| Title: | WinNT-Clusters |
| Notice: | Info directories moved to DECWET::SHARE1$:[NT_CLSTR] |
| Moderator: | DECWET::CAPPELLOF |
| Created: | Thu Oct 19 1995 |
| Last Modified: | Fri Jun 06 1997 |
| Last Successful Update: | Fri Jun 06 1997 |
| Number of topics: | 863 |
| Total number of notes: | 3478 |
I was discussing WNT clustering with a customer and he began to focus on the client software. His first concern was that this was but another software product that he would have to install and support on many desktop systems. He seemed to think that we could have hidden the cluster systems behind a single cluster name without adding code to the client. I didn't understand the details of what the client actually did to defend it. The secondary concern was the potential overhead required to have the client code do its job. Is there a technical write-up available on how the cluster client works? If so, please give me a pointer.
| T.R | Title | User | Personal Name | Date | Lines |
|---|---|---|---|---|---|
| 591.1 | MSE1::PCOTE | Rebuilt NT: 163, Rebuilt VMS:1 | Fri Jan 31 1997 17:08 | 7 | |
start with http://www.windows.digital.com/clusters/whtpap1.htp The client piece of the NT cluster software could become superfluous with IP failover and [hopefully] if/when Network share failover is implemented. | |||||
| 591.2 | Hope this is the answer | ZPOVC::SOONHENG | Sun Feb 02 1997 19:46 | 7 | |
Probably , my reply is too late.
If I understood your question correctly, the purpose of the client
cluster s/w is to identify the cluster alias name. Without the
client cluster s/w, you will need to specify each server node name
to interact with the cluster.
| |||||