propergol
-
Posts
80 -
Joined
-
Last visited
-
Days Won
4
Posts posted by propergol
-
-
Thanks. I have found what is the small card : it is a "riser" card and it seems it needs to be plug onto specific server ports.
Also regarding the SAS expander : yes I do know that I need a contoller
that is the reason why I did wrote that I already have a m1115.I did have a closer look to the expander pictures and if you look at the Molex, it looks like the heatsink has beegining burning it
...I will wait for something less "used"...
-
and mine :
Both SSD (Sandisk Extreme II and Sandisk Extreme Pro) have same ID : RQKIA4F2
-
I am still wondering if I should buy an other IBM M1115 or buy an Intel SAS expander...
Expanders are not very cheap
...I manage to find one, not new RES2SV240, but it comes with an additional card and I don't know what it is.
Here is the auction : http://www.ebay.fr/itm/Intel-RES2SV240NC-RES2SV240-24-port-6Gbs-SATA-SAS-RAID-Expander-Card-/161823139680?hash=item25ad692760
$200 for both cards and cables, do you think it worth it? What is the second litle card exactly?
Thanks
EDIT : I need a new SAS card or expander because I plan to use SSD for caching, at least 3, and I miss some SATA 3 ports.
-
Let me know if I am wrong : if I disable "Bypass File System Filters" then my antivirus would scan x3 duplicates files/folders 3 time rather than only 1 time?
-
One question regarding Perfectdisk : Do you think that its Optiwrite function can work with pooled drives?
Basicaly it would prevent fragmentation before it happens, but I am unsure if it can works with DrivePool "bypass file system filters".
-
Thanks
So first I did added a new SSD and Scanner's SMART window is empty here too
The first SSD Sandisk Extreme II is plugged into Intel's mobo port and the second Sandisk Extreme Pro into M1115.
I have made a video that shows what Scanner, Direct IO Test are showing. I have also included a Hard Disk Sentinel view that seems to read SMART without issue.
-
The SSD is plugged into Intel's motherboard port and here is what scanner is showing :
On Intel's other ports I also have WD green and Scanner read SMART values from them correctly.
Same for REDs that are plugged into M1115 (RAID mode) SMART reading is ok.
-
Thanks a LOT!
DrivePool 2.2.0.649 beta works and it is fast now
I have no idea where the problem was but it is now fixed.
Now that read striping is fixed I would try to speed up the write thing with SSD plugin. I am waitting for some good deal (since I need 3) on fast 250Gb SSD.
Thanks again.
-
I don't know what happend but it realy looks like it read only from one disk.
I have made an other test : 2 movies on a pool with x3 duplication so both movies are present in 3 differents hard drives.
I did then launch a copy of one movie 1 on a drive A outside the pool and movie 2 on drive B outside the pool.
Please take a look at what happen then (the PC's specs at the bottom left are not correct : I did remote desktop to my server so this is not the server specs) :
-
ok thanks, I will go with TLER enabled.
-
Glad to hear you will look at it
What is very strange is that I did face this issue using 2 differents systems (Win 10 and Server 2012 r2), and 2 differents hardware controlers (Intel ICH10R, IBM M1115), even the motherboards are from different brand (Asrock, Gigabit), processor (Sandy, Ivy) etc...
My AV is also different : on Server 2012 R2, Symantec Endpoint Protection, on Win 10, Bitdefender Total Security 2016.
The only common thing is the chipset : Z77
-
Hello,
As you maybe know I am comming from home made NAS that was using Intel RAID 10 4x4Tb RED drives.
Since those drives are no longuer part of RAID array, I was wondering if I should disable TLER in order to let the drives to do their own error recovery.
Or should I let it enabled?
Thanks
-
same here : add Mega please.
-
Just to clarify a few things here:
Read Striping absolutely requires duplicated data. If it's not duplicated, you don't get the advantage here.
If you read again my first post, I clearly explain that :
"- the pool has real time duplication, read sriping, bypass system filters set to ON and file protection set to file duplication"
if you prefer, file duplication is set to "x2" for the whole pool.
Additionally, Read Striping doesn't always read from both copies of the files. There are situations where it will read from one disk or the other (such as different bus speeds, and usage).
Again, if your look at my first post, I explain that first I did start the copy of one file on the pool to a SSD (outside the pool) then I did started to copy an other file on the pool to a HD (outside the pool).
If you look at the copy screenshot you can see that even when copying 2 differents files from the pool at the same time to 2 differents devices, the copy only reads from one disk of the pool when it should read one file from one pooled disk and the other from the second pooled disk.
Since this read striping doesn't happend, copying those 2 files run at only 60mb/s (half the read speed of one drive of the pool) when it should be running at the read speed of pooled disk 1 + read speed of pooled disk 2.
One thign to try is to set the block mode to "false" and see if that affects the behavior at all.
I did changed this option then stoped/restarted DripePool service : no change, same behavior
Regardless of this issue, I did just bought Stablebit bundle
-
I am afraid the issue could afect other systems than Windows 10...
I did made the same copy test that I did under Windows 10 but this time with DrivePool installed on Windows Server 2012 R2 and I see the exact same thing : read stripping doesn't seems to work.
I guess you could easily reproduce the issue on your side.
-
Well backup of each client won't be bigger than 50 Gb since most of their datas are already on server's SMB shares. Also, second bigest file transfert would probably be movies and won't exceed this size.
But you are right regarding the SSD size it is very important not to underscale.
I was thinking to add 3 x CRUCIAL BX100 250Gb SSDs because they are not expessive (less than €85), they are MLC based SSD, their write speed is 370Mb/s and are extremly power efficient.
But again, nothing choosed yet.
-
Hmm, WS2012R2 allows for concurrent Client Backups? Nice.
Of course, even a slower write would not matter as backups would simply take a bit longer. If it were a technical requirement then someone running 20 clients would have a serious issue, no?
I used to be scared of SSD wear but it might be worth while to investigate the actual I/O Client Backups cause. Assume you would actually write 1TB a day for the Client Backups, which seems like a lot to me, even TLC should still last you years, no?
Also, I seriously doubt Clients would actually deliver 125MB/s. Thinking about it, I would say that Client Backups is probably one of the less obvious reasons to go SSD Optimiser.
You are right : client backups (Acronis backup, not Windows backup) wont run @ 125Mb/s since during file copy tests it did cap at 118Mb/s.
Also I wouldn't schedule client backup to run at the same time...but again IF this happend, or other scenario with concurent file transfert then I want it to run uncapped @118Mb/s for each client.
So...118Mb/s x3 = 354Mb/s
Also imagine that tomorow I will add 1 more 1Gb link or maybe a single 10Gb link for a new PC then theoretical write would be higher again.
I am simply wondering if sequential write is the only spec I should look at, or maybe there are other important things,I don't know since caching is something special.
-
About the SSDs for caching, what are the most important specs?
Let me know if this is correct, and or, if I am missing things :
- SSD based on MLC since caching may involve a lot of write/delete and MLC are less prone to wear than TLC
- in my case, the worst case is that 3 clients would upload their backups at the same time to the server running DrivePool. Since each client would use 1Gb ethernet link : 125Mb/s x 3 = 375Mb/s So each SSD should be capable of at least 375Mb/s sequential write.
-
Could you try downloading the attached file and placing it in "C:\Program Files\StableBit\DrivePool"?
(this is stock, except for the language override to english).
Thanks a lot, it works.
Just a note : it loks like "DrivePool.Service.exe.config " wasn't present in my computer, only DrivePool.Service.exe.default.config was here.
-
Read Striping specifically checks the BUS speed for the disks in question, IIRC. If one is much faster than the others, it will prefer disks on that over the other, instead of reading from both.
Well...in my case, disks are from same model and ports from same chipset/speed. Also both disks where tested read/write at the very exact same speed.
Could you ensure that Read Striping is enabled (Pool Options -> Performance -> Read Striping (it should be checked).
If it is, and you're still not seeing that,.... (and there should be a third bar in the Performance UI for Read Striping, actually)
Read striping was on as @Umfriend said "Lire la bande" = Read striping
then could you enable file system logging, copy a file and then stop the logging.
http://wiki.covecube.com/StableBit_DrivePool_2.x_Log_Collection
I did upload the logs in .rar archive. Just a note : I did started the log record, started copy one file from the pool to drive A, started copy of an other file from the pool to drive B, then stopped the copy, stoped log reccording.
Also, do you have StableBit Scanner installed on the system? If so, could you open the disk details on both disks, and look at what the SATA level it's indicating for each disk?
I didn't have Stablebit Scanner installed, I was using HD Sentinel
... so I did uninstalled HD Sentinel, installed Stablebit Scanner.Here is a screenshot :
-
AFAIK, read striping works on the file level only and by that I mean that for each single file you read, DP will choose a HDD to read it from. It will not read parts from HDD1 and other parts from HDD2.
I guess I did misread DrivePool 's doc since it was talking about block based copying, I think it was capable of reading from multiple disk at same time.
So to get benefit from read striping you might want to test copying a bunch of files. But then, I think, it becomes a question of how many copy operations windows will allow concurrently. As far as I can tell, the regular copy does only one at a time so you could try with robocopy.
That is exactly what I did tried to do in the second part of my copy test (check the screenshot) : I did launch a copy of file A to a SSD (outside the pool) and a copy of file B to a HD (outside the pool). You can see on the screenshot that even in this case the read is only done on one disk of the pool.
-
On version 2.2.0.636 BETA, it has no effect, UI is still in french.
-
Since I was running Windows 10 pro insider build 10532, and was thinking it could be the root of my issue, I did a fresh install of standard Windows 10 pro (not insider).
The result is the same, the copy only read from a single disk at the same time.
Then I think : maybe it can be that my Intel ports (where all HD, SSD are plugged) are set to RAID mode in BIOS, even if HD are not "raided".
So I did made some changes in order to Windows 10 to run in AHCI mode rather than RAID mode : same result
-
I would like to use DrivePool in english but my system is in french so I did try to change :
<setting name="DrivePool_CultureOverride" serializeAs="String">
to
<setting name="DrivePool_CultureOverride" serializeAs="en">
I did even restart my PC : no luck UI is still in french
Question regarding Intel SAS expander (ebay auction)
in Hardware
Posted
Yep. Sounds right. I will offer at $90.
Since I have a very cramped build : 9 HDs + 4 SSDs inside a litle mid tower, the expander powered with a Molex would offer better cable management than an other M1115.