Seagate ST43401N/ND Guide d'installation Page 67

  • Télécharger
  • Ajouter à mon manuel
  • Imprimer
  • Page
    / 146
  • Table des matières
  • MARQUE LIVRES
  • Noté. / 5. Basé sur avis des utilisateurs
Vue de la page 66
CHAPTER 5 Implement: Making Required Application Changes
59
To see whether you can enable wide data sizes for your client software, go to
the section below that describes your system:
Both ASE and Open Client are Version 12.5
Open Client Is an Older Version
ASE 12.5 Connects to a Remote Pre-12.5 Server
Both ASE and Open
Client are Version
12.5
You should not have any problems if both Adaptive Server and Open Client are
at version 12.5. Perform the following before you use Open Client 12.5 with
Adaptive Server version 12.5:
1 Relink your Open Client application with the 12.5 libraries.
2 Establish the new version number with CS_VERSION.
3 If you are using JConnect, reconnect the drivers.
Open Client Is an
Older Version
If you are running a pre-version 12.5 Open Client application with Adaptive
Server version 12.5, and have relinked your application with 12.5 Open Client,
but have not changed the version of your libraries, the pre-version 12.5
functionality is fine, but the new limits are not enabled. Adaptive Server sends
the data according to the pre-12.5 limits, and will truncate any data that is
beyond these limits.
If you are running pre-version 12.5 Open Client with Adaptive Server version
12.5, but have not relinked your libraries, Adaptive Server sends data according
to the pre-12.5 version, with the same restrictions outlined above.
ASE 12.5 Connects to
a Remote Pre-12.5
Server
If Open Client 12.5 connects to a 12.5 Adaptive Server, which then queries a
remote server which is at an earlier version, the remote server truncates any
wide data before returning results. All servers and clients in a transaction must
be at 12.5 in order to use the new wide limits.
Wide Columns and Optimizer Statistics
When placing statistics on a wide column, the server uses on the first 255 bytes
of data. Even if you place an index on a column created as
char(500), only
the first 255 bytes will contribute towards the column histograms in
sysstatistics. For unichar columns, the first 127 chars will be used to gather
column distributions.
Vue de la page 66
1 2 ... 62 63 64 65 66 67 68 69 70 71 72 ... 145 146

Commentaires sur ces manuels

Pas de commentaire