Still, it is sad that there is not a second NIC port on the server model and an eSATA port across the line.
The one thing that is encouraging about this product release is that it was not accompanied with a great dog and pony show scheduled for some time that has nothing to do with when the product was ready. Perhaps, just perhaps, Apple is beginning to realize that it should intro a product line when it is "good to go".
Personally, I remain very disappointed that at least one eSATA port was not included with the new Minis as it severely limits the uses I had envisioned for it as a HTPC and/or home server.
Yes Apple certainly hasn't been fond of eSATA and I certainly don't expect that attitude to change. However eSATA's roadmap is pretty much defunct. Intel just announced that they're putting USB 3 on the backburner until they can integrate it (cheaply) into the chipsets in 2011. To this ads credence to the reports that Intel and Apple are moving rapidly towards Light Peak. If you're intel ..what would you do? Focus on USB 3 which is primarily going to benefit storage products and eSATA which is the same or would you leverage a 10Gbp duplex fiber connection that will support multiple protocol and come with latency that should be phenomenal.
I've seen the "light" and in a few years a mini class server will come with SSD that can read 500MBps and will attach to "network" storage via Light Peak at speeds and latency that has few peers.
Yes Apple certainly hasn't been fond of eSATA and I certainly don't expect that attitude to change. However eSATA's roadmap is pretty much defunct. Intel just announced that they're putting USB 3 on the backburner until they can integrate it (cheaply) into the chipsets in 2011. To this ads credence to the reports that Intel and Apple are moving rapidly towards Light Peak. If you're intel ..what would you do? Focus on USB 3 which is primarily going to benefit storage products and eSATA which is the same or would you leverage a 10Gbp duplex fiber connection that will support multiple protocol and come with latency that should be phenomenal.
I've seen the "light" and in a few years a mini class server will come with SSD that can read 500MBps and will attach to "network" storage via Light Peak at speeds and latency that has few peers.
I very much agree that Light Peak is a promising technology and form factor for the future. eSATA, on the other hand, is a reasonably mature technology that has been deployed for some time now and is a well know, and accepted, stand which provides demonstrably superior performance when compared to other currently deployed technologies appropriate for widespread use. There really is no acceptable excuse, at least to me, for Apple not having adopted it. Unlike the question you pose (and fairly so) about Intel devoting resources to the development of Light Peak vs. the development of USB 3, eSATA has off-the-shelf parts to implement it at the present time.
I very much agree that Light Peak is a promising technology and form factor for the future. eSATA, on the other hand, is a reasonably mature technology that has been deployed for some time now and is a well know, and accepted, stand which provides demonstrably superior performance when compared to other currently deployed technologies appropriate for widespread use. There really is no acceptable excuse, at least to me, for Apple not having adopted it. Unlike the question you pose (and fairly so) about Intel devoting resources to the development of Light Peak vs. the development of USB 3, eSATA has off-the-shelf parts to implement it at the present time.
Oh well, they did not ask me.
Cheers
OK, fine, but do you think it would have been wise for Apple to engineer an entirely new connector type for the mini when it's dubious whether 95% of the potential customers would ever use it?
It should be clear that Apple's design goal was to create a functional server out of the Mac mini, knowing that some customers had already done this on their own. Conversely, I doubt their design goal was to shoehorn a fully featured server spec into a computer that wasn't originally designed for that purpose. That approach would have required rethinking the whole computer, IMHO, and for what would have resulted in benefits to a small number of mini purchasers.
Apple is doing the right thing by waiting for what they expect to be the coming standard in connections... Light Peak. Why waste time, energy, and money jerry-rigging stop-gap solutions?
OK, fine, but do you think it would have been wise for Apple to engineer an entirely new connector type for the mini when it's dubious whether 95% of the potential customers would ever use it?
It should be clear that Apple's design goal was to create a functional server out of the Mac mini, knowing that some customers had already done this on their own. Conversely, I doubt their design goal was to shoehorn a fully featured server spec into a computer that wasn't originally designed for that purpose. That approach would have required rethinking the whole computer, IMHO, and for what would have resulted in benefits to a small number of mini purchasers.
Apple is doing the right thing by waiting for what they expect to be the coming standard in connections... Light Peak. Why waste time, energy, and money jerry-rigging stop-gap solutions?
You must have skipped over the "off-the-shelf parts" portion of my comment. In fact there is an off the shelf part that is a combination USB/eSATA port which is very nice, indeed. It is not some "jerry-rigging stop-gap solution". There is already a SATA controller in the Mini (and the well known hack to run it outside the case). The beauty of SATA is that there is no difference between an internally mounted drive and an externally mounted drive in so far as performance is concerned.
The Mini is not "a fully featured server". Doing these comparatively simple things would not make it one in any event, as you well know.
Unless Apple delivers a fully configured Light Peak Mini in Jan 2010, your argument will not have proven to be accurate. If it is, then the current release will be demonstrated to be merely a "place holder" which is soon to be forgotten.
In any event, Apple probably did not ask you either.
You must have skipped over the "off-the-shelf parts" portion of my comment. In fact there is an off the shelf part that is a combination USB/eSATA port which is very nice, indeed. It is not some "jerry-rigging stop-gap solution". There is already a SATA controller in the Mini (and the well known hack to run it outside the case). The beauty of SATA is that there is no difference between an internally mounted drive and an externally mounted drive in so far as performance is concerned.
The Mini is not "a fully featured server". Doing these comparatively simple things would not make it one in any event, as you well know.
Unless Apple delivers a fully configured Light Peak Mini in Jan 2010, your argument will not have proven to be accurate. If it is, then the current release will be demonstrated to be merely a "place holder" which is soon to be forgotten.
In any event, Apple probably did not ask you either.
Cheers
I understand where you're coming from but I suppose every computer, to some extent, is a place holder until the next one comes along.
However, I should have been more clear... I suspect Apple's goal was not to redesign the mini's case for a connector technology which they likely consider to be a short term solution. Probably we can all agree that the percentage of mini customers looking for eSATA is quite low.
Low percent looking for eSATA + need for case redesign + short term life of redesign = don't go forward with eSATA.
That's how I suspect Apple looked at this when deciding how to make this OS X Server version of the mini.
I understand where you're coming from but I suppose every computer, to some extent, is a place holder until the next one comes along.
However, I should have been more clear... I suspect Apple's goal was not to redesign the mini's case for a connector technology which they likely consider to be a short term solution. Probably we can all agree that the percentage of mini customers looking for eSATA is quite low.
Low percent looking for eSATA + need for case redesign + short term life of redesign = don't go forward with eSATA.
That's how I suspect Apple looked at this when deciding how to make this OS X Server version of the mini.
Yes, the number of people willing to hack their brand new Mini to be able to use eSATA is bound to be a small percentage. Sadly, the others probably do not realize just what they are missing and hence are not standing on their soap boxes demanding it. ;-)
There is a very great difference in the performance of a SATA connection and any other consumer level attachment.
I guess we will see where things go after the first of the year.
I already have a MacMini 2.26 that I upgraded to twin 500GB drives and 4GB RAM.
I want one with dual NICs so I can use the DroboPro with it.
Others want dual NICs for fail over, firewalling, load balancing etc, something that almost every commercially available NAS device provides.
NAS devices like the Qnap do a good job of providing a very fast device with low power usage BUT they have a propietory software interface which you cannot alter (you have to use their applications).
I want a low power server with redundant storage that I can control and if possible, I would like it to be a Mac server.
I already have a Dell server with hardware RAID and it eats power and is as noisy a hair drier
Apple has still not filled this gap in their product offering but something as simple as another NIC would solve this. I also understand why they did this.
It's a lot easier to redesign a top case without a DVD slot than it is to redesign the motherboard and backplane.
I'm hoping that sales will be good and they will add another NIC in a future revision...
Their current redundant storage offering for the new MacMini server is an external Promise RAID box connected via FW800
NAS devices like the Qnap do a good job of providing a very fast device with low power usage BUT they have a propietory software interface which you cannot alter (you have to use their applications).
AFAIK Qnap uses a customized linux install on their NAS boxes just as a lot of NAS manufacturers do (I have a Synology DS409 for home use. It uses a custom linux base install I can replace or hack at if I want). Not quite the same as what most people mean when they say proprietary, and certainly not unalterable.
Quote:
I want a low power server with redundant storage that I can control and if possible, I would like it to be a Mac server.
I already have a Dell server with hardware RAID and it eats power and is as noisy a hair drier
I know it's not apple, but I do heartily recommend the synology line for home use (I cannot speak to qnap devices as I havent really used them).
As for apple's offerings... you could always run sata cables out from the mac mini to external enclosures, though it'd be a bit ugly and kludgy...
At work I also have a power hungry, wind-tunnel Dell server running Debian with a Dell MD1000 hanging off it for extra storage space for our primary group scratch space. It sits in a machine room along with a couple small computer clusters, so the sound really isnt an issue :-p. The right tool for the right job....
What most of the posts here are ignoring is the presence of OSX Server, and what it brings to the table.
The Mini isn't a great fileserver. We've established that. It "only" supports 1GBps. "Pathetic", I know. But for 99.999% of small businesses, it'll do just fine.
What it can do is run all the other little services just fine. Give me one place I can define a user, through an easy-to-understand GUI, and have it handle VPN access, email access (if not hosted elsewhere), private iChat configuration, secure LDAP authentication to other devices, a protected internal wiki that's easy for everyone to use, calendaring services...
And then with one click remove that access when someone moves on. Oh, and let me force password changes, etc, as needed, although everyone wants just one password to remember.
For the average small business office, keeping up with that kind of stuff is a real pain in the ass. Even for us it gets annoying, and we're a mac-based development shop! Solving that issue for $1K will really help a lot of people.
Oh, and if you also move 10TB of files everywhere all the time, get a great fileserver as well. It can talk to the MiniServer for user authentication. But that's not the Mini's job. But saving the odd presentation and sales contract? It'll do that just fine.
I do not like the idea of having my ethernet connection and my external HD on the same controller (or is there two USB controllers in the mac mini ?)
But it might do the trick if it is connected to the modem indeed...
It's already covered as a bad idea. It's also only 10/100, even though USB has a bitrate of up to 400Mbps. You can split the drive controller from the Eth by using a Firewire drive, that's certainly a separate controller. I have the FW800 basic Drobo, among other single drive enclosures. I wanted to get the DroboPro, you can get nearly four basic Drobos for that price though. Seems pretty nice, but I can do other things with the extra money.
Comments
Still, it is sad that there is not a second NIC port on the server model and an eSATA port across the line.
The one thing that is encouraging about this product release is that it was not accompanied with a great dog and pony show scheduled for some time that has nothing to do with when the product was ready. Perhaps, just perhaps, Apple is beginning to realize that it should intro a product line when it is "good to go".
Personally, I remain very disappointed that at least one eSATA port was not included with the new Minis as it severely limits the uses I had envisioned for it as a HTPC and/or home server.
Yes Apple certainly hasn't been fond of eSATA and I certainly don't expect that attitude to change. However eSATA's roadmap is pretty much defunct. Intel just announced that they're putting USB 3 on the backburner until they can integrate it (cheaply) into the chipsets in 2011. To this ads credence to the reports that Intel and Apple are moving rapidly towards Light Peak. If you're intel ..what would you do? Focus on USB 3 which is primarily going to benefit storage products and eSATA which is the same or would you leverage a 10Gbp duplex fiber connection that will support multiple protocol and come with latency that should be phenomenal.
I've seen the "light" and in a few years a mini class server will come with SSD that can read 500MBps and will attach to "network" storage via Light Peak at speeds and latency that has few peers.
Yes Apple certainly hasn't been fond of eSATA and I certainly don't expect that attitude to change. However eSATA's roadmap is pretty much defunct. Intel just announced that they're putting USB 3 on the backburner until they can integrate it (cheaply) into the chipsets in 2011. To this ads credence to the reports that Intel and Apple are moving rapidly towards Light Peak. If you're intel ..what would you do? Focus on USB 3 which is primarily going to benefit storage products and eSATA which is the same or would you leverage a 10Gbp duplex fiber connection that will support multiple protocol and come with latency that should be phenomenal.
I've seen the "light" and in a few years a mini class server will come with SSD that can read 500MBps and will attach to "network" storage via Light Peak at speeds and latency that has few peers.
I very much agree that Light Peak is a promising technology and form factor for the future. eSATA, on the other hand, is a reasonably mature technology that has been deployed for some time now and is a well know, and accepted, stand which provides demonstrably superior performance when compared to other currently deployed technologies appropriate for widespread use. There really is no acceptable excuse, at least to me, for Apple not having adopted it. Unlike the question you pose (and fairly so) about Intel devoting resources to the development of Light Peak vs. the development of USB 3, eSATA has off-the-shelf parts to implement it at the present time.
Oh well, they did not ask me.
Cheers
I very much agree that Light Peak is a promising technology and form factor for the future. eSATA, on the other hand, is a reasonably mature technology that has been deployed for some time now and is a well know, and accepted, stand which provides demonstrably superior performance when compared to other currently deployed technologies appropriate for widespread use. There really is no acceptable excuse, at least to me, for Apple not having adopted it. Unlike the question you pose (and fairly so) about Intel devoting resources to the development of Light Peak vs. the development of USB 3, eSATA has off-the-shelf parts to implement it at the present time.
Oh well, they did not ask me.
Cheers
OK, fine, but do you think it would have been wise for Apple to engineer an entirely new connector type for the mini when it's dubious whether 95% of the potential customers would ever use it?
It should be clear that Apple's design goal was to create a functional server out of the Mac mini, knowing that some customers had already done this on their own. Conversely, I doubt their design goal was to shoehorn a fully featured server spec into a computer that wasn't originally designed for that purpose. That approach would have required rethinking the whole computer, IMHO, and for what would have resulted in benefits to a small number of mini purchasers.
Apple is doing the right thing by waiting for what they expect to be the coming standard in connections... Light Peak. Why waste time, energy, and money jerry-rigging stop-gap solutions?
OK, fine, but do you think it would have been wise for Apple to engineer an entirely new connector type for the mini when it's dubious whether 95% of the potential customers would ever use it?
It should be clear that Apple's design goal was to create a functional server out of the Mac mini, knowing that some customers had already done this on their own. Conversely, I doubt their design goal was to shoehorn a fully featured server spec into a computer that wasn't originally designed for that purpose. That approach would have required rethinking the whole computer, IMHO, and for what would have resulted in benefits to a small number of mini purchasers.
Apple is doing the right thing by waiting for what they expect to be the coming standard in connections... Light Peak. Why waste time, energy, and money jerry-rigging stop-gap solutions?
You must have skipped over the "off-the-shelf parts" portion of my comment. In fact there is an off the shelf part that is a combination USB/eSATA port which is very nice, indeed. It is not some "jerry-rigging stop-gap solution". There is already a SATA controller in the Mini (and the well known hack to run it outside the case). The beauty of SATA is that there is no difference between an internally mounted drive and an externally mounted drive in so far as performance is concerned.
The Mini is not "a fully featured server". Doing these comparatively simple things would not make it one in any event, as you well know.
Unless Apple delivers a fully configured Light Peak Mini in Jan 2010, your argument will not have proven to be accurate. If it is, then the current release will be demonstrated to be merely a "place holder" which is soon to be forgotten.
In any event, Apple probably did not ask you either.
Cheers
You must have skipped over the "off-the-shelf parts" portion of my comment. In fact there is an off the shelf part that is a combination USB/eSATA port which is very nice, indeed. It is not some "jerry-rigging stop-gap solution". There is already a SATA controller in the Mini (and the well known hack to run it outside the case). The beauty of SATA is that there is no difference between an internally mounted drive and an externally mounted drive in so far as performance is concerned.
The Mini is not "a fully featured server". Doing these comparatively simple things would not make it one in any event, as you well know.
Unless Apple delivers a fully configured Light Peak Mini in Jan 2010, your argument will not have proven to be accurate. If it is, then the current release will be demonstrated to be merely a "place holder" which is soon to be forgotten.
In any event, Apple probably did not ask you either.
Cheers
I understand where you're coming from but I suppose every computer, to some extent, is a place holder until the next one comes along.
However, I should have been more clear... I suspect Apple's goal was not to redesign the mini's case for a connector technology which they likely consider to be a short term solution. Probably we can all agree that the percentage of mini customers looking for eSATA is quite low.
Low percent looking for eSATA + need for case redesign + short term life of redesign = don't go forward with eSATA.
That's how I suspect Apple looked at this when deciding how to make this OS X Server version of the mini.
I understand where you're coming from but I suppose every computer, to some extent, is a place holder until the next one comes along.
However, I should have been more clear... I suspect Apple's goal was not to redesign the mini's case for a connector technology which they likely consider to be a short term solution. Probably we can all agree that the percentage of mini customers looking for eSATA is quite low.
Low percent looking for eSATA + need for case redesign + short term life of redesign = don't go forward with eSATA.
That's how I suspect Apple looked at this when deciding how to make this OS X Server version of the mini.
Yes, the number of people willing to hack their brand new Mini to be able to use eSATA is bound to be a small percentage. Sadly, the others probably do not realize just what they are missing and hence are not standing on their soap boxes demanding it. ;-)
There is a very great difference in the performance of a SATA connection and any other consumer level attachment.
I guess we will see where things go after the first of the year.
Cheers
Yes, I'm being sarcastic and NO, the Drobo familly will NOT work with 10.6 in 64 bit mode, which means Snow Leopard Server...
In all seriousness, for a low power server it would be ideal in this configuration with a DroboPro but it just isn't going to happen.
This leaves a MacPro or Xserve (or older G5 PowerMac with dual NICs) running 10.5 server as the only Mac server options for DroboPro...
ET
I already have a MacMini 2.26 that I upgraded to twin 500GB drives and 4GB RAM.
I want one with dual NICs so I can use the DroboPro with it.
Others want dual NICs for fail over, firewalling, load balancing etc, something that almost every commercially available NAS device provides.
NAS devices like the Qnap do a good job of providing a very fast device with low power usage BUT they have a propietory software interface which you cannot alter (you have to use their applications).
I want a low power server with redundant storage that I can control and if possible, I would like it to be a Mac server.
I already have a Dell server with hardware RAID and it eats power and is as noisy a hair drier
Apple has still not filled this gap in their product offering but something as simple as another NIC would solve this. I also understand why they did this.
It's a lot easier to redesign a top case without a DVD slot than it is to redesign the motherboard and backplane.
I'm hoping that sales will be good and they will add another NIC in a future revision...
Their current redundant storage offering for the new MacMini server is an external Promise RAID box connected via FW800
ET
NAS devices like the Qnap do a good job of providing a very fast device with low power usage BUT they have a propietory software interface which you cannot alter (you have to use their applications).
AFAIK Qnap uses a customized linux install on their NAS boxes just as a lot of NAS manufacturers do (I have a Synology DS409 for home use. It uses a custom linux base install I can replace or hack at if I want). Not quite the same as what most people mean when they say proprietary, and certainly not unalterable.
I want a low power server with redundant storage that I can control and if possible, I would like it to be a Mac server.
I already have a Dell server with hardware RAID and it eats power and is as noisy a hair drier
I know it's not apple, but I do heartily recommend the synology line for home use (I cannot speak to qnap devices as I havent really used them).
As for apple's offerings... you could always run sata cables out from the mac mini to external enclosures, though it'd be a bit ugly and kludgy...
At work I also have a power hungry, wind-tunnel Dell server running Debian with a Dell MD1000 hanging off it for extra storage space for our primary group scratch space. It sits in a machine room along with a couple small computer clusters, so the sound really isnt an issue :-p. The right tool for the right job....
You must have missed the "for me personally" part
I already have a MacMini 2.26 that I upgraded to twin 500GB drives and 4GB RAM.
I want one with dual NICs so I can use the DroboPro with it.
Then just sell the mini and the hair dryer (Dell) server and buy a Mac Pro. What's so hard about that?
The Mini isn't a great fileserver. We've established that. It "only" supports 1GBps. "Pathetic", I know. But for 99.999% of small businesses, it'll do just fine.
What it can do is run all the other little services just fine. Give me one place I can define a user, through an easy-to-understand GUI, and have it handle VPN access, email access (if not hosted elsewhere), private iChat configuration, secure LDAP authentication to other devices, a protected internal wiki that's easy for everyone to use, calendaring services...
And then with one click remove that access when someone moves on. Oh, and let me force password changes, etc, as needed, although everyone wants just one password to remember.
For the average small business office, keeping up with that kind of stuff is a real pain in the ass. Even for us it gets annoying, and we're a mac-based development shop! Solving that issue for $1K will really help a lot of people.
Oh, and if you also move 10TB of files everywhere all the time, get a great fileserver as well. It can talk to the MiniServer for user authentication. But that's not the Mini's job. But saving the odd presentation and sales contract? It'll do that just fine.
[USB to Eth adaptater]
I do not like the idea of having my ethernet connection and my external HD on the same controller (or is there two USB controllers in the mac mini ?)
But it might do the trick if it is connected to the modem indeed...
I do not like the idea of having my ethernet connection and my external HD on the same controller (or is there two USB controllers in the mac mini ?)
But it might do the trick if it is connected to the modem indeed...
It's already covered as a bad idea. It's also only 10/100, even though USB has a bitrate of up to 400Mbps. You can split the drive controller from the Eth by using a Firewire drive, that's certainly a separate controller. I have the FW800 basic Drobo, among other single drive enclosures. I wanted to get the DroboPro, you can get nearly four basic Drobos for that price though. Seems pretty nice, but I can do other things with the extra money.