Which is more important to you, unRAID 5.0 features or 3TB drive support?

[PeterB]

A 'stable' release is simply a version of the code which has been exercised sufficiently for most users to be comfortable with entrusting their data to it.

 

I disagree.  IMHO stable means 1) no known bugs and ...

 

So that means that a 'stable' release becomes 'unstable' as soon as a bug is discovered?

 

I wonder, then, which release(s) can currently be considered 'stable'?

[lionelhutz]

lionelhutz - If you are not putting data on your hard drive what do you use it for?

 

unRAID is not a backup. The parity provides a means to easily recover from a failed hard drive. This is not the same thing as a data back-up, which you seem to be implying unRAID is, and should further be by implimenting double parity.

 

Peter

 

[Auggie]

I'm not sure what ROI is supposed to mean in this context; there is no "return" and thus the term doesn't really apply here. If you're talking about cost-per-TB, the 3TB drives are considerably more expensive at about $50/TB compared to $35/TB for 2TB drives. If the argument is that some folks need 15TB servers today, or if the argument is that it makes more sense to be "future-proof" and get 3TB drives today, then go ahead and do that with 3TB drives today. You'll be able to "unlock" the extra 800GB/drive as soon as 5.1 is released. If 5.1 is a small release focused primarily on 2.2TB+ support, the wait shouldn't be long at all.

 

"Real artists ship."  

 

 

To the Esteemable Mr. SnowLprd:

 

I can't believe you are arguing on the usage of terminology here.  But so be it..

 

The term "ROI (Return On Investment)" is not just limited to just monetary value as you clearly wish to narrow it's scope to.

 

It is a general accounting term to evaluate the value of "an investment, business decision or action" and "has become a central financial metric for asset purchase decisions (computer systems, factory machines, or service vehicles, for example)."

 

"Value" as used in these contexts is subjective, based upon the each entity's own criteria for calculating value, to include usage such as longevity and savings.  For example, I may "value" my 1966 Mustang GT A-code Fastback as "priceless" (or at least north of $30,000) but someone else may value it as worthless.

 

All motherboards have a finite number of expansion ports (both in built-in SATA ports as well as add-on cards with additional SATA ports).  I value my limited expansion ports highly, therefore, higher capacity drives have a higher "value" to me, ergo 3TB drives have a higher ROI for my situation.  That is, my "return" on my investment in larger capacity drives means I can store more data with less cost than purchasing lower capacity hard drives.

 

Regarding your other point of a "small release" 5.1 where the wait "shouldn't be long", you are truly now just positing a WAG (speculating) as it appears no one knows what will go into 5.1 and when it will be released as even 5.0 is still unsettled and has been in a perpetual "beta" mode for close to a year or more.

 

I really didn't want to inflame the already "heated" debates posted in this thread, so I apologize to those that felt I'm just adding to the "noise" here as it contributes no "value" whatsoever to the topic at hand.  I will desist henceforth on this particular point.

 

My original post was simply to provide my own opinions of what should be included to support my vote in the poll as this is what I believe is the original intent of the OP for creating this thread, and not to bicker or counter-point other views as I realize everyone has their own opinions as to what they perceive as high priority features.

[squirrellydw]

I can't believe this is still going on, the poll clearly shows people want 5.0 more than 3tb support by a wide margin.

[bubbaQ]

So that means that a 'stable' release becomes 'unstable' as soon as a bug is discovered?

 

No.  You just don't release any version labeled "stable" if beta testing or RC testing discovered a bug.  5.0b6a has a bug, and thus is not a candidate for a "stable" release until that bug is quashed.

 

Of course a stable release may eventually be discovered to have a bug.  I'm saying you don't label any release as "stable" if it has a known bug when you release it.  This of course contemplates that you have a beta/RC cycle designed to find bugs before labeling any release as "stable." 

[SnowLprd]

The term "ROI (Return On Investment)" is not just limited to just monetary value as you clearly wish to narrow it's [sic] scope to.

 

I have a master's degree in finance, so suffice it to say that I understand the definition of ROI quite well, thank you. Your conflation of the term with the subjective notion of "value," while interesting, doesn't hold water.

 

All motherboards have a finite number of expansion ports (both in built-in SATA ports as well as add-on cards with additional SATA ports).  I value my limited expansion ports highly, therefore, higher capacity drives have a higher "value" to me, ergo 3TB drives have a higher ROI for my situation.  That is, my "return" on my investment in larger capacity drives means I can store more data with less cost than purchasing lower capacity hard drives.

 

I don't know how much data you feel you need to store, so I'm not going to try to understand how you are doing your math. I will, however, say this... At $100 for an 8-port SATA card (that's $12.50 per port), I find it hard to believe that there are a significant number of people whose storage needs are so great that it's less expensive to (a) buy 3TB drives at today's cost than to (b) use 2TB drives with an expanded number of ports. But like I said, I don't know your situation. Maybe you've done the math, and maybe it still works out better for you with 3TB drives. In which case, like I said before, just do that -- the other 800GB/drive will be available to you soon enough.

 

My original post was simply to provide my own opinions of what should be included to support my vote in the poll as this is what I believe is the original intent of the OP for creating this thread, and not to bicker or counter-point other views as I realize everyone has their own opinions as to what they perceive as high priority features.

 

Well said. And as Squirrellydw has already pointed out, the poll clearly shows people want 5.0 released more than they want 2.2TB+ support — by a wide margin.

[blockhead]

...the poll clearly shows people want 5.0 released more than they want 2.2TB+ support — by a wide margin.

 

Wowza!  I can't imagine a better reason to register than to respond to that statement.  What the poll clearly shows is that you do not know how to read polls.  The poll clearly shows 51.5% of a few hundred current unraid users who inhabit this forum and like to vote in polls think 3T support should not be a priority.  That leaves 48.5% expressing a desire for 3T support.  9 votes - that's your "wide margin"?

 

If these are the numbers generated by current users, imagine a poll of would-be users.  How would the numbers shift if you put this question to a few hundred prospective buyers?  Yes, that's right, it would be a landslide for 3T support, because no one is looking to build a new massive storage system based on software that doesn't support the largest readily available hard drives.

 

As a happy user of unraid for several months now, I am grateful for the existence of this software more than I can adequately express.  I bought two licenses because I'm far more afraid of Tom losing interest in unraid than I am afraid of a flash drive failure.  As a current user with a 2T based system with plenty of room, the best thing for me is a stable 5.0 with no 3T support since I won't need it for quite some time.  But the best thing for Lime Tech is to snag as many fresh fish as it can by supporting the drives the fish want to use.  That's my vote, and if you care at all about unraid succeeding as an enterprise, that should be your vote too. 

[intertan]

common I need 5.0 for my soon to be built setup based on sandy bridge. these 3tb drives, if you use these you must have a lot of stuff you want to keep. Wouldn't you rather have a stable server with the overhead if drives fail? I know my 2tb drives were to go I would be out 30+ hours of my time.

[mrklaw]

If you have a huge tower machine converted to hold 20 disks, then perhaps 3TB isn't that important to you. Or maybe you've just forked out on a bunch of 2TB disks so don't want to upgrade yet. Doesn't mean 3TB support isn't important.

 

what, functionally, in 5.0 adds specific value to a large number of users? AFP? Maybe. I'm a mac user and should appreciate it. but for media serving samba works fine. Or NFS. Active directory? Depends what interest Lime-tech have from businesses. It may well be a priority to add to further their business. New gui? pah. I'd argue that the most interesting is the new plugin architecture. Look at the customisation forum, with people asking how to get x and y added to unraid. Now go and look at the qnap and synology forums where similar questions are being asked. Supporting 'application x' via plugins I would argue is critical to making this a real consumer/soho solution to compete with synology/qnap and others.

 

Also, consider a normal home user. Perhaps they're looking at 2 bay or maybe a 4 bay solution. In those cases there is clear benefit from offering support for 3TB drives where the physical limitations prevent easy expansion - you can't just buy another SATA controller if you can only physically fit 4 disks in the chassis.

 

 

I don't really answer the question - I'd argue that plugins and 3TB support are the most important new features to add, and they conflict with this question - one being reliant on 5.0.

 

 

[boof]

I don't know how much data you feel you need to store, so I'm not going to try to understand how you are doing your math. I will, however, say this... At $100 for an 8-port SATA card (that's $12.50 per port), I find it hard to believe that there are a significant number of people whose storage needs are so great that it's less expensive to (a) buy 3TB drives at today's cost than to (b) use 2TB drives with an expanded number of ports. But like I said, I don't know your situation. Maybe you've done the math, and maybe it still works out better for you with 3TB drives. In which case, like I said before, just do that -- the other 800GB/drive will be available to you soon enough.

 

In answer to :

 

a) your costs ignore all the other costs per port. Presuming you're running unraid to primarily serve data then you have the cost of your entire server against the number of ports it can serve. In addition to the literal physical sata ports you need to power the drives, you need somewhere physically to put them (cost increasing if you're using a hotswap type bay) etc etc. Take the entire cost of your server, divide by the number of sata ports it has and you have your per port cost. Then add on the cost of a 2TB drive to this compared to a 3TB drive. The terabytes per port cost is where I'm interested and at the moment, for me, 3TB drives work out cheaper per terabyte per port than 2TB drives. i.e the density of 3TB drives, when viewing my server costs as a whole, are now more cost effective than 2TB disks.

 

b) We can't expand ports ad infinitum. Either because the case we're using doesn't have enough slots and investing in a new case adds significant cost - or because simply unraid only supports a max amount of drives. Which comes back to getting the most terabytes per port (or slot in unraid parlance) you can.

 

A few people have suggested we should just buy 3TB drives now as we can use them later. I'd really like to think that's likely but we have no indication that's a feasible thing beyond speculation / hope.

 

No current roadmap and so no idea of 3TB support. I can't purchase storage just now that I know will be unused on the hope that it might be supported someday. See the per terabyte costs above.

 

I'll reiterate for 3TB support the drives are here, the controllers are here support in software in unraid is what's holding things back. There is every chance we will never see 3TB support. There is nothing to suggest otherwise at this point.

 

And I agree with the previous poster that this is out of concern (and for my own part frustration) that unraid needs to stay competitive.

 

I've recently built a small server for extended offsite backups and dropped some 3TB disks in. No problems with the controller hardware, no problems getting it formatted and storing data - not using unraid. 4TB drives aren't far off, lets not be two disk generations behind.

 

[squirrellydw]

...the poll clearly shows people want 5.0 released more than they want 2.2TB+ support — by a wide margin.

 

Wowza!  I can't imagine a better reason to register than to respond to that statement.  What the poll clearly shows is that you do not know how to read polls.  The poll clearly shows 51.5% of a few hundred current unraid users who inhabit this forum and like to vote in polls think 3T support should not be a priority.  That leaves 48.5% expressing a desire for 3T support.  9 votes - that's your "wide margin"?

 

If these are the numbers generated by current users, imagine a poll of would-be users.  How would the numbers shift if you put this question to a few hundred prospective buyers?  Yes, that's right, it would be a landslide for 3T support, because no one is looking to build a new massive storage system based on software that doesn't support the largest readily available hard drives.

 

As a happy user of unraid for several months now, I am grateful for the existence of this software more than I can adequately express.  I bought two licenses because I'm far more afraid of Tom losing interest in unraid than I am afraid of a flash drive failure.  As a current user with a 2T based system with plenty of room, the best thing for me is a stable 5.0 with no 3T support since I won't need it for quite some time.  But the best thing for Lime Tech is to snag as many fresh fish as it can by supporting the drives the fish want to use.  That's my vote, and if you care at all about unraid succeeding as an enterprise, that should be your vote too. 

 

The poll clearly show people want 5.0 more than 3tb support. The poll is slanted to try and have it support 3tb. Three of the four question support 3tb so of course is going to seem that half the people want it. Why not just ask, do you want 3tb support now or 5.0 now?

 

With that being said, I have plenty of room left in my 20 bay server so I want 5.0 stable.

[NAS]

I had to read the poll twice to come to that same conclusion.

 

I think even with the slant it is clear that most people want v5 first.

 

I am also amazed how many people have voted

[blockhead]

Three of the four question support 3tb so of course is going to seem that half the people want it.

 

So your hypothesis is that a significant portion of respondents are being tricked into saying they want something they don't really want?  Perhaps they are deciding how to vote by spinning a dreidel.  Or, my hypothesis, perhaps they can read and are picking the exact option they think is best.

 

Why not just ask, do you want 3tb support now or 5.0 now?

 

Good idea. And if you put that question to a group of a few hundred would-be users, what would the answer be?

 

Why not just ask whether the continued development of this software is beneficial to its current users?  Because if the answer is yes, then you'll want what put's the most motivation in Tom's pocket.

 

I have plenty of room left in my 20 bay server so I want 5.0 stable.

 

I have plenty of room in my 12 bay (pathetic I know!) server, and I want what's most likely to keep unraid in the forefront of the developer's mind, because in the long run that's what's best for everyone involved here.

[SnowLprd]

While I'm still seeing some arguments with questionable foundations, some good points were also raised.

 

Blockhead: You are right about the poll totals. That said, 52% of folks would rather see 5.0 shipped soon with its current feature set, which still speaks volumes. Your assumption that prospective unRAID buyers would constitute a "landslide" for the other options however, doesn't appear to be grounded in any material fact. It's a shaky assumption at best. And I don't buy the premise that the LimeTech business is going to suffer if 2.2TB+ support isn't added yesterday. The people who choose unRAID seem to do so because they understand its advantages over competing solutions, and I think it's highly unlikely that the LimeTech business will suffer to any significant degree if 2.2TB+ support appears in 5.1 instead of 5.0. I would argue that if you care about unRAID succeeding as an enterprise, vote for shipping early and often, when means getting 5.0 out the door so work can begin on 5.1.

 

Boof: As I mentioned earlier, I question whether the majority of folks have so much data to store that they need to worry about running out of ports. But maybe I'm wrong... Maybe the vast majority of folks here max out 20+ ports and need all the storage density they can get. I'm not one of those people (clearly), and for folks in my situation, the cost per terabyte is much more significant than the other measurement you describe.

 

Last but not least, let's cut the fear mongering: "There is every chance we will never see 3TB support." Yeah, and there's "every" chance that all of our unRAID machines are actually zombie botnet sleeper nodes that are waiting to be activated. The question isn't about what's possible — it's about what's probable, and given the track record (e.g., support for AF drives in the most recent 4.7 release), I'd argue there are much better odds that 2.2TB+ support will be added in the not-too-distant future than the odds for "never."

[boof]

Boof: As I mentioned earlier, I question whether the majority of folks have so much data to store that they need to worry about running out of ports. But maybe I'm wrong... Maybe the vast majority of folks here max out 20+ ports and need all the storage density they can get. I'm not one of those people (clearly), and for folks in my situation, the cost per terabyte is much more significant than the other measurement you describe.

 

It's up to an individual but even if you're not close to your case capacity I would still suggest the terabyte per port cost stands. You want value for money when you buy your storage, there's no point spending more than you need to. In the same way I'm sure most people wouldn't be going out to buy a 500G disk if they needed more storage as there would be little point when there are 2TB disks on the shelve. There would be a case for a different argument if unraid striped data across the disks and so more spindles meant more iops but I can't get past density being the number one thing.

 

Last but not least, let's cut the fear mongering: "There is every chance we will never see 3TB support." Yeah, and there's "every" chance that all of our unRAID machines are actually zombie botnet sleeper nodes that are waiting to be activated. The question isn't about what's possible — it's about what's probable, and given the track record (e.g., support for AF drives in the most recent 4.7 release), I'd argue there are much better odds that 2.2TB+ support will be added in the not-too-distant future than the odds for "never."

 

In that one paragraph you've suggested we will be getting 3TB support (I'm glossing over 2.5TB drives a little bit!) and assigned it a timeframe - something which others have also been arbitrarily doing. I'd hope to agree with you on your first point though on the latter, given release schedules I wouldn't like to suggest when. Either way it's pure speculation on your part. No fear mongering just fact based on current information.

 

My point is many people on this thread are clearly working on the assumption that 3TB support is happening and soon. I can't see where they're getting that from. Hopefully an updated roadmap will appear at some point.

[mrklaw]

right now I don't see 3TB as critical unless you have a very specific setup - i.e a low number of physical slots and a large amount of disposable income. 3TB drives simply aren't yet at the correct price/GB to be an economical choice for a setup. It'd be cheaper to buy a larger machine and put more 2TB drives in it.

 

But the point where 3TB drives become better value will come soon enough, and so we just need to hope/expect that 3TB support in unRAID comes on or before that point. Ideally there would be a clear statement from the dev team when there is a relatively commitable timescale, so that those looking into unRAID and requiring 3TB support will be reassured.

[SnowLprd]

It's up to an individual but even if you're not close to your case capacity I would still suggest the terabyte per port cost stands.

 

... and yet you don't explain why, which to me sounds like hand-waving. I'm not close to my case capacity, and I just bought a 2TB drive. Even if 2.2TB+ support were available, I would still have bought the 2TB drive. Why? Because given my storage needs, it demonstrably costs me less than the 3TB drive.

 

You want value for money when you buy your storage, there's no point spending more than you need to. In the same way I'm sure most people wouldn't be going out to buy a 500G disk if they needed more storage as there would be little point when there are 2TB disks on the shelve.

 

Huh? Your first sentence seems to prove my point, but then you take it into a different direction in the second sentence. If I only need 500GB today, and the price for that drive were significantly less per GB than the 2TB drive, why wouldn't I buy the 500GB? After all, by the time I need a 2TB drive, the price will have dropped significantly and it will probably end up costing less to buy that one plus the original 500GB than if I had bought the 2TB drive from day one. And I end up with an extra 500GB drive that I can stick into a drive dock if I want it for non-essential spillover data/backup.

 

Last but not least, let's cut the fear mongering: "There is every chance we will never see 3TB support." Yeah, and there's "every" chance that all of our unRAID machines are actually zombie botnet sleeper nodes that are waiting to be activated. The question isn't about what's possible — it's about what's probable, and given the track record (e.g., support for AF drives in the most recent 4.7 release), I'd argue there are much better odds that 2.2TB+ support will be added in the not-too-distant future than the odds for "never."

 

In that one paragraph you've suggested we will be getting 3TB support (I'm glossing over 2.5TB drives a little bit!) and assigned it a timeframe - something which others have also been arbitrarily doing. I'd hope to agree with you on your first point though on the latter, given release schedules I wouldn't like to suggest when. Either way it's pure speculation on your part. No fear mongering just fact based on current information.

 

1. Read my comment again. I was only talking about odds, and I stand by my suggestion that the odds for "never" are lower than "someday."

2. I imagine that 2.2TB+ support isn't too far off, but that is of course speculation.

3. Saying "There is every chance we will never see 3TB support" is not fact. It's fear mongering. Saying "There is a chance..." would be factually correct, but then again so is my hypothesis that all of our unRAID machines are actually zombie botnet sleeper nodes. 

[boof]

It's up to an individual but even if you're not close to your case capacity I would still suggest the terabyte per port cost stands.

 

... and yet you don't explain why, which to me sounds like hand-waving. I'm not close to my case capacity, and I just bought a 2TB drive. Even if 2.2TB+ support were available, I would still have bought the 2TB drive. Why? Because given my storage needs, it demonstrably costs me less than the 3TB drive.

 

Not hand waving. We're clearly accounting for our storage costs in a different way.

 

In your scenario above I would have bought whatever drive got me the best price for raw storage based on the total cost of my server. This will depend on individual servers and local part costs but I can tell you that 2TB drives are not cost effective versus 3TB drives for me (OS support being equal of course).

 

You want value for money when you buy your storage, there's no point spending more than you need to. In the same way I'm sure most people wouldn't be going out to buy a 500G disk if they needed more storage as there would be little point when there are 2TB disks on the shelve.

 

Huh? Your first sentence seems to prove my point, but then you take it into a different direction in the second sentence. If I only need 500GB today, and the price for that drive were significantly less per GB than the 2TB drive, why wouldn't I buy the 500GB? After all, by the time I need a 2TB drive, the price will have dropped significantly and it will probably end up costing less to buy that one plus the original 500GB than if I had bought the 2TB drive from day one. And I end up with an extra 500GB drive that I can stick into a drive dock if I want it for non-essential spillover data/backup.

 

Again you're accounting differently. You've described above buying based on the isolated drive price with a view to making the overall cost effective by taking some sort of action in the future based on unknown future variables.

 

You make a valid point and it's fair enough, but personally I'd rather buy storage at the most cost effective price point from the outset rather than hoping to balance the books later through further purchases and actual physical device shuffling. My glib example is probably too extreme but I'll stick with it, if I buy a 500G drive instead of a 2TB drive than the storage per port cost is terrible. That's where my value for money is as my storage ports are my most scarce commidity, not in the point of sale cost of the drive. This does presume that you'll end up using the entire drives worth of storage however.

 

1. Read my comment again. I was only talking about odds, and I stand by my suggestion that the odds for "never" are lower than "someday."

2. I imagine that 2.2TB+ support isn't too far off, but that is of course speculation.

3. Saying "There is every chance we will never see 3TB support" is not fact. It's fear mongering. Saying "There is a chance..." would be factually correct, but then again so is my hypothesis that all of our unRAID machines are actually zombie botnet sleeper nodes. 

 

I didn't say it was a fact. I said chance.

 

The facts are there is no information to say it is happening. No roadmap no nothing. I hope you're correct and I hope we get 3TB support. But it's hope and speculation. There is also chance we won't - if you want to call that fear mongering then I can't argue with that though I'd prefer to call it a more balanced viewpoint.

 

It's been bandied about this thread like it's a given we're getting 3TB support soon and what we're really discussing is priority. I'm pointing out that's getting slightly ahead of ourselves - unless I'm missing some information (which is always likely!)

[Rajahal]

Why not just ask, do you want 3tb support now or 5.0 now?

 

When I originally created the poll, that was exactly what I asked - just those two options.  Then bjp999 suggested some other poll options, so I invited him to edit the poll, which he did.

 

Also, when I created this poll I thought that one of those two outcomes would have happened by now.  Just speculation, but that was my hunch.  I thought that the poll would help Tom determine which hurdle he should tackle first.  Clearly I was wrong.  Tom instead decided to work on a completely separate issue, and we have had no indication of progress towards either a stable 5.0 release or 3 TB drive support.  I'm not saying that Tom hasn't been working on these, I'm just saying that he hasn't said anything to anyone of us about it (or updated the roadmap).

 

Personally, I don't feel this thread needs to be locked.  I'm actually tempted to create another poll re: whether or not this thread should be locked   I see that a lot of arguments are going around in circles, but I still believe that there's interesting discussion happening here as well.  One of the other mods may lock the thread if they see fit, but I won't.

[blockhead]

...I think it's highly unlikely that the LimeTech business will suffer to any significant degree if 2.2TB+ support appears in 5.1 instead of 5.0.

 

And I think it's highly impossible for LimeTech to benefit to any degree from the hundreds or thousands of people out there right now looking to build servers with 3TB disks.  That's all I'm saying.  Until support is added, none of that money goes to LT, and that's bad for us.

[BRiT]

I thought that the poll would help Tom determine which hurdle he should tackle first.  Clearly I was wrong.  Tom instead decided to work on a completely separate issue, and we have had no indication of progress towards either a stable 5.0 release or 3 TB drive support. 

 

Perhaps LimeTech providing full support for SAS/SCSI cards, such as the LSI SAS 9211-8i, is the most direct hardware path to reliably supporting 3TB drives. 

 

[kizer]

Unless there is a flame war I always believe in leaving a thread open and as long as its on topic.

[aiden]

Some people are more thick skinned than others, while some can take offense easily and hold a grudge.  I just prefer to not see us cannibalize each other out of general frustration, so I made the suggestion to lock.  Obviously that too is not in step with the majority, so I will continue to sit back and watch the page count grow.

[Rajahal]

I thought that the poll would help Tom determine which hurdle he should tackle first.  Clearly I was wrong.  Tom instead decided to work on a completely separate issue, and we have had no indication of progress towards either a stable 5.0 release or 3 TB drive support. 

 

Perhaps LimeTech providing full support for SAS/SCSI cards, such as the LSI SAS 9211-8i, is the most direct hardware path to reliably supporting 3TB drives. 

 

 

Interesting thought...I hope you are right.

[PeterB]

Perhaps LimeTech providing full support for SAS/SCSI cards, such as the LSI SAS 9211-8i, is the most direct hardware path to reliably supporting 3TB drives. 

 

 

Interesting thought...I hope you are right.

 

Well, as far as I can deduce, the SAS2008-based controllers are GPT/3TB compatible.  It is quite possible that many older mobo SATA ports will never become 3TB compatible.