Oracle Database Tips by Donald Burleson

When a System Will Not Benefit From Moving to SSD

The move to SSD assets from standard SCSI, ATA or SATA disks can be a blessing when it answers a specific performance problem related to disk I/O saturation. However, one must be careful when diagnosing the I/O related problems on a system. This section will provide some example STATSPACKs and use them to show whether or not there is a benefit to moving to SSD assets.

The following listing includes an events report for a system that will not benefit from SSD:

Top 5 Timed Events
~~~~~~~~~~~~~~~~~~                                                     % Total
Event                                               Waits    Time (s) Ela Time
-------------------------------------------- ------------ ----------- --------
CPU time                                                        1,127    73.25
global cache cr request                           213,187         122     7.95
db file sequential read                           152,521          96     6.27
control file sequential read                      118,104          78     5.06
SQL*Net message from dblink                           890          38     2.48
          -------------------------------------------------------------

The system in the above listing spends 73% of its time in the CPU with only 11 percent of the time spent waiting on disks. If the system were tuned to eliminate the CPU bottleneck, chances are the bottleneck will move to the disks and at that time, it would benefit from SSD technology. As the system in the listing stands right now, moving to SSD could actually hurt performance as it would place more stress on the already over worked CPU assets.

In the next example, the move to SSD might be beneficial since there is reserve CPU capacity of 40%, and it is spending the other wait time waiting on disks as shown in the following listing:

Top 5 Timed Events
~~~~~~~~~~~~~~~~~~                                                     % Total
Event                                               Waits    Time (s) Ela Time
-------------------------------------------- ------------ ----------- --------
CPU time                                                        1,300    60.46
db file sequential read                           342,625         616    28.67
db file scattered read                             12,986          66     3.07
log file parallel write                             2,889          65     3.03
db file parallel write                              1,080          59     2.75
          -------------------------------------------------------------

The system in the above listing indicates index and table stress as well as stress on log files. Assuming that the sample amount of time in the STATSPACK is representative of the overall system performance, the DBA needs to look further to determine what should be moved, tables or indexes, to the SSD asset.  The file I/O profile from this same report is shown in the following listing:

Tablespace
------------------------------
                 Av      Av     Av                    Av        Buffer Av Buf
         Reads Reads/s Rd(ms) Blks/Rd       Writes Writes/s      Waits Wt(ms)
-------------- ------- ------ ------- ------------ -------- ---------- ------
SRCD
        12,680       4    6.7     1.0       18,943        6          0    0.0
SYSTEM
        30,282      10    3.0     2.5          623        0          0    0.0
UNDOTBS1
            14       0   35.7     1.0       28,733        9          0    0.0
SRCX
         2,799       1    4.5     1.0       18,038        6          0    0.0
NOMADD
        16,604       5    1.8     1.0            8        0          0    0.0
TST_GLOBALX
         7,560       2    1.6     1.0           18        0          0    0.0
TST_GLOBALD
         6,242       2    2.0     1.2           36        0          0    0.0
XDB
         5,636       2    1.5     1.0            4        0          0    0.0
REEX
         4,240       1    2.0     1.0            4        0          0    0.0
ZENX
         3,812       1    2.1     1.0            4        0          0    0.0
ESRX
         3,656       1    1.6     1.0            4        0          0    0.0

The heavy-hitters in this listing are the srcd, system, nomad, tst_globallx and tst_globald tablespaces. The actual report from which the listing is extracted is over ten pages long for this section on datafiles, but these are the largest contributors to the I/O profile.  Analysis of the system showed improper use of the SYSTEM tablespace.  Once this was corrected, the others were left as the I/O stress points. Moving the heavy hitters to an SSD asset would do the following for this system:

• Shift the load to the CPUs

• Reduce I/O stress on the I/O subsystem allowing other datafiles to be accessed more efficiently.

• Speed access to the data/indexes contained in the moved datafiles.

The above benefits might actually provide greater than the percentage benefit quoted above. One of the other waits deals with redo log files, specifically log file parallel write.  Since this is a log file write specific wait, moving the redo logs would also show some benefit but not as great as that shown by moving tables and indexes.

The following listing provides a false positive indicator for use of SSD:

Top 5 Timed Events
~~~~~~~~~~~~~~~~~~                                                     % Total
Event                                               Waits    Time (s) Ela Time
-------------------------------------------- ------------ ----------- --------
db file sequential read                         6,261,550         691    96.72
control file parallel write                         1,274          19     2.73
CPU time                                                            2      .24
db file parallel write                                 28           1      .14
db file scattered read                              2,248           1      .12
          -------------------------------------------------------------

What is a false positive indicator? In this case, the STATSPACK seems to indicate that the database is doing a lot of full table scans and that this is 96-97 percent of wait times, which should indicate a move to SSD would be beneficial. However, a review of the entire report should be conducted.  The header for the file is shown in the following listing:

STATSPACK report for 

DB Name         DB Id    Instance     Inst Num Release     Cluster Host
------------ ----------- ------------ -------- ----------- ------- -----------
TSTPRD        3265066449 tstprd              1 9.2.0.3.0   NO      test08 

            Snap Id     Snap Time      Sessions Curs/Sess Comment
            ------- ------------------ -------- --------- -------------------
Begin Snap:       3 09-Nov-03 13:20:20       10       2.1
  End Snap:       4 09-Nov-03 14:26:01       10       2.1
   Elapsed:               65.68 (mins)
 

Cache Sizes (end)
~~~~~~~~~~~~~~~~~
               Buffer Cache:        24M      Std Block Size:         8K
           Shared Pool Size:        48M          Log Buffer:       512K

The tiny size for the buffer cache and shared pool should be noted along with the restricted time period monitored. Unless the server has severe memory limitations, the company using this database would be better off increasing the memory allocated to the instance and then looking at SSD if the waits are still an issue. The small elapsed time indicates that this STATSPACK run was probably for a specific transaction and is not indicative of full system load.

While a move to SSD may benefit many systems, the DBA should carefully review all information to ensure that another fix is not more appropriate.

See code depot for complete scripts

This is an excerpt from the book Oracle RAC & Tuning with Solid State Disk.  You can get it for more than 30% by buying it directly from the publisher and get immediate access to working code examples.

---

HTML-DB support:

For HTML-DB development support just call to get an Oracle Certified professional for all HTML-DB development projects.