|
If you were logged in you would be able to see more operations.
|
|
|
|
Issue Links:
|
Relate
|
|
|
|
This issue is related to:
|
|
CORE-2530
Denser data stream and better prefetch logic in the network protocol
|
|
|
|
|
|
|
| QA Status: |
Done successfully
|
| Test Details: |
Measurement showed than on 3.0 (SS/SC/CS) transfer of NULLs is more than 5 (five) times
faster than text data with length = 32K. As of 2.5 (SC) than ration is about 1.7 ... 1.8.
Test fills up two tables: one with text data and another with only nulls.
Then we receive data from these tables via EZ/EDS, evaluate elapsed time for both cases
and calculate its ratio. This ratio in 3.0 should be not less than 4x.
Measurement showed than on 3.0 (SS/SC/CS) transfer of NULLs is more than 5 (five) times
faster than text data with length = 32K. As of 2.5 (SC) than ration is about 1.7 ... 1.8.
Test fills up two tables: one with text data and another with only nulls.
Then we receive data from these tables via EZ/EDS, evaluate elapsed time for both cases
and calculate its ratio. This ratio in 3.0 should be not less than 4x.
|
|
It seems that a char field is sent in its full declared length over the wire, even when the field is null, generating unnecessary network traffic. IMHO, this is ok when field contains some data, but should not occur when the field is null.
PS: Thanks Ivan Prenosil for testing.
|
|
Description
|
It seems that a char field is sent in its full declared length over the wire, even when the field is null, generating unnecessary network traffic. IMHO, this is ok when field contains some data, but should not occur when the field is null.
PS: Thanks Ivan Prenosil for testing. |
Show » |
| There are no subversion log entries for this issue yet.
|
|
Improvement
Don't send full length of field over the wire when field is null