[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference lassie::ucx

Title:	DEC TCP/IP Services for OpenVMS
Notice:	Note 2-SSB Kits, 3-FT Kits, 4-Patch Info, 7-QAR System
Moderator:	ucxaxp.ucx.lkg.dec.com::TIBBERT

Created:	Thu Nov 17 1994
Last Modified:	Fri Jun 06 1997
Last Successful Update:	Fri Jun 06 1997
Number of topics:	5568
Total number of notes:	21492

5285.0. "Cluster Alias load balance" by NIOPS2::THIBODEAU () Thu Feb 27 1997 12:44

    We are currently running UCX 4.0 Eco2 on production clusters in  the
    Northeast. Since so many people are now using IP we are running into
    load balancing problems. From reading notes in here I'm not clear on 
    what I need to impliment cluster alias load balancing. We resolve names
    via bindservers and winservers, does this cause me an issue. If anyone
    is familiar with how ucx is configured internal for CCS and knows how
    we can fix this I would very much appreciate a message.
    
    Thanks
    
    Alan

T.R Title User Personal
Name Date Lines

5285.1 your mileage may vary UTRTSC::KNOPPERS Oswald Knoppers Fri Feb 28 1997 02:33 11

T.R	Title	User	Personal Name	Date	Lines
5285.1	your mileage may vary	UTRTSC::KNOPPERS	Oswald Knoppers	`Fri Feb 28 1997 02:33`	11
	In order to do real load balancing, you will need all bind servers in use to be UCX based. This can be the main problem. However, if you have an alias defined, most (non-UCX) bind servers will do round-robin. In most cases this will give you a 'reasonable' balance. DECnet cluster alias, for instance, also does round-robin balancing (althoug you can define a fixed weight factor). Regards, Oswald

In order to do real load balancing, you will need all bind servers in use
to be UCX based. This can be the main problem.

However, if you have an alias defined, most (non-UCX) bind servers will do
round-robin. In most cases this will give you a 'reasonable' balance.
DECnet cluster alias, for instance, also does round-robin balancing
(althoug you can define a fixed weight factor).

Regards,

Oswald