15.5 Workload characteristics

---

The following output from the STATSPACK utility provides workload characteristics of the instance captured over the snapshot period. The average time against all events is the overall system-level performance characteristics.

 Global Cache Service - Workload Characteristics ---------------------------------------------- Ave global cache get time (ms):                      1.3 Ave global cache convert time (ms):                  8.2 Ave build time for CR block (ms):                    0.1 Ave flush time for CR block (ms):                    1.1 Ave send time for CR block (ms):                     0.2 Ave time to process CR block request (ms):           1.5 Ave receive time for CR block (ms):                  4.9 Ave pin time for current block (ms):                 0.3 Ave flush time for current block (ms):               0.0 Ave send time for current block (ms): Ave time to process current block request (ms):      0.6 Ave receive time for current block (ms):           634.6 Global cache hit ratio:                              0.2 Ratio of current block defers:                       0.1 % of messages sent for buffer gets:                  0.2 % of remote buffer gets:                             0.2 Ratio of I/O for coherence:                          0.2 Ratio of local vs remote work:                       0.4 Ratio of fusion vs physical writes:                  0.0 Global Enqueue Service Statistics -------------------------------- Ave global lock get time (ms):                       1.9 Ave global lock convert time (ms):                  48.8 Ratio of global lock gets vs global lock releases:   1.1

From the above output, the one wait event that stands out from the others is the ''Ave receive time for current block (ms).'' This is a measure of the average time a block has taken to arrive at the requesting instance from the time that the block was originally requested. A high value in the column indicates that there could be potential cluster interconnect issues including high latency. This needs to be resolved by looking at the cluster interconnect configuration as described under Section 15.3 above.

Table 15.3 is an Oracle-provided table indicating the acceptable latency on certain critical average waits listed in the STATSPACK report above. A large deviation from these figures almost always indicates perfor mance issues directly or indirectly related to the cluster interconnect. In a RAC configuration, an abnormally high value could also mean frequent block transfers across the cluster interconnect; under such circum stances, consider resizing the DB_FILE_MULTIBLOCK_READ_COUNT.

Table 15.3: Acceptable STATSPACK Latency Values^[a]

Latency	Lower Bound (ms)	Typical (ms)	Upper Bound (ms)
Ave global cache get time	0.3	3	10
Ave global cache convert time	0.3	6	15
Ave receive time for CR block	0.3	4	12
Ave time to process current block request	0.1	3	23
Ave receive time for current block	0.3	8	30
[a]Source: Oracle Corporation

---