SEARCH
0-9 A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
Prev | Current Page 350 | Next

James Luetkehoelter

"Pro SQL Server Disaster Recovery"


Keep in mind that you must configure each active node so that in the event of a failure,
all nodes can run on a single server without clobbering each other. In an eight-node
active environment, that means a lot of configuration and a lot of testing. I??™minterested
to see one of these in production (as of this writing, I haven??™t).
If you have more than three physical nodes, I strongly advocate using the N+1 principle.
One server should always be sitting fallow, waiting to help out in the event of a
hardware failure. That node should always be physically as ???beefy??? as the strongest node
in your cluster. Trust me, when the hardware failure occurs, you??™ll be happy you configured
it that way. An active/active configuration on a two-node cluster requires more care
in configuration, but it??™s an acceptable situation. If you have the funds for a cluster of
three or more nodes, there really is no excuse for not having a passive node for failure
scenarios.
CHAPTER 7 n CLUSTERING 186
Figure 7-4. A four-node cluster
Multiple Instances
While a standalone installation supports 50 instances, a clustered environment can
only support 24 instances.


Pages:
338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362
hotel jelenia góra Russian bride Free English grammar and study guid powiekszenia wielkoformatowe counter strike 1.6