MSA Storage
1753628 Members
5526 Online
108797 Solutions
New Discussion

MSA x1600 G2 bad read performance

 
nmason-tech
New Member

MSA x1600 G2 bad read performance

Hi All,

 

 Hope someone may have some insight into an issue we're experiencing.

 

 HP x1600 G2 network storage system - LFF w/ p212 512mb bbwc model also an attached D2600

 

 Two arrays are configured one 12 disk R5 (2TB 7.2kRPM SATA) and one 6 disk R5 (2TB 7.2kRPM SATA).

 

 All was working well up until about a week or so ago when we started seeing issues with some of the iSCSI connected clients having slow file system reads (>30MB/s). We're not sure what has occurred but through testing have concluded

that while any level of write I/O is being performed and a simultaneous read occurs the read operation is extremely degraded. 

 

We have 4 systems connecting via MS iSCSI initiator to the WSS2008 iSCSI target system with a handful of VHD's for iSCSI storage. As a test we disconnected these clients so that the x1600 G2 had no outside I/O occurring and ran some read/write operations. 

 

 Using ATTO Benchmark with nothing else going on in the background on the two arrays they are able to achieve +350MB/s write & read speeds so long as it's one or the other. As a test we copied a 6GB to the 12 disk array and simultaneously attempted to read a 500MB file at the same time. The result is the write speed of around 200MB/s and a read speed of >30MB/s.  Same thing happens with the 6 disk array as well.

 

I've checked the array controller settings and they are as follows currently:

 

Internal Controller
Controller Status OK
Controller Smart Array P212
Slot 3

Bus Interface PCI
Hardware Revision C
Firmware Version 5.12
RAID 6 (ADG) Status Enabled
Rebuild Priority Medium
Transformation Priority Medium
Surface Scan Analysis Priority Idle
Surface Scan Delay 3 sec
RAID 6/60 Alternate Inconsistency Repair Policy Disabled
RAID 5/6/50/60 Degraded Mode Performance Optimization Enabled
Physical Drive Request Elevator Sort Disabled
Monitor and Performance Analysis Delay 60 min
Maximum Drive Request Queue Depth Automatic
Number of Ports 2 (1 Internal / 1 External)
Number of Arrays 3
Number of Logical Drives 3
Number of Physical Drives 20

Array Accelerator
Array Accelerator Present Yes
Cache Status OK
Array Accelerator (Cache) Ratio 50% Read / 50% Write
Total Cache Size 512 MB
Cache Backup Power Source Batteries
Battery/Capacitor Pack Count 1
Battery/Capacitor Status OK

 

Anyone have any ideas as to what could be causing simultaneous read/write to be so bad? 

 

Thanks,

 

NM-Tech

 

 

P.S. This thread has been moevd from General to Storage Area Networks (SAN) (Small and Medium Business). - Hp Forum Moderator