X-Serve product never really had a great business driver to use; render farms were always better off with blades, and SMBs that didn't need a SAN were better off with 2U servers that can hold more SAS. The only real motivations for buying them seem to fall down to: looks/loyalty, need to mount in server or AV rack, or the perceived simplicity of administration. Apparently there are some grid computing applications where they are better than blades, but that is outside of my expertise.
But, SMB server needs aren't going away. Smart companies are having e-mail hosted; that is clearly cost-effective. Salesforce and the ilk are cost effective for the first few people, but it becomes expensive when the whole company needs it, so I doubt that is really going to be a driver. Things like Amazon S3 sounded good for remote data backup... but they are too expensive as well relative to placing a backup server in a colo.
Apple likely didn't think they had a way to compete without a virtualization asset, and I would say they were right.
The Apple G5 Xserve did have some advantages compared to typical Intel systems at the time due to the Altivec units, so you could get more flops per buck and watt than typical PC based clusters. However, now that Apple uses the same chips as everyone else there is no unique advantage to the Apple systems in hardware, and Mac OS is not specially optimized for cluster use, so no advantage there. A lot more variety in hardware from many companies to choose from. Apple Server software may have some advantages to casual server operators, but they seldom need rack mount systems.
Heck the XServes were innovative, striking, and influential, but they didn't really leverage Apple's existing advantages enough, and the best parts were simply copied in a more economical way by competitors. As innovative as the X Serves were, they were rather pricy. Apple has an overarching strategy that of creating synergy through recursive simplification and reuse; product lines, components, etc. Hardware—CPUs, GPUs, Memory, cameras, displays; OS—OS X, OS X Sever, i OS; Physical parts—iPhone, iPod Touch; Productions techniques, etc. Even their philosophy of coding using C, Objective-C, Cocoa, etc. follows this idea.
They didn't get much of that at all with the XServes.
Perhaps they can put together a system that will use a number of unmodified (or only slightly modified) subsystems common across several lines. This would produce an advantage in product, performance, cost, and price offering to customers. Say a blade system that uses Mac Pro or even Mac Mini boards in combination with some other Apple subsystems, all combined in an efficient, innovative, and compact way? The could get the overall cost down and figure out how to spend a bit less for the exterior (it's for racks after all!) while still maintaining the signature Apple look of quality.
Or how about this? An Apple blade system with 20 quad core CPUs per 2U using low power, high performance, souped up AppleTV boards in them. First stop? A deal with intel to license the right to make custom Apple SoCs with intel cores (but not intel graphics.) Could be extremely viable and would be much more in sync with the more typical Apple strategic plan.
Jobs allegedly responded on his iPhone: "Hardly anyone was buying them."
Not surprising since hardly anyone marketed them.
They were great blade servers (significantly better than the competition when they were running PowerPC chips) and I feel bad for the few IT people who stuck their neck out for them. This is another hard lesson that you really cant trust a company as chimerical as Apple with infrastructure technologies -- no matter how insanely great.
Does anyone remember OpenDoc (as it really was not as is described now by the powers that want the concept to remain dead).
I just love that he points out that he's French.
I'M FRENCH! WHY DO YOU THINK I WRITE WITH THIS HORRIBLE GRAMMAR YOU SILLY CEO YOU!
(anyone?)
I fart in your general direction. (But thanks to GPS and internal gyros, I now can do this with greater precision!)
Apple pulled the XServes years ago from MacWorld leaving us with just consumer gear and little reason for professionals to attend.
Apple never displayed them in a retail store that I ever saw.
Apple spent little on advertising them.
And you're surprised hardly anybody is buying them?
What enterprise servers DO you see advertised? Can you name any of the Cisco servers being advertised right now?
Trade magazines are the only place you will see this type of gear, and Apple actually did advertise in those areas often.
The advertising for this is done to the solution providers, not to the general public.