Simpler Navigation for Servers and Operating Systems - Please Update Your Bookmarks
Completed: a much simpler Servers and Operating Systems section of the Community. We combined many of the older boards, so you won't have to click through so many levels to get at the information you need. Check the consolidated boards here as many sub-forums are now single boards.
If you have bookmarked forums or discussion boards in Servers and Operating Systems, we suggest you check and update them as needed.
BladeSystem Server Blades
cancel
Showing results for 
Search instead for 
Did you mean: 

SSD endurance and Block size details

chuckk281
Trusted Contributor

SSD endurance and Block size details

Robert had a customer question:

 

*****************

 

A customer is looking to deploy a large number of SSD drives (HP 800GB 6G SATA Value Endurance SFF 2.5-in SC - 717973-B21) in a Cassandra cluster. Each BL460c Gen8 server will have 1 SDD. 

 

In using the ‘value’ SSDs the customer understands the write endurance limitations of the drive and is hoping to put into place processes to avoid wear (i.e filling full blocks without the need for a write/erase).  We have a few questions that I cannot seem to locate readily available answers and I am hoping that you can help me out with:

 

  1. Does this SDD drive support Trim? - HP 800GB 6G SATA Value Endurance SFF 2.5-in SC Enterprise Value 3yr Wty Solid State Drive 717973-B21
  2. The Logical block size is 512bytes (SAS/SATA interface ).  What is the page and Physical Bock size for this SSD? 
  3. What SSD OEM is this drive based on?

************

 

Info from Robert:

 

***************

 

Smart Array controllers like P220i do not yet provide SCSI UNMAP support on logical drives (mapping to SCSI UNMAP or ATA TRIM on the physical drives used to create each logical drive).  So, any OS support for UNMAP/TRIM/discard on file deletions won’t make it to the drives.

Smart Array does send UNMAP/TRIM commands to drives for Rapid Parity Initialization, but that’s not used for RAID 0. It will also send those commands in an upcoming Over-Provisioning Optimization feature in Snap6 when creating the first logical drive.

 

I can check with the drive group for you on the specific page size (usually 4 KiB or 8 KiB) and drive vendor(s) for this SKU (we usually don’t disclose that). 

 

http://www.slideshare.net/rbranson/cassandra-and-solid-state-drives is a presentation by DataStax on Cassanda and SSDs, pointing out that all writes are sequential.  http://www.flashmemorysummit.com/English/Collaterals/Proceedings/2013/20130814_201B_Moxon.pdf slide 22 discusses some other Cassandra SSD considerations.

 

*************

 

Other comments or questions?