I hope that Apple offers something like Xen in OS X soon. This seems like the best sort of option to provide the virtualization technology currently lacking.
So, how could one put these numbers into context in order to be useful?
Well, you understand that I don't have these numbers at my fingertips.
But, the server line, and the client line, are like two separate worlds, in a way.
The servers will have problems for the reasons I gave in post #15. The clients are different, but still will have difficulties because of some of those problems. But they don't have to be as colosly involved.
The question is what does a company need the machines for?
After that is determined, the rest falls into place.
Apple is developing an alternative to Exchange, but it will need installs to become widely useful. Unless Apple is willing to address industry's needs in the other areas, it won't matter.
The context is that it is up to Apple to decide whether it wants to be a player, or not.
Jobs had me cringing when he first came back to Apple.
In a press conference, he was asked about the enterprise, and what Apple would do to get it back.
What exactly is the target market for the X-Serve? I see a few of them going into data centers, but can't understand who they offer a good value proposition for.
For my small business, I wanted to find an excuse to go with an X-Serve instead of a Dell... but the premium was almost 2x! Even factoring in consulting time to go with Linux, we saved over $1k with the Dull.
I honestly am just having trouble understanding, and would love to know. It seems like Apple really needs to come up with a more cost-effective solution if they want to make a strong push of Leopard Server.
Well that depends on your needs and the currently available hardware. I'll be the first to admit that Tiger server is limited in many key ways that pretty much relegate it to predominanty Mac networks and supporting roles in Windows dominated server networks.
When I price the current Xserve against the competition I see an immediate savings but that depends on what you need to do.
Leopard server will make this much easier. Every company needs a calendar server and finally OS X server will have it. ACLS are fully supported in Leopard and they have a new Team Directory structure which should be useful.
I'm a big fan of Time Machine...snapshots and a robust backup system are a requirement today with end user restore as a the gravy.
Active/Active failover with the Mail server and XSAN, clustered iCal server. RADIUS for wireless networks.
Leopard is the "Tiger" for sys admins. Meaning it's going to be the first server OS that admins feel "comfortable" and relatively empowered on.
Steve Jobs has stated that Apple is not an Enterprise company and he's right. They don't have to be. They need to be focused on that mid tier and SMB market where people are beginning to move to multiple servers and modern storage solutions.
Playing at the Enterprise level means committing to fast growth. Look at the Top 3 server vendors. They aquire a new company about every month. Apple aquires a new company maybe every 6 months or so. Certainly not enough to make a play in the Enterprise.
What I think Apple will do is continue to utilize and support Open Source technologies folding those into OS X and keeping costs down and interoperability up.
Evolving the ZFS support into a stable and mature product is important as well. I hope XSAN 2.0 syncs with StorNeXt 3.0
Distributed LAN Client enables applications on the LAN to access a shared pool of storage faster, and with a higher level of resiliency, than most traditional network sharing methods can provide. Using clustered gateways for access and an optimized communication protocol for performance, LAN Client not only extends data sharing but also delivers enterprise features such as load balancing and transparent I/O failover.
Data Reduction Storage is a specialized tier of disk that uses data de-duplication to reduce a customer’s data footprint, saving money by lowering capacity requirements and enabling data to be retained on fast recovery disk for a much longer period of time. Quantum has pioneering data de-duplication technology, a fundamental enabler for storage applications in backup and archive. StorNext is the first product of its kind to include such intuitive, user-specific data reduction technology.
Dynamic Resource Allocation increases uptime by enabling online service operations. It allows customers to scale their storage – adding new storage capacity or transparently swapping out disk arrays during hardware upgrades – while the system is active..
Consider Apple to be "drafting" behind the leaders. Whatever they spend millions on Apple wants to deliver without the same expenditure.
Well, you understand that I don't have these numbers at my fingertips.
But, the server line, and the client line, are like two separate worlds, in a way.
The servers will have problems for the reasons I gave in post #15. The clients are different, but still will have difficulties because of some of those problems. But they don't have to be as colosly involved.
The question is what does a company need the machines for?
After that is determined, the rest falls into place.
Apple is developing an alternative to Exchange, but it will need installs to become widely useful. Unless Apple is willing to address industry's needs in the other areas, it won't matter.
The context is that it is up to Apple to decide whether it wants to be a player, or not.
Jobs had me cringing when he first came back to Apple.
In a press conference, he was asked about the enterprise, and what Apple would do to get it back.
He replied that:
"The enterprise is not our customer."
That says it all.
iCal Server is a good start, but it is not a complete Exchange replacement. I don't think it handles free/busy time and delegates. Plus, while many IMAP based mail servers support creation of shared mailboxes/folders, there lacks a standard interface for configuration of these mailboxes (i.e. setting permissions).
Leopard server will make this much easier. Every company needs a calendar server and finally OS X server will have it. ACLS are fully supported in Leopard and they have a new Team Directory structure which should be useful.
ACLS is already running on Linux and FreeBSD - I'm running it myself. This is not a Leopard-only feature/service.
Quote:
I'm a big fan of Time Machine...snapshots and a robust backup system are a requirement today with end user restore as a the gravy.
I'm hoping that TM will not only handle snapshots, but revision control too. Do you know anything about whether TM will support revision control?
ACLS is already running on Linux and FreeBSD - I'm running it myself. This is not a Leopard-only feature/service.
I'm hoping that TM will not only handle snapshots, but revision control too. Do you know anything about whether TM will support revision control?
Oh I know...in fact ACLS are in Tiger in a limited scope and off by default. In Leopard ACLS are on by default and much more pervasive. I certainly don't want to say ACL support is some advantage of Apple...they're mere catching up with the crowd here.
I've seen nothing yet in TM that allows for any kind of revision control. I've love to see this in TM 2.0 along with remote replication/Disaster Recovery.
Apple is going to have to craft together the disparate pieces (iCal svr, mail svr, Wiki svr, Spotlight svr, iChat svr etc) and make'em all sing in harmony without the admin having to juggle too much.
I hope they update the Leopard Server "Sneek Peek" page next week with more information.
Oh I know...in fact ACLS are in Tiger in a limited scope and off by default. In Leopard ACLS are on by default and much more pervasive. I certainly don't want to say ACL support is some advantage of Apple...they're mere catching up with the crowd here.
I've seen nothing yet in TM that allows for any kind of revision control. I've love to see this in TM 2.0 along with remote replication/Disaster Recovery.
Apple is going to have to craft together the disparate pieces (iCal svr, mail svr, Wiki svr, Spotlight svr, iChat svr etc) and make'em all sing in harmony without the admin having to juggle too much.
I hope they update the Leopard Server "Sneek Peek" page next week with more information.
Sorry, I used the wrong acronym here. I meant to say "DCS" (Darwin Calendar Server) is running on Linux and FreeBSD, but ACLs are too
Sorry, I used the wrong acronym here. I meant to say "DCS" (Darwin Calendar Server) is running on Linux and FreeBSD, but ACLs are too
Apple doesn't have to get ahead of the other players, they simply have to catch up.
The server product does offer easy management. It does offer a good price once 2003 server client license costs are taken into account. But Apple doesn't go far enough. I think they are moving in that direction, but I'm not certain of what goal they have in mind. If they are attempting to move into mid size firms in a stealthy manner, they are so stealthy, that they aren't being seen.
Apple doesn't have to get ahead of the other players, they simply have to catch up.
The server product does offer easy management. It does offer a good price once 2003 server client license costs are taken into account. But Apple doesn't go far enough. I think they are moving in that direction, but I'm not certain of what goal they have in mind. If they are attempting to move into mid size firms in a stealthy manner, they are so stealthy, that they aren't being seen.
It only offers easy management if you are in an environment that can actually use Apple's GUIs... This is a vital consideration, because this is also the most compelling reason I can think of to use OS X Server at all.
It only offers easy management if you are in an environment that can actually use Apple's GUIs... This is a vital consideration, because this is also the most compelling reason I can think of to use OS X Server at all.
Actually, the articles I read in the industry computer press shows that managers DO like Apple's management. They also like the hardware. They also like the software suites installed with the machine, AND they like the open environment software that Apple has added GUI's to. They like the UNIX underpinnings, as many are comfortable with that, and they like the fact that Leopard will be fully UNIX compliant.
Overall, they get pretty good grades for what they are. It's in the mixed environments, and the lack of OS X virtualization that they have problems.
Also, they feel uncomfortable that Apple has shown no interest in blades, or larger machines, as it gives them no upgrade paths as other vendors do.
Actually, the articles I read in the industry computer press shows that managers DO like Apple's management. They also like the hardware. They also like the software suites installed with the machine, AND they like the open environment software that Apple has added GUI's to. They like the UNIX underpinnings, as many are comfortable with that, and they like the fact that Leopard will be fully UNIX compliant.
Overall, they get pretty good grades for what they are. It's in the mixed environments, and the lack of OS X virtualization that they have problems.
Also, they feel uncomfortable that Apple has shown no interest in blades, or larger machines, as it gives them no upgrade paths as other vendors do.
I see what you're saying, but what I'm saying is that it doesn't even matter whether people like or dislike Apple's GUIs or whether they like Apple's management. If the GUIs *can't* be used, it renders them useless.
I see what you're saying, but what I'm saying is that it doesn't even matter whether people like or dislike Apple's GUIs or whether they like Apple's management. If the GUIs *can't* be used, it renders them useless.
There's really no reason why they can't be used. Other than the few times when an administrator may want to do something that the GUI doesn't allow, the GUI is fine.
There's really no reason why they can't be used. Other than the few times when an administrator may want to do something that the GUI doesn't allow, the GUI is fine.
Where wouldn't the GUI be usable?
I'd be willing to bet that it would be useless in most large businesses. Most large businesses have account management departments that are responsible for maintaining identity information, various ways of handling DNS and DHCP specific to the environment, HR departments, security departments, etc. This data needs to be passed in between departments, often via SOAP calls and/or a series of scripts that push and pull data from these systems. All of this is automated, making the GUIs only useful for manual tweaks, if that (manual tweaks can also result in data becoming out of sync).
Also, if you are wanting to do some advanced Apache configuration - e.g. external authentication modules, optimization, etc. if you make these changes in the command line, will the GUI not clobber these changes? I probably wouldn't count on this. If you disagree, I'd be willing to bet that there would be other setups that may go beyond what the GUIs will cover... As environments become bigger and bigger, they become more and more complicated and specific to the environment.
I'd be willing to bet that it would be useless in most large businesses. Most large businesses have account management departments that are responsible for maintaining identity information, various ways of handling DNS and DHCP specific to the environment, HR departments, security departments, etc. This data needs to be passed in between departments, often via SOAP calls and/or a series of scripts that push and pull data from these systems. All of this is automated, making the GUIs only useful for manual tweaks, if that (manual tweaks can also result in data becoming out of sync).
Also, if you are wanting to do some advanced Apache configuration - e.g. external authentication modules, optimization, etc. if you make these changes in the command line, will the GUI not clobber these changes? I probably wouldn't count on this. If you disagree, I'd be willing to bet that there would be other setups that may go beyond what the GUIs will cover... As environments become bigger and bigger, they become more and more complicated and specific to the environment.
What have I been saying to you?
That Apple is not after big business. That Apple publicly shunned big business. I have pointed out problems that Apple will have to solve before big business would be interested, other than as a niche product.
But, no, the GUI will not "clobber" the kind of changes that you would make through terminal, or any of the other methods you can use that are allowing proper changes to be made.
That Apple is not after big business. That Apple publicly shunned big business. I have pointed out problems that Apple will have to solve before big business would be interested, other than as a niche product.
But, no, the GUI will not "clobber" the kind of changes that you would make through terminal, or any of the other methods you can use that are allowing proper changes to be made.
I see, I misunderstood something then, I apologize for that.
I did have OS X Server clobber some changes I made in Apache when I used to have to work with OS X Server, but I can't remember if this was in 10.3 or 10.4. At any rate, I would still think that a good administrator would be very careful about pushing some changes out in the GUI after making various manual tweaks and additions in the command line, but perhaps this assumption is unfounded.
At any rate, yes, we are in agreement. Like I said, I just don't know how much business is to be had in small business for Apple. I've always found Apple's server strategy to be pretty bizarre overall. They sort of make gestures at supporting larger environments in several different manifestations, but often fall short. Just my perception.
I see, I misunderstood something then, I apologize for that.
I did have OS X Server clobber some changes I made in Apache when I used to have to work with OS X Server, but I can't remember if this was in 10.3 or 10.4. At any rate, I would still think that a good administrator would be very careful about pushing some changes out in the GUI after making various manual tweaks and additions in the command line, but perhaps this assumption is unfounded.
At any rate, yes, we are in agreement. Like I said, I just don't know how much business is to be had in small business for Apple. I've always found Apple's server strategy to be pretty bizarre overall. They sort of make gestures at supporting larger environments in several different manifestations, but often fall short. Just my perception.
One example of where I think Apple's server strategy is a little bizarre:
Of all of the features missing in OS X Server, I would think that Apple creating their own Wiki and Blog software should be the least of their concerns. There are several existing open source products that handle these tasks nicely. I don't understand why Apple doesn't focus their resources on some of the more fundamental weaknesses of Server.
One example of where I think Apple's server strategy is a little bizarre:
Of all of the features missing in OS X Server, I would think that Apple creating their own Wiki and Blog software should be the least of their concerns. There are several existing open source products that handle these tasks nicely. I don't understand why Apple doesn't focus their resources on some of the more fundamental weaknesses of Server.
One thing I'm realizing is that the posters over here are much more sophisticated as a group than on at least one bigger (?) forum I've long been on (if about equally testy). And this multi-quote feature? Way cool....
Quote:
Originally Posted by AISI
In Q3 2002 worldwide server shipments totaled about 1.1 million units according to Gartner and Apple sold 7,484 units, its share of the server market was 0.68 percent.
In Q1 2007 worldwide server shipments reached just over 2.1 million units, Apple sold 8,700 Xserves, its share of the server market should be 0.4 percent.
1 - HP: 634,093 units (30.03%)
2 - Dell: 445,850 units (21.12%)
3 - IBM: 295,175 units (13.98%)
4 - Fujitsu: 81,068 units (3.84%)
5 - Sun: 79,063 units (3.74%)
8 - Acer: 14,900 units (0.71%) *
9 - Hitachi: 9,000 units (0.43%) *
10 - Apple: 8,700 units (0.41%) *
Other: 543,416 units (25.74%)
Total: 2,111,265 units (100.00%)
Gartner numbers except * IDC quoted by Digitimes
...Apple did pretty well in 2004, but at the moment with less than 10,000 units/quarter, there is nothing to write home about. Xserve shipments should reach 14,968 units just to maintain the Q3 2002 market share of 0.68 percent. Just out of curiosity, if the Xserve server market share was similar to the Mac worldwide market share, Apple would have sold approximately 52,500 units this quarter (6x more).
The article was about the current quarter -- on quarter growth of 73% in a market that declined 11% -- and their growth in business (if and when) is being driven from the outside in by workers (a few at least IT guys) who use Macs on their own. So one would expect the effect of Macs becoming hotter to have a delayed reaction in business shops. The first few quarters after Leopard Server comes out (not simultaneously with Leopard?) will be more indicative of the trend (or lack of one) than the stats provided above. And again each Xserve may be associated with the sale of considerably more of Apple's other SKU's.
The current worldwide market share therefore is not that depressive to me -- I'd be looking at the US share over the next year or two and THEN turn my gaze to the world.
If I was an IT guy, though, yeah, I'd be shaking my head over some of the things Mac Guy says, and feel some sympathy for PC guy. But in fact that snarky Mac Guy can do much biz stuff than he lets on -- and he can virtualize the rest he doesn't.
The ads are reminiscent of 7-Up's abandoned "we're the Uncola" in that sense. I suspect that when Apple's serious Mr. I'm a Mac won't be appearing in the print ad campaigns in Computerworld, nor will he be the poster boy trotted out by visiting Apple reps on corp campuses.
Quote:
Originally Posted by hmurchison
...I'll be the first to admit that Tiger server is limited in many key ways that pretty much relegate it to predominanty Mac networks and supporting roles in Windows dominated server networks.
When I price the current Xserve against the competition I see an immediate savings but that depends on what you need to do.
Leopard server will make this much easier. Every company needs a calendar server and finally OS X server will have it. ACLS are fully supported in Leopard and they have a new Team Directory structure which should be useful.
I'm a big fan of Time Machine...snapshots and a robust backup system are a requirement today with end user restore as a the gravy. Active/Active failover with the Mail server and XSAN, clustered iCal server. RADIUS for wireless networks.
Leopard is the "Tiger" for sys admins. Meaning it's going to be the first server OS that admins feel "comfortable" and relatively empowered on. Steve Jobs has stated that Apple is not an Enterprise company and he's right. They don't have to be. They need to be focused on that mid tier and SMB market where people are beginning to move to multiple servers and modern storage solutions.
....Look at the Top 3 server vendors. They aquire a new company about every month. Apple aquires a new company maybe every 6 months or so. Certainly not enough to make a play in the Enterprise.
What I think Apple will do is continue to utilize and support Open Source technologies folding those into OS X and keeping costs down and interoperability up. Evolving the ZFS support into a stable and mature product is important as well. I hope XSAN 2.0 syncs with StorNeXt 3.0
Consider Apple to be "drafting" behind the leaders. Whatever they spend millions on Apple wants to deliver without the same expenditure.
I realize the following doesn't address many of the shortcomings detailed elsewhere in this thread, and the following applies to Leopard across the board, however. according to the latest Mac OS Rumors and Apple Insider, Leopard will offer ZFS and:
"ZFS includes numerous features and optimizations found nowhere else -- from powerful RAID enhancing features, unique data structure, industry-leading volume management tools, extremely efficient partitioning/instances handling, support for extremely large volume sizes (well in excess of 1024 Terabytes -- e.g. Petabyte territory).
Of course it also supports the features offered by Leopard's advanced HFS+ implementation including Journaling.....improves the functionality of Spotlight's meta-data databases, and handles all real or virtual devices independently in what is called a "zpool" instead of the way other FS'es work by creating one "true" device and then assigining all other devices secondary/virtual status.
I/O performance, high-end server/cluster/workstation reliability & feature support, as well as failure-recovery and data security are all dramatically enhanced with ZFS. Early benchmarks show that Leopard's powerful ZFS implementation can boost performance of large RAID arrays, Flash memory drives and many other disk types by anywhere from 5 to 50% or more (in the case of large software RAID arrays composed of powerful high-end drives; hardware RAIDs also lean to the higher end of that range but the difference may not be quite as dramatic in that case)."
Quote:
Originally Posted by melgross
Apple doesn't have to get ahead of the other players, they simply have to catch up.
The server product does offer easy management. It does offer a good price once 2003 server client license costs are taken into account. But Apple doesn't go far enough. I think they are moving in that direction, but I'm not certain of what goal they have in mind. If they are attempting to move into mid size firms in a stealthy manner, they are so stealthy, that they aren't being seen.
One can infer that the server division is not Steve's highest priority by far, and it may even hang by a thread in his thinking, kept alive by a smallish group of true believers in the company. But I don't buy that entirely. Steve also said there would "never" be a video iPod.
So while I don't expect to see an Apple Enterprise division anytime soon, as workers keep whining about missing their Macs at work, and the value prop becomes clearer, I do expect the product line to stay on an upward trajectory and to hopefully differentiate into a wider range of offerings.
Quote:
Originally Posted by melgross
...articles I read in the industry computer press show that managers DO like Apple's management. They also like the hardware. They also like the software suites installed with the machine, AND they like the open environment software that Apple has added GUI's to. They like the UNIX underpinnings....
Overall, they get pretty good grades for what they are. It's in the mixed environments, and the lack of OS X virtualization that they have problems. Also, they feel uncomfortable that Apple has shown no interest in blades, or larger machines, as it gives them no upgrade paths as other vendors do.
See above. But more importantly, let's see what ends up in Leopard Server. Especially in terms of virtualization or LS and/or 3rd Party Virt. solutions.
Quote:
Originally Posted by besson3c
...I don't understand why Apple doesn't focus their resources on some of the more fundamental weaknesses of Server.
Quote:
Originally Posted by melgross
Because they don't want to.
Yet. Resources are finite things, even when you're currently the king of a rapidly evolving media serving and consuming market (which is converging and bifrucating simultaneously), and scrambling to stay there against a rising host of well-funded competitors. If Apple can do anything similar in the video and phone markets to what they've done in the music player market and achieve some of the eponymous branding status in them that they've managed in music, their restless eyes may move to getting more serious about new horizons, and business buys a lot [a LOT] of digital gear.
Even the Pixar Division of Disney would be a good customer to have, and SJ might have something to say about that.
Comments
So, how could one put these numbers into context in order to be useful?
Well, you understand that I don't have these numbers at my fingertips.
But, the server line, and the client line, are like two separate worlds, in a way.
The servers will have problems for the reasons I gave in post #15. The clients are different, but still will have difficulties because of some of those problems. But they don't have to be as colosly involved.
The question is what does a company need the machines for?
After that is determined, the rest falls into place.
Apple is developing an alternative to Exchange, but it will need installs to become widely useful. Unless Apple is willing to address industry's needs in the other areas, it won't matter.
The context is that it is up to Apple to decide whether it wants to be a player, or not.
Jobs had me cringing when he first came back to Apple.
In a press conference, he was asked about the enterprise, and what Apple would do to get it back.
He replied that:
"The enterprise is not our customer."
That says it all.
What exactly is the target market for the X-Serve? I see a few of them going into data centers, but can't understand who they offer a good value proposition for.
For my small business, I wanted to find an excuse to go with an X-Serve instead of a Dell... but the premium was almost 2x! Even factoring in consulting time to go with Linux, we saved over $1k with the Dull.
I honestly am just having trouble understanding, and would love to know. It seems like Apple really needs to come up with a more cost-effective solution if they want to make a strong push of Leopard Server.
Well that depends on your needs and the currently available hardware. I'll be the first to admit that Tiger server is limited in many key ways that pretty much relegate it to predominanty Mac networks and supporting roles in Windows dominated server networks.
When I price the current Xserve against the competition I see an immediate savings but that depends on what you need to do.
Leopard server will make this much easier. Every company needs a calendar server and finally OS X server will have it. ACLS are fully supported in Leopard and they have a new Team Directory structure which should be useful.
I'm a big fan of Time Machine...snapshots and a robust backup system are a requirement today with end user restore as a the gravy.
Active/Active failover with the Mail server and XSAN, clustered iCal server. RADIUS for wireless networks.
Leopard is the "Tiger" for sys admins. Meaning it's going to be the first server OS that admins feel "comfortable" and relatively empowered on.
Steve Jobs has stated that Apple is not an Enterprise company and he's right. They don't have to be. They need to be focused on that mid tier and SMB market where people are beginning to move to multiple servers and modern storage solutions.
Playing at the Enterprise level means committing to fast growth. Look at the Top 3 server vendors. They aquire a new company about every month. Apple aquires a new company maybe every 6 months or so. Certainly not enough to make a play in the Enterprise.
What I think Apple will do is continue to utilize and support Open Source technologies folding those into OS X and keeping costs down and interoperability up.
Evolving the ZFS support into a stable and mature product is important as well. I hope XSAN 2.0 syncs with StorNeXt 3.0
http://www.xsanity.com/article.php/20070402225250974
Distributed LAN Client enables applications on the LAN to access a shared pool of storage faster, and with a higher level of resiliency, than most traditional network sharing methods can provide. Using clustered gateways for access and an optimized communication protocol for performance, LAN Client not only extends data sharing but also delivers enterprise features such as load balancing and transparent I/O failover.
Data Reduction Storage is a specialized tier of disk that uses data de-duplication to reduce a customer’s data footprint, saving money by lowering capacity requirements and enabling data to be retained on fast recovery disk for a much longer period of time. Quantum has pioneering data de-duplication technology, a fundamental enabler for storage applications in backup and archive. StorNext is the first product of its kind to include such intuitive, user-specific data reduction technology.
Dynamic Resource Allocation increases uptime by enabling online service operations. It allows customers to scale their storage – adding new storage capacity or transparently swapping out disk arrays during hardware upgrades – while the system is active..
Consider Apple to be "drafting" behind the leaders. Whatever they spend millions on Apple wants to deliver without the same expenditure.
Well, you understand that I don't have these numbers at my fingertips.
But, the server line, and the client line, are like two separate worlds, in a way.
The servers will have problems for the reasons I gave in post #15. The clients are different, but still will have difficulties because of some of those problems. But they don't have to be as colosly involved.
The question is what does a company need the machines for?
After that is determined, the rest falls into place.
Apple is developing an alternative to Exchange, but it will need installs to become widely useful. Unless Apple is willing to address industry's needs in the other areas, it won't matter.
The context is that it is up to Apple to decide whether it wants to be a player, or not.
Jobs had me cringing when he first came back to Apple.
In a press conference, he was asked about the enterprise, and what Apple would do to get it back.
He replied that:
"The enterprise is not our customer."
That says it all.
iCal Server is a good start, but it is not a complete Exchange replacement. I don't think it handles free/busy time and delegates. Plus, while many IMAP based mail servers support creation of shared mailboxes/folders, there lacks a standard interface for configuration of these mailboxes (i.e. setting permissions).
Leopard server will make this much easier. Every company needs a calendar server and finally OS X server will have it. ACLS are fully supported in Leopard and they have a new Team Directory structure which should be useful.
ACLS is already running on Linux and FreeBSD - I'm running it myself. This is not a Leopard-only feature/service.
I'm a big fan of Time Machine...snapshots and a robust backup system are a requirement today with end user restore as a the gravy.
I'm hoping that TM will not only handle snapshots, but revision control too. Do you know anything about whether TM will support revision control?
ACLS is already running on Linux and FreeBSD - I'm running it myself. This is not a Leopard-only feature/service.
I'm hoping that TM will not only handle snapshots, but revision control too. Do you know anything about whether TM will support revision control?
Oh I know...in fact ACLS are in Tiger in a limited scope and off by default. In Leopard ACLS are on by default and much more pervasive. I certainly don't want to say ACL support is some advantage of Apple...they're mere catching up with the crowd here.
I've seen nothing yet in TM that allows for any kind of revision control. I've love to see this in TM 2.0 along with remote replication/Disaster Recovery.
Apple is going to have to craft together the disparate pieces (iCal svr, mail svr, Wiki svr, Spotlight svr, iChat svr etc) and make'em all sing in harmony without the admin having to juggle too much.
I hope they update the Leopard Server "Sneek Peek" page next week with more information.
Oh I know...in fact ACLS are in Tiger in a limited scope and off by default. In Leopard ACLS are on by default and much more pervasive. I certainly don't want to say ACL support is some advantage of Apple...they're mere catching up with the crowd here.
I've seen nothing yet in TM that allows for any kind of revision control. I've love to see this in TM 2.0 along with remote replication/Disaster Recovery.
Apple is going to have to craft together the disparate pieces (iCal svr, mail svr, Wiki svr, Spotlight svr, iChat svr etc) and make'em all sing in harmony without the admin having to juggle too much.
I hope they update the Leopard Server "Sneek Peek" page next week with more information.
Sorry, I used the wrong acronym here. I meant to say "DCS" (Darwin Calendar Server) is running on Linux and FreeBSD, but ACLs are too
Sorry, I used the wrong acronym here. I meant to say "DCS" (Darwin Calendar Server) is running on Linux and FreeBSD, but ACLs are too
Apple doesn't have to get ahead of the other players, they simply have to catch up.
The server product does offer easy management. It does offer a good price once 2003 server client license costs are taken into account. But Apple doesn't go far enough. I think they are moving in that direction, but I'm not certain of what goal they have in mind. If they are attempting to move into mid size firms in a stealthy manner, they are so stealthy, that they aren't being seen.
Apple doesn't have to get ahead of the other players, they simply have to catch up.
The server product does offer easy management. It does offer a good price once 2003 server client license costs are taken into account. But Apple doesn't go far enough. I think they are moving in that direction, but I'm not certain of what goal they have in mind. If they are attempting to move into mid size firms in a stealthy manner, they are so stealthy, that they aren't being seen.
It only offers easy management if you are in an environment that can actually use Apple's GUIs... This is a vital consideration, because this is also the most compelling reason I can think of to use OS X Server at all.
It only offers easy management if you are in an environment that can actually use Apple's GUIs... This is a vital consideration, because this is also the most compelling reason I can think of to use OS X Server at all.
Actually, the articles I read in the industry computer press shows that managers DO like Apple's management. They also like the hardware. They also like the software suites installed with the machine, AND they like the open environment software that Apple has added GUI's to. They like the UNIX underpinnings, as many are comfortable with that, and they like the fact that Leopard will be fully UNIX compliant.
Overall, they get pretty good grades for what they are. It's in the mixed environments, and the lack of OS X virtualization that they have problems.
Also, they feel uncomfortable that Apple has shown no interest in blades, or larger machines, as it gives them no upgrade paths as other vendors do.
Actually, the articles I read in the industry computer press shows that managers DO like Apple's management. They also like the hardware. They also like the software suites installed with the machine, AND they like the open environment software that Apple has added GUI's to. They like the UNIX underpinnings, as many are comfortable with that, and they like the fact that Leopard will be fully UNIX compliant.
Overall, they get pretty good grades for what they are. It's in the mixed environments, and the lack of OS X virtualization that they have problems.
Also, they feel uncomfortable that Apple has shown no interest in blades, or larger machines, as it gives them no upgrade paths as other vendors do.
I see what you're saying, but what I'm saying is that it doesn't even matter whether people like or dislike Apple's GUIs or whether they like Apple's management. If the GUIs *can't* be used, it renders them useless.
I see what you're saying, but what I'm saying is that it doesn't even matter whether people like or dislike Apple's GUIs or whether they like Apple's management. If the GUIs *can't* be used, it renders them useless.
There's really no reason why they can't be used. Other than the few times when an administrator may want to do something that the GUI doesn't allow, the GUI is fine.
Where wouldn't the GUI be usable?
There's really no reason why they can't be used. Other than the few times when an administrator may want to do something that the GUI doesn't allow, the GUI is fine.
Where wouldn't the GUI be usable?
I'd be willing to bet that it would be useless in most large businesses. Most large businesses have account management departments that are responsible for maintaining identity information, various ways of handling DNS and DHCP specific to the environment, HR departments, security departments, etc. This data needs to be passed in between departments, often via SOAP calls and/or a series of scripts that push and pull data from these systems. All of this is automated, making the GUIs only useful for manual tweaks, if that (manual tweaks can also result in data becoming out of sync).
Also, if you are wanting to do some advanced Apache configuration - e.g. external authentication modules, optimization, etc. if you make these changes in the command line, will the GUI not clobber these changes? I probably wouldn't count on this. If you disagree, I'd be willing to bet that there would be other setups that may go beyond what the GUIs will cover... As environments become bigger and bigger, they become more and more complicated and specific to the environment.
I'd be willing to bet that it would be useless in most large businesses. Most large businesses have account management departments that are responsible for maintaining identity information, various ways of handling DNS and DHCP specific to the environment, HR departments, security departments, etc. This data needs to be passed in between departments, often via SOAP calls and/or a series of scripts that push and pull data from these systems. All of this is automated, making the GUIs only useful for manual tweaks, if that (manual tweaks can also result in data becoming out of sync).
Also, if you are wanting to do some advanced Apache configuration - e.g. external authentication modules, optimization, etc. if you make these changes in the command line, will the GUI not clobber these changes? I probably wouldn't count on this. If you disagree, I'd be willing to bet that there would be other setups that may go beyond what the GUIs will cover... As environments become bigger and bigger, they become more and more complicated and specific to the environment.
What have I been saying to you?
That Apple is not after big business. That Apple publicly shunned big business. I have pointed out problems that Apple will have to solve before big business would be interested, other than as a niche product.
But, no, the GUI will not "clobber" the kind of changes that you would make through terminal, or any of the other methods you can use that are allowing proper changes to be made.
What have I been saying to you?
That Apple is not after big business. That Apple publicly shunned big business. I have pointed out problems that Apple will have to solve before big business would be interested, other than as a niche product.
But, no, the GUI will not "clobber" the kind of changes that you would make through terminal, or any of the other methods you can use that are allowing proper changes to be made.
I see, I misunderstood something then, I apologize for that.
I did have OS X Server clobber some changes I made in Apache when I used to have to work with OS X Server, but I can't remember if this was in 10.3 or 10.4. At any rate, I would still think that a good administrator would be very careful about pushing some changes out in the GUI after making various manual tweaks and additions in the command line, but perhaps this assumption is unfounded.
At any rate, yes, we are in agreement. Like I said, I just don't know how much business is to be had in small business for Apple. I've always found Apple's server strategy to be pretty bizarre overall. They sort of make gestures at supporting larger environments in several different manifestations, but often fall short. Just my perception.
I see, I misunderstood something then, I apologize for that.
I did have OS X Server clobber some changes I made in Apache when I used to have to work with OS X Server, but I can't remember if this was in 10.3 or 10.4. At any rate, I would still think that a good administrator would be very careful about pushing some changes out in the GUI after making various manual tweaks and additions in the command line, but perhaps this assumption is unfounded.
At any rate, yes, we are in agreement. Like I said, I just don't know how much business is to be had in small business for Apple. I've always found Apple's server strategy to be pretty bizarre overall. They sort of make gestures at supporting larger environments in several different manifestations, but often fall short. Just my perception.
Well then, good. I'm happy that's been resolved.
Of all of the features missing in OS X Server, I would think that Apple creating their own Wiki and Blog software should be the least of their concerns. There are several existing open source products that handle these tasks nicely. I don't understand why Apple doesn't focus their resources on some of the more fundamental weaknesses of Server.
One example of where I think Apple's server strategy is a little bizarre:
Of all of the features missing in OS X Server, I would think that Apple creating their own Wiki and Blog software should be the least of their concerns. There are several existing open source products that handle these tasks nicely. I don't understand why Apple doesn't focus their resources on some of the more fundamental weaknesses of Server.
Because they don't want to.
In Q3 2002 worldwide server shipments totaled about 1.1 million units according to Gartner and Apple sold 7,484 units, its share of the server market was 0.68 percent.
In Q1 2007 worldwide server shipments reached just over 2.1 million units, Apple sold 8,700 Xserves, its share of the server market should be 0.4 percent.
1 - HP: 634,093 units (30.03%)
2 - Dell: 445,850 units (21.12%)
3 - IBM: 295,175 units (13.98%)
4 - Fujitsu: 81,068 units (3.84%)
5 - Sun: 79,063 units (3.74%)
8 - Acer: 14,900 units (0.71%) *
9 - Hitachi: 9,000 units (0.43%) *
10 - Apple: 8,700 units (0.41%) *
Other: 543,416 units (25.74%)
Total: 2,111,265 units (100.00%)
Gartner numbers except * IDC quoted by Digitimes
...Apple did pretty well in 2004, but at the moment with less than 10,000 units/quarter, there is nothing to write home about. Xserve shipments should reach 14,968 units just to maintain the Q3 2002 market share of 0.68 percent. Just out of curiosity, if the Xserve server market share was similar to the Mac worldwide market share, Apple would have sold approximately 52,500 units this quarter (6x more).
The article was about the current quarter -- on quarter growth of 73% in a market that declined 11% -- and their growth in business (if and when) is being driven from the outside in by workers (a few at least IT guys) who use Macs on their own. So one would expect the effect of Macs becoming hotter to have a delayed reaction in business shops. The first few quarters after Leopard Server comes out (not simultaneously with Leopard?) will be more indicative of the trend (or lack of one) than the stats provided above. And again each Xserve may be associated with the sale of considerably more of Apple's other SKU's.
The current worldwide market share therefore is not that depressive to me -- I'd be looking at the US share over the next year or two and THEN turn my gaze to the world.
If I was an IT guy, though, yeah, I'd be shaking my head over some of the things Mac Guy says, and feel some sympathy for PC guy. But in fact that snarky Mac Guy can do much biz stuff than he lets on -- and he can virtualize the rest he doesn't.
The ads are reminiscent of 7-Up's abandoned "we're the Uncola" in that sense. I suspect that when Apple's serious Mr. I'm a Mac won't be appearing in the print ad campaigns in Computerworld, nor will he be the poster boy trotted out by visiting Apple reps on corp campuses.
...I'll be the first to admit that Tiger server is limited in many key ways that pretty much relegate it to predominanty Mac networks and supporting roles in Windows dominated server networks.
When I price the current Xserve against the competition I see an immediate savings but that depends on what you need to do.
Leopard server will make this much easier. Every company needs a calendar server and finally OS X server will have it. ACLS are fully supported in Leopard and they have a new Team Directory structure which should be useful.
I'm a big fan of Time Machine...snapshots and a robust backup system are a requirement today with end user restore as a the gravy. Active/Active failover with the Mail server and XSAN, clustered iCal server. RADIUS for wireless networks.
Leopard is the "Tiger" for sys admins. Meaning it's going to be the first server OS that admins feel "comfortable" and relatively empowered on. Steve Jobs has stated that Apple is not an Enterprise company and he's right. They don't have to be. They need to be focused on that mid tier and SMB market where people are beginning to move to multiple servers and modern storage solutions.
....Look at the Top 3 server vendors. They aquire a new company about every month. Apple aquires a new company maybe every 6 months or so. Certainly not enough to make a play in the Enterprise.
What I think Apple will do is continue to utilize and support Open Source technologies folding those into OS X and keeping costs down and interoperability up. Evolving the ZFS support into a stable and mature product is important as well. I hope XSAN 2.0 syncs with StorNeXt 3.0
http://www.xsanity.com/article.php/20070402225250974
Consider Apple to be "drafting" behind the leaders. Whatever they spend millions on Apple wants to deliver without the same expenditure.
I realize the following doesn't address many of the shortcomings detailed elsewhere in this thread, and the following applies to Leopard across the board, however. according to the latest Mac OS Rumors and Apple Insider, Leopard will offer ZFS and:
"ZFS includes numerous features and optimizations found nowhere else -- from powerful RAID enhancing features, unique data structure, industry-leading volume management tools, extremely efficient partitioning/instances handling, support for extremely large volume sizes (well in excess of 1024 Terabytes -- e.g. Petabyte territory).
Of course it also supports the features offered by Leopard's advanced HFS+ implementation including Journaling.....improves the functionality of Spotlight's meta-data databases, and handles all real or virtual devices independently in what is called a "zpool" instead of the way other FS'es work by creating one "true" device and then assigining all other devices secondary/virtual status.
I/O performance, high-end server/cluster/workstation reliability & feature support, as well as failure-recovery and data security are all dramatically enhanced with ZFS. Early benchmarks show that Leopard's powerful ZFS implementation can boost performance of large RAID arrays, Flash memory drives and many other disk types by anywhere from 5 to 50% or more (in the case of large software RAID arrays composed of powerful high-end drives; hardware RAIDs also lean to the higher end of that range but the difference may not be quite as dramatic in that case)."
Apple doesn't have to get ahead of the other players, they simply have to catch up.
The server product does offer easy management. It does offer a good price once 2003 server client license costs are taken into account. But Apple doesn't go far enough. I think they are moving in that direction, but I'm not certain of what goal they have in mind. If they are attempting to move into mid size firms in a stealthy manner, they are so stealthy, that they aren't being seen.
One can infer that the server division is not Steve's highest priority by far, and it may even hang by a thread in his thinking, kept alive by a smallish group of true believers in the company. But I don't buy that entirely. Steve also said there would "never" be a video iPod.
So while I don't expect to see an Apple Enterprise division anytime soon, as workers keep whining about missing their Macs at work, and the value prop becomes clearer, I do expect the product line to stay on an upward trajectory and to hopefully differentiate into a wider range of offerings.
...articles I read in the industry computer press show that managers DO like Apple's management. They also like the hardware. They also like the software suites installed with the machine, AND they like the open environment software that Apple has added GUI's to. They like the UNIX underpinnings....
Overall, they get pretty good grades for what they are. It's in the mixed environments, and the lack of OS X virtualization that they have problems. Also, they feel uncomfortable that Apple has shown no interest in blades, or larger machines, as it gives them no upgrade paths as other vendors do.
See above. But more importantly, let's see what ends up in Leopard Server. Especially in terms of virtualization or LS and/or 3rd Party Virt. solutions.
...I don't understand why Apple doesn't focus their resources on some of the more fundamental weaknesses of Server.
Because they don't want to.
Yet. Resources are finite things, even when you're currently the king of a rapidly evolving media serving and consuming market (which is converging and bifrucating simultaneously), and scrambling to stay there against a rising host of well-funded competitors. If Apple can do anything similar in the video and phone markets to what they've done in the music player market and achieve some of the eponymous branding status in them that they've managed in music, their restless eyes may move to getting more serious about new horizons, and business buys a lot [a LOT] of digital gear.
Even the Pixar Division of Disney would be a good customer to have, and SJ might have something to say about that.