Skip to content

Commit cffc351

Browse files
author
stevestein
committed
sqldb - update per feedback
1 parent cf3e82d commit cffc351

File tree

1 file changed

+5
-10
lines changed

1 file changed

+5
-10
lines changed

articles/sql-database/import-export-from-vm.md

Lines changed: 5 additions & 10 deletions
Original file line numberDiff line numberDiff line change
@@ -134,21 +134,16 @@ sqlpackage.exe /a:Import /sf:testExport.bacpac /tdn:NewDacFX /tsn:apptestserver.
134134

135135
## Performance considerations
136136

137-
Export speeds vary due to many factors (for example, data shape) so it's impossible to predict what speed should be expected. The import export service uses the same technology as SqlPackage and may take considerable time, particularly for large databases.
137+
Export speeds vary due to many factors (for example, data shape) so it's impossible to predict what speed should be expected. SqlPackage may take considerable time, particularly for large databases.
138138

139-
To get the best performance using the import export service you can try the following strategies:
139+
To get the best performance you can try the following strategies:
140140

141141
1. Make sure no other workload is running on the database. Create a copy before export may be the best solution to ensure no other workloads are running.
142142
2. Increase database service level objective (SLO) to better handle the export workload (primarily read I/O). If the database is currently GP_Gen5_4, perhaps a Business Critical tier would help with read workload.
143143
3. Make sure there are clustered indexes particularly for large tables.
144-
145-
146-
Even with all of the above conditions, the import export service may not meet the performance targets of the customer. For comparison the customer can try to run SqlPackage on their own in an Azure VM, with the following strategies:
147-
148-
1. Same concerns over database workload and SLO as previously mentioned.
149-
2. VM should be in the same region as the database to help avoid network constraints.
150-
3. VM should have SSD with adequate size for generating temp artifacts before uploading to blob storage.
151-
4. VM should have adequate core and memory configuration for the specific database.
144+
4. Virtual machines (VMs) should be in the same region as the database to help avoid network constraints.
145+
5. VMs should have SSD with adequate size for generating temp artifacts before uploading to blob storage.
146+
6. VMs should have adequate core and memory configuration for the specific database.
152147

153148
## Store the imported or exported .BACPAC file
154149

0 commit comments

Comments
 (0)