Solved Email server options
-
@Pete-S said in Email server options:
Always go with 3.5" storage when you need some volume but not SSD speed.
Ultrastar 12TB 7.2K SAS-3 drives are about $400 each. 12TB RAID-1 becomes about $800 for 12TB storage. That's 6.7 cents per GB of data.How long will it take for a raid array to rebuild on a 12Tb disk?
-
@dyasny said in Email server options:
@Pete-S said in Email server options:
Always go with 3.5" storage when you need some volume but not SSD speed.
Ultrastar 12TB 7.2K SAS-3 drives are about $400 each. 12TB RAID-1 becomes about $800 for 12TB storage. That's 6.7 cents per GB of data.How long will it take for a raid array to rebuild on a 12Tb disk?
The time it takes to copy one full disk to the new drive.
-
@dyasny said in Email server options:
@Pete-S said in Email server options:
Always go with 3.5" storage when you need some volume but not SSD speed.
Ultrastar 12TB 7.2K SAS-3 drives are about $400 each. 12TB RAID-1 becomes about $800 for 12TB storage. That's 6.7 cents per GB of data.How long will it take for a raid array to rebuild on a 12Tb disk?
If RAID 1, it's a strait copy. If other RAID levels, the answer is really complex.
-
@scottalanmiller said in Email server options:
@dyasny said in Email server options:
@Pete-S said in Email server options:
Always go with 3.5" storage when you need some volume but not SSD speed.
Ultrastar 12TB 7.2K SAS-3 drives are about $400 each. 12TB RAID-1 becomes about $800 for 12TB storage. That's 6.7 cents per GB of data.How long will it take for a raid array to rebuild on a 12Tb disk?
If RAID 1, it's a strait copy. If other RAID levels, the answer is really complex.
Exactly, in other words, you didn't provide enough information to answer your question.
It will also depend if the system is in use while the rebuild is happening - and how much that activity is hitting the disk.
-
@travisdh1 @scottalanmiller my point here is, huge drives sounds great on paper, in terms of $ per Gb, but whenever possible, I will always take a lot of smaller spindles over a few huge ones. When dealing with spindles that is, SSDs and NVMes are a whole different story of course.
Imagine you're building a large data store with huge disks, because it feels like you're getting more for less that way. And assuming your disk in a RAID5 takes X hours to rebuild. During that X, you're as vulnerable as if you were running raid0, more vulnerable, because you have multiple disks from the same production series, with the same age and wear on them, so chances are high more will die simultaneously. The larger the disks, the higher the X, and 12Tb will have you counting X in days, not hours, at least in a parity based RAID.
You can always go for other RAID levels, with higher redundancy rates, but that also has downsides, both in price and performance. In short, YMMV, but I always advise to take factors beside the price per Gb into consideration, it's a huge factor people tend to skip entirely.
-
@dyasny Are we on the why RAID 5 is bad with massive capacity storage disks again?
-
The simple answer is, a risk assessment should be performed, but generally speaking spindles are a never use RAID5 case. RAID6 or 10.
SSD RAID5 is okay.
-
@dyasny said in Email server options:
es are high more will die simultaneously. The larger the disks, the higher the X, and 12Tb will have you counting X in days, not
yup - the math shows that at 12 TB, you have a near 100% chance of a second drive failing during a RAID 5 resilver.
RAID 6 or 10, and depending on the size, might even have to go to RAID 7 instead of 10.
This makes looking at SSDs and RAID 5 very attractive.
-
@Dashrender said in Email server options:
@dyasny said in Email server options:
es are high more will die simultaneously. The larger the disks, the higher the X, and 12Tb will have you counting X in days, not
yup - the math shows that at 12 TB, you have a near 100% chance of a second drive failing during a RAID 5 resilver.
Not at all. Not even close. You are mixing URE risks with disk failure rates. Unrelated items.
-
@scottalanmiller said in Email server options:
@Dashrender said in Email server options:
@dyasny said in Email server options:
es are high more will die simultaneously. The larger the disks, the higher the X, and 12Tb will have you counting X in days, not
yup - the math shows that at 12 TB, you have a near 100% chance of a second drive failing during a RAID 5 resilver.
Not at all. Not even close. You are mixing URE risks with disk failure rates. Unrelated items.
Whoops.. you're right, I typed the wrong thing...
-
@dyasny said in Email server options:
Imagine you're building a large data store with huge disks, because it feels like you're getting more for less that way. And assuming your disk in a RAID5 takes X hours to rebuild. During that X, you're as vulnerable as if you were running raid0, more vulnerable, because you have multiple disks from the same production series, with the same age and wear on them, so chances are high more will die simultaneously. The larger the disks, the higher the X, and 12Tb will have you counting X in days, not hours, at least in a parity based RAID.
Absolutely, although you have to consider the total number of spindles as well. Each additional spindle carries a risk factor, too.
-
@scottalanmiller said in Email server options:
Absolutely, although you have to consider the total number of spindles as well. Each additional spindle carries a risk factor, too.
same idea as with distributing a load between a lot of small hosts or running one big monolith.
-
@dyasny said in Email server options:
@scottalanmiller said in Email server options:
Absolutely, although you have to consider the total number of spindles as well. Each additional spindle carries a risk factor, too.
same idea as with distributing a load between a lot of small hosts or running one big monolith.
Sort of, but that's not quite the same. It's distributed in both cases, it is redundant in both cases. There are lots and lots of factors involved, not just "breaking it up into nodes." It's more complex than that. At some point, more smaller spindles is safer, but at some point fewer, larger ones are. And you have to consider a lot of factors including drive fail rates, UREs, time to rebuild, time to replace, etc. It's a large equation.
For example, if your drives move 100 IOPS, then many small drives is likely to make sense. But if your drives move 10,000,000 IOPS, then two giant drives will likely make more sense (assuming equal failure risks.) Speed and failure rates are key here, if you don't consider then, you can't tell when more drives or fewer drives is safer.
-
@scottalanmiller said in Email server options:
Sort of, but that's not quite the same. It's distributed in both cases, it is redundant in both cases. There are lots and lots of factors involved, not just "breaking it up into nodes." It's more complex than that. At some point, more smaller spindles is safer, but at some point fewer, larger ones are.
At what point fewer larger spindles are safer? With more drives you get more spindles, reducing the seek time, the main problem with magnetic drives. With more drives you can implement RAID with better redundancy levels - 10, 50, the EE variants etc. The only real downside is the fact that you are running more kit - you need more physical space, connectors, cables, power and more parts might fail and need replacements (without affecting the system).
And you have to consider a lot of factors including drive fail rates, UREs, time to rebuild, time to replace, etc. It's a large equation.
Most of these factors, when dealing with spindles and not SSDs/NVMes favour the more/smaller idea.
For example, if your drives move 100 IOPS, then many small drives is likely to make sense. But if your drives move 10,000,000 IOPS, then two giant drives will likely make more sense (assuming equal failure risks.) Speed and failure rates are key here, if you don't consider then, you can't tell when more drives or fewer drives is safer.
10000000 IOPS? Are we still talking about spindles here?
-
@scottalanmiller said in Email server options:
@Curtis said in Email server options:
@scottalanmiller said in Email server options:
@Curtis said in Email server options:
I'll keep all your account active for a week or so, if you want one time, please let me know.
I think you need to run an ML mail service for everyone
Thinking about it. Just sure I could do it for free.
I bet you'd only be looking at 20-30 people seriously interested. Would be pretty cool, though.
I can do it for free. But I need to set some restrictions.
TLDR; Free service, not for abuse or high volume.
For example:
- Server would be a VPS in Europe (for those who care) with good performance but high latency to US.
- Only 35 GB of storage total (maybe 70 if needed). I do not care about the amount of email accounts as long they are within the storage limits.
- No Solr searching (Java and memory usage).
- Only disaster recovery backups: 1 week.
- I will need some time to get a clean IP. Meanwhile I could use Mailchannels until I get one, but only if email volume is low.
- I will shut it down the moment I see the service is abused by anyone or if an account is hacked and starts sending SPAM.
- I will not be fighting incomming SPAM. Every user will be responsible to train rspamd for their own account.
- I do not want to be vetting who gets an email or admin account and who doesn't. Someone else would need to do that.
- The service will go down from time to time for updates. Like once a month, for less than 30 minutes, at night in America
- For sure I forgot more that one thing, but you get the idea.
By the way, which domain name? who will own it?
Let me know if interested. -
@dave_c let’s work on this together since I already have a start on it.
Your terms are reasonable, and similar to the terms I would have
-
Why would the VPS be in Europe & not the US?
-
@FATeknollogee cheaper
-
@Curtis said in Email server options:
@FATeknollogee cheaper duh
It can't be that much cheaper, I don't buy that reason.
So you'll choose the higher latency VPS because it's less money but all the folks that are using it are here in the US?
-
@FATeknollogee everyone is in the US? That can’t possibly be true...
