Proliant buying advice
-
Read this on HP.com: "Because of the limitations of endurance, SSDs—unlike disk drives—have a limited service life in a
server."That service life is listed at just three years. Three years! I could make an assumption that the cost to replace in three years time will be considerably lower than the current cost, but I'm not sure I want to go there.
-
@Carnival-Boy said:
Had another look at SSDs. I didn't realise HP's "Enterprise Value" line was so cheap. $800 for 300GB.
When it comes to IOPS, SSDs are the value option. It's capacity where they are still lacking.
-
@Carnival-Boy said:
Read this on HP.com: "Because of the limitations of endurance, SSDs—unlike disk drives—have a limited service life in a
server."That service life is listed at just three years. Three years! I could make an assumption that the cost to replace in three years time will be considerably lower than the current cost, but I'm not sure I want to go there.
Normally enterprise SSDs have drastically longer lives than HDs. SSDs are measured in writes, HDs in service years, though. What a service life of three years on an SSD means I have no idea. They don't wear out that way.
-
@scottalanmiller said:
What a service life of three years on an SSD means I have no idea. They don't wear out that way.
Specifically "3k – 5k NAND program/erase cycles" for the Value range and "100k – 200k" for the Performance range. I have no idea how to calculate my requirements though.
I don't know if this means I have to replace them after three years, or if I can just replace them as and when they fail under warranty. Are they covered by post-warranty Carepacks, or would 3 years be considered wear and tear.
-
You might be interested in reading this regarding SSD life:
http://h20565.www2.hp.com/hpsc/doc/public/display?docId=emr_na-c03312456Basically, HP software monitors the number of writes, and when the limit is reached it puts the drive into a state of predictive failure and you have to replace it.
-
@Carnival-Boy said:
You might be interested in reading this regarding SSD life:
http://h20565.www2.hp.com/hpsc/doc/public/display?docId=emr_na-c03312456Basically, HP software monitors the number of writes, and when the limit is reached it puts the drive into a state of predictive failure and you have to replace it.
That's good. As Scott said, SSDs wear out based on the number of writes to a (don't know correct word) sector. So HP monitors those writes and when you reach a certain point they recommend replacement before failure, unlike an HD with often fail with no warning.
SSDs do something called wear leveling. They know which sectors have been written to and which ones haven't. So when writing new data instead of deleting the old data it marks the old data location as unused and writes the new data to a new lesser written to part of the drive.
I read somewhere that the average consumer SSD can take something like 100 GB of writes every day for 5+ years before you'll wear it out.
*edit - previously I stated 2 TB+ this is incorrect.
-
@Dashrender said:
I read somewhere that the average consumer SSD can take something like 2+ TB of writes every day for 5+ years before you'll wear it out.
If that's the case, it sounds like I have nothing to worry about.
-
@scottalanmiller said:
Normally enterprise SSDs have drastically longer lives than HDs.
HP contradict you:
"SSDs have a shorter lifespan, or endurance, than enterprise-level disk drives" -
@Carnival-Boy said:
HP contradict you:
"SSDs have a shorter lifespan, or endurance, than enterprise-level disk drives"They contradict the entire industry.
-
@Carnival-Boy said:
@Dashrender said:
I read somewhere that the average consumer SSD can take something like 2+ TB of writes every day for 5+ years before you'll wear it out.
If that's the case, it sounds like I have nothing to worry about.
That's consumer SSDs, of course. Enterprise is often listed in decades.
-
@Carnival-Boy said:
You might be interested in reading this regarding SSD life:
http://h20565.www2.hp.com/hpsc/doc/public/display?docId=emr_na-c03312456Basically, HP software monitors the number of writes, and when the limit is reached it puts the drive into a state of predictive failure and you have to replace it.
That's a separate issue. That's not the SSD wearing out, that's HP replacing drives proactively. As long as you are under warranty, that's a free, new drive.
-
@scottalanmiller said:
That's a separate issue. That's not the SSD wearing out, that's HP replacing drives proactively. As long as you are under warranty, that's a free, new drive.
I don't think so if this thread is anything to go by:
http://community.spiceworks.com/topic/465098-dell-enterprise-ssd-drives
It's Dell, but I'd guess HP is the same.Incidently, on that thread @Dashrender says he read a drive can write 100GB per day, but on this thread that's inflated to 2TB. I'll bear that in mind if he ever claims to have caught a hundred pound fish
-
Well, this article http://arstechnica.com/gadgets/2014/06/consumer-grade-ssds-actually-last-a-hell-of-a-long-time/ indicated they got about 700 TB before drives started to fail in their torture test. My statement of 2 TB is clearly mistaken.
FYI 3 of the drives they were testing where still working, and had written over 1 PB as of the article.
Again these are consumer multi cell drives, not single cell enterprise drives (though the value drives you mentioned might be multi cell drives too).
With your 20 GB DB you'd have to write the entire DB 5 times a day to come close to this type of torture, seems pretty unlikely for an SMB with a DB of that size.
-
@Carnival-Boy said:
I don't think so if this thread is anything to go by:
http://community.spiceworks.com/topic/465098-dell-enterprise-ssd-drives
It's Dell, but I'd guess HP is the same.That thread is mostly talking about MLC, not SLC enterprise drives, right? And still the wear is very long.
-
@Carnival-Boy said:
Incidently, on that thread @Dashrender says he read a drive can write 100GB per day, but on this thread that's inflated to 2TB. I'll bear that in mind if he ever claims to have caught a hundred pound fish
Thanks - Clearly I haven't read the article in a long time.. 100 GB does seem to coincide with the other article I just posted.
I fixed my other post.
-
@Dashrender said:
With your 20 GB DB you'd have to write the entire DB 100 times a day to come close to this type of torture, seems pretty unlikely.
And the more write cache you have on the controller, the less it writes to the disk. And this is why RAID 1 is important with SSDs rather than RAID 5 or RAID 6 which write two or three times for each block instead of just once like RAID 1. Write expansion isn't a concern with longevity on spindle drives but it is on SSDs so the write cache is less for speed and more for longevity in that case.
Having lots of memory reduces writes too.
-
So if "Enterprise Value" drives have massive IOPS, massive reliability and massive life expectancy, who buys "Enterprise Mainstream" and "Enterprise Performance" drives, given the massive premiums you're paying for the latter?
Is it possibly a bit of a money-making scam from HP to convince companies to overspend, or am I missing something else?
-
@Carnival-Boy said:
So if "Enterprise Value" drives have massive IOPS, massive reliability and massive life expectancy, who buys "Enterprise Mainstream" and "Enterprise Performance" drives, given the massive premiums you're paying for the latter?
We have no idea what those are.
-
@Carnival-Boy said:
Is it possibly a bit of a money-making scam from HP to convince companies to overspend, or am I missing something else?
Very possible.
-
@scottalanmiller said:
Each port is another full SAS channel with full SAS bandwidth. You want that if you have a place to use it. You don't, so does not matter to you. It is often used when you have an internal array and an external array or a massive internal array that you want to split.
From the specs, the 440ar has 2 internal 4 mini-SAS connectors, whilst the 440 has a single internal 8 mini-SAS connector. So both have 8 connectors, but the ar is split into 2 x 4 for whatever reason (form factor?).
