Oracle stress testing tool ORION

ORION (OracleIOCalibrationTool) is an independent tool used to calibrate the IO performance of the Oracle database storage system. Calibration Results for understanding storage system performance

ORION (Oracle I/O Calibration Tool) is an independent Tool used to calibrate the I/O performance of the Oracle database's storage system. Calibration Results for understanding storage system performance

1. Features

ORION (Oracle I/O Calibration Tool) is an independent Tool used to calibrate the I/O performance of the Oracle database's storage system. The calibration result is of great help for understanding the performance of the storage system. It can not only identify the problems affecting the performance of the Oracle database, but also measure the size of the new database installation. As ORION is an independent tool, you do not need to create and run Oracle databases.

To simulate the Oracle database as much as possible, ORION uses the same I/O software set as Oracle to generate a combined I/O load. You can configure ORION to generate many I/O loads, including an I/O load that simulates OLTP and data warehouse loads.

ORION supports multiple platforms. More platforms will be supported in the future. The User Guide provides detailed user guides, including "Getting started" and troubleshooting instructions. You can call the "-help" option in ORION to directly obtain the abstract about how to use ORION.

2. Simple use process

Run and install
Gunzip orion_linux_x86-64

Run the test
Generate the mytest. lun File
/Dev/mapper/mpat1_p1
/Dev/mapper/mpath8p1

Run commands
/Orion_linux_x86-64-run simple-testname mytest-num_disks 2

Other commands can be used for help.
/Orion_linux_x86-64-help

3. Test Results
[Root @ tmp] # cat mytest_20120629_1519_summary.txt
Orion version 11.1.0.7.0

Commandline:
-Run simple-testname mytest-num_disks 2

This maps to this test:
Test: mytest
Small IO size: 8 KB
Large IO size: 1024 KB
IO Types: Small Random IOs, Large Random IOs
Simulated Array Type: CONCAT
Write: 0%
Cache Size: Not Entered
Duration for each Data Point: 60 seconds
Small Columns:, 0
Large Columns:, 0, 1, 2, 3, 4
Total Data Points: 15

Name:/dev/mapper/mpath7p1 Size: 536869863424
Name:/dev/mapper/mpath8p1 Size: 536869863424
2 FILEs found.

Maximum Large MBPS = 183.53 @ Small = 0 and Large = 4
Maximum Small IOPS = 851 @ Small = 10 and Large = 0
Minimum Small Latency = 11.38 @ Small = 3 and Large = 0

4. Comparison and Analysis of results
During stress testing, you can use the IOSTAT command to calculate the value. The maximum value varies slightly due to sampling.

IOPS = [dm-13] r/s + [dm-14] r/s + [dm-13] w/s + [dm-14] w/s IOSTAT Max 422 + 424 = 846
MBPS = ([dm-13] rsec/s + [dm-14] rsec/s)/2/1024 = RMB/s + wMB/s IOSTAT Max 94 + 92 = 186 MBPS
Latency = await: About 11 seconds
[Root @ mapper] # iostat-d-x 1 30 | grep dm-1 [3, 4]
Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s avgrq-sz avgqu-sz await svctm % util
Dm-13 0.00 0.00 25.50 0.00 3386.05 0.00 132.77 0.32 12.57 4.45 11.35
Dm-14 0.00 0.00 25.56 0.00 3426.57 0.00 134.04 0.32 12.60 4.44 11.36
Dm-13 0.00 0.00 339.00 0.00 5424.00 0.00 16.00 4.01 11.86 2.95 100.10
Dm-14 0.00 0.00 328.00 0.00 5248.00 0.00 16.00 3.97 12.00 3.02 99.20
Dm-13 0.00 0.00 364.00 0.00 5824.00 0.00 16.00 4.20 11.43 2.74 99.70
Dm-14 0.00 0.00 325.00 0.00 5200.00 0.00 16.00 3.78 11.72 3.05 99.10
Dm-13 0.00 0.00 340.00 0.00 5440.00 0.00 16.00 3.94 11.64 2.94 99.80
Dm-14 0.00 0.00 355.00 0.00 5680.00 0.00 16.00 4.02 11.36 2.82 100.20

[Root @ host8] # iostat-d-xm 1 30 | grep dm-1 [3, 4]
Device: rrqm/s wrqm/s r/s w/s RMB/s wMB/s avgrq-sz avgqu-sz await svctm % util
Dm-13 0.00 0.00 341.00 0.00 2.66 0.00 16.00 4.07 11.90 2.93 99.80
Dm-14 0.00 0.00 343.00 0.00 2.68 0.00 16.00 3.90 11.35 2.92 100.10
Dm-13 0.00 0.00 330.00 0.00 2.58 0.00 16.00 3.92 11.89 3.01 99.40
Dm-14 0.00 0.00 360.00 0.00 2.81 0.00 16.00 4.04 11.29 2.73 98.40
Dm-13 0.00 0.00 342.00 0.00 2.67 0.00 16.00 4.09 12.08 2.90 99.30
Dm-14 0.00 0.00 352.00 0.00 2.75 0.00 16.00 3.87 10.89 2.82 99.40
Dm-13 0.00 0.00 345.54 0.00 2.70 0.00 16.00 3.98 11.48 2.87 99.21
Dm-14 0.00 0.00 336.63 0.00 2.63 0.00 16.00 3.92 11.61 2.92 98.42
Dm-13 0.00 0.00 352.00 0.00 2.75 0.00 16.00 3.96 11.23 2.83 99.50
Dm-14 0.00 0.00 344.00 0.00 2.69 0.00 16.00 3.98 11.62 2.89 99.30
Dm-13 0.00 0.00 179.00 0.00 89.50 0.00 1024.00 3.74 20.88 5.20 93.13
Dm-14 0.00 0.00 189.60 0.00 94.80 0.00 1024.00 3.84 20.27 4.96 94.07
Dm-13 0.00 0.00 188.20 0.00 94.10 0.00 1024.00 3.84 20.42 4.98 93.68
Dm-14 0.00 0.00 185.20 0.00 92.60 0.00 1024.00 3.75 20.27 5.01 92.79
Dm-13 0.00 0.00 180.00 0.00 90.00 0.00 1024.00 3.70 20.55 5.07 91.17
Dm-13 0.00 0.00 441.40 0.00 3.45 0.00 16.00 5.13 11.62 2.26 99.91
Dm-14 0.00 0.00 415.20 0.00 3.24 0.00 16.00 4.82 11.60 2.41 99.92
Dm-13 0.00 0.00 422.70 0.00 3.30 0.00 16.00 4.93 11.66 2.36 99.89
Dm-14 0.00 0.00 424.80 0.00 3.32 0.00 16.00 5.02 11.81 2.35 99.83
Dm-13 0.00 0.00 426.80 0.00 3.33 0.00 16.00 4.99 11.69 2.34 99.92
Dm-14 0.00 0.00 419.90 0.00 3.28 0.00 16.00 4.96 11.80 2.38 100.00

Rrqm/s: the number of merge read operations per second. That is, delta (rmerge)/s
Wrqm/s: Number of write operations performed on merge per second. That is, delta (wmerge)/s
R/s: The number of read I/O devices per second. That is, delta (rio)/s
W/s: the number of write I/O devices completed per second. That is, delta (wio)/s
Rsec/s: Number of read sectors per second. That is, delta (rsect)/s
Wsec/s: Number of write sectors per second. That is, delta (wsect)/s
RkB/s: the number of bytes read per second. It is half of rsect/s because the size of each slice is 512 bytes.
WkB/s: the number of K bytes written per second. Half of wsect/s.
Avgrq-sz: average data size (slice) of each device I/O operation ). That is, delta (rsect + wsect)/delta (rio + wio)
Avgqu-sz: Average I/O queue length. That is, delta (aveq)/s/1000 (because aveq is measured in milliseconds ).
Await: average wait time (in milliseconds) for each device I/O operation ). That is, delta (ruse + wuse)/delta (rio + wio)
Svctm: Average service time (in milliseconds) for each device I/O operation ). That is, delta (use)/delta (rio + wio)
% Util: the percentage of time in one second is used for I/O operations, or the number of I/O queues in one second is not empty.
That is, delta (use)/s/1000 (because the Unit of use is millisecond

Note:
IO ing can be obtained through the following command. Mpath8 is the dm-14
Multipath-ll

Mpath8 (360050768018086462800000000000055) dm-14 IBM, 2145
[Size = 500G] [features = 1 queue_if_no_path] [hwhandler = 0] [rw]
\ _ Round-robin 0 [prio = 100] [active]
\ _ 9: 0: 0: 9 sdak 66: 64 [active] [ready]
\ _ 8: 0: 0: 9 sdk 8: 160 [active] [ready]
\ _ Round-robin 0 [prio = 20] [enabled]
\ _ 9: 0: 1: 9 sdax 67: 16 [active] [ready]
\ _ 8: 0: 1: 9 sdx 65: 112 [active] [ready]
Mpath7 (360050768018086462800000000000054) dm-13 IBM, 2145
[Size = 500G] [features = 1 queue_if_no_path] [hwhandler = 0] [rw]
\ _ Round-robin 0 [prio = 100] [active]
\ _ 9: 0: 1: 8 sdaw 67: 0 [active] [ready]
\ _ 8: 0: 1: 8 sdw 65: 96 [active] [ready]
\ _ Round-robin 0 [prio = 20] [enabled]
\ _ 9: 0: 0: 8 sdaj 66: 48 [active] [ready]
\ _ 8: 0: 0: 8 sdj 8:144 [active] [ready]