HPE Storage Users Group

Hey.

Brand new 3par system, installed yestoday. Migrated some VM's to the system today. AO runs from 07:00 to 17:00 and starts optimization at 18:00, max runtime 6 hours.

First run today, took 1 second. It have been set for performance mode. VM's was migrated to the sytem until 13:00. Any idea why it would only run for 1 second, and complete sucessfully - however nothing was moved to ssd. Are the VM's on the array too fresh for AO to measure anything yet?

I have two tiers, FC10K and SSD
The output is here:

2014-05-21 18:00:00 CEST Created task.
2014-05-21 18:00:00 CEST Updated Executing "AO_MAN-FRE_0700-1700_FC-SSD" as 1:24554
2014-05-21 18:00:01 CEST Updated Start sample: 2014-05-21 07:00:00 (1400648400)
2014-05-21 18:00:01 CEST Updated End sample: 2014-05-21 17:00:00 (1400684400)
2014-05-21 18:00:01 CEST Updated CPG SSD_r5 growth limited to 2052096.0 MiB
2014-05-21 18:00:01 CEST Updated CPG FC_r5 growth limited to 9854976.0 MiB
2014-05-21 18:00:01 CEST Updated maxRegions = 8192
2014-05-21 18:00:01 CEST Updated Analyzing 0 LDs in tier 0 CPG SSD_r5.
2014-05-21 18:00:01 CEST Updated Analyzing 8 LDs in tier 1 CPG FC_r5.
2014-05-21 18:00:01 CEST Updated LD Stats for CPG SSD_r5
2014-05-21 18:00:01 CEST Updated Time R_IOP/s W_IOP/s T_IOP/s R_KB/s W_KB/s T_KB/s R_svctms W_svctms T_svctms R_szKB W_szKB T_szKB qlen busy%
2014-05-21 18:00:01 CEST Updated LD Stats for CPG FC_r5
2014-05-21 18:00:01 CEST Updated Time R_IOP/s W_IOP/s T_IOP/s R_KB/s W_KB/s T_KB/s R_svctms W_svctms T_svctms R_szKB W_szKB T_szKB qlen busy%
2014-05-21 18:00:01 CEST Updated 2014-05-21 11:00:00 0.1 27.5 27.6 1.1 10476.9 10478.0 2.8 14.4 14.4 12.4 380.6 379.5 0 0.7
2014-05-21 18:00:01 CEST Updated 2014-05-21 12:00:00 2.6 110.4 113.0 112.4 40677.4 40789.7 2.1 16.3 16.0 43.1 368.5 361.0 0 2.6
2014-05-21 18:00:01 CEST Updated 2014-05-21 13:00:00 9.0 254.7 263.7 335.0 86993.1 87328.1 4.6 18.5 18.1 37.3 341.5 331.2 0 7.2
2014-05-21 18:00:01 CEST Updated 2014-05-21 14:00:00 984.7 170.8 1155.5 104210.5 54620.3 158830.8 5.2 27.7 8.5 105.8 319.7 137.5 0 10.5
2014-05-21 18:00:01 CEST Updated 2014-05-21 15:00:00 5.9 18.6 24.6 253.0 212.2 465.2 4.6 19.6 16.0 42.6 11.4 18.9 9 1.4
2014-05-21 18:00:01 CEST Updated 2014-05-21 16:00:00 2.4 13.3 15.8 124.3 138.7 263.0 4.5 19.0 16.8 51.0 10.4 16.7 0 1.0
2014-05-21 18:00:01 CEST Updated 2014-05-21 17:00:00 1.1 9.6 10.8 55.0 117.8 172.8 4.9 18.5 17.1 49.4 12.2 16.1 0 0.7
2014-05-21 18:00:01 CEST Updated Tier CPG Disk RAID UsedMB WarnMB LimitMB AvailMB IO/GB*m_Max IO/GB*m_Avg IO/GB*m_Min Count T_IOP/s T_svctms
2014-05-21 18:00:01 CEST Updated 0 SSD_r5 SSD 5 0 0 0 1048576 0.000 0.000 0.000 0 0 0.0
2014-05-21 18:00:01 CEST Updated 1 FC_r5 FC 5 835584 0 0 1048576 0.000 0.000 0.000 0 230 11.0
2014-05-21 18:00:01 CEST Updated Starting genMoves iteration: 0
2014-05-21 18:00:01 CEST Updated Starting selMoves 1 0 busiest Performance
2014-05-21 18:00:01 CEST Updated SSD_r5 svctms: 0.0 maxSvctms: 15 raidSvctFactor: 1.5
2014-05-21 18:00:01 CEST Updated FC_r5 svctms: 11.0 maxSvctms: 40 raidSvctFactor: 1.5
2014-05-21 18:00:01 CEST Updated accrLimit: 0.000
2014-05-21 18:00:01 CEST Updated Criteria: busiest Move from tier1 to tier0 0.0 MiB
2014-05-21 18:00:01 CEST Updated Starting selMoves 0 1 idlest Performance
2014-05-21 18:00:01 CEST Updated FC_r5 svctms: 11.0 maxSvctms: 40 raidSvctFactor: 1.5
2014-05-21 18:00:01 CEST Updated Criteria: idlest Move from tier0 to tier1 0.0 MiB
2014-05-21 18:00:01 CEST Updated Checking for compact with parameter: auto
2014-05-21 18:00:01 CEST Updated No need to compact CPG SSD_r5: Both sdfree_mb (0.0) and usrfree_mb (0.0) <= freespacemin_mb (16384)
2014-05-21 18:00:01 CEST Updated No need to compact CPG SSD_r5: Both sdfree_mb (0.0) and usrfree_mb (0.0) <= freespacemin_mb (16384)
2014-05-21 18:00:01 CEST Updated No need to compact CPG FC_r5: Both sdfree_mb (9216.0) and usrfree_mb (0.0) <= freespacemin_mb (65536)
2014-05-21 18:00:01 CEST Updated No need to compact CPG FC_r5: Both sdfree_mb (9216.0) and usrfree_mb (0.0) <= freespacemin_mb (65536)
2014-05-21 18:00:01 CEST Updated AO complete
2014-05-21 18:00:01 CEST Completed scheduled task.

I think the first day the array is running it still has to figure things out internally. I'd wait until tomorrow and see what it processes.

Did AO ever run correctly?

_________________
Bryan W
Senior Architect/Manager of System Infrastructure, Dallas TX
https://www.linkedin.com/in/bryanlwhite

Having done the same myself, my experience is that it will take 1-2 days for it to start moving.

I still havnt managed to get much data up in the SSD tier, but my latencies and IOPs on my tiers are placed correctly 99% of the IO happends on the SSD tier, even though it only has around 10% of the data.

I am also using the performance profile.

Hi guys. Sorry i did'nt reply. I only got one mail regarding new post.

To answer the question, 1 day later ao ran correctly and moved 210 gb. It have been running a few days more, and 300 gb lies in the ssd tier now. I was hoping for more, because i dont want to waste all the good ssd io

I've been migrating 10 servers more over the weekend, so i guess more will be moved on monday.

Which report in report manager would you create, to see the io distribution - ssd vs. Fc?

Yes - see my new post

The SR Report you are looking for is probably Adaptive Optimization > VV Tiers

_________________
Bryan W
Senior Architect/Manager of System Infrastructure, Dallas TX
https://www.linkedin.com/in/bryanlwhite