I'm looking for a means to ensure that servers in the two or three datacenters, connected in a ring via IP through two ISPs, can distribute load and/or replicate data among at least two SAN-class disk devices.

I want to evaluate several solutions, and I'm open to solutions ranging from free, off-the-shelf, off-the-top-shelf (ie, really expensive), to custom designed by leading tech suppliers.

(Please help)

See this link: Designing Disaster Tolerant HA Clusters Using Metrocluster and Continentalclusters:

thanks for the link, but it's quite vague... it provides neither technical nor executive info. It seems it's written for a field engineer with prior knowledge of all the products. Is Metrocluster an HP product? I see that NetApp also has a product with a similar name. Does Metrocluster require a fiber link between the disk arrays?

are you looking for host based replication or should it be a storage based solution?

Did we read the same 470 page book? I have to say that it seems technical to me. But different strokes for different folks. Perhaps you missed the links to other info in the documentation roadmap in the preface? HP has thousands of pages of documents on-line about their HA product line. I don't know if HP got this from Netapp. I have never used it, but I've heard of it and just thought I'd mention it.

Host-based replication. We currently have a system for application redundancy and load balancing based on LVS+Squid+RH-clustering.

for a host based solution i can recommend:

Sun StorageTek Availability Suite Software - Overview
and
Lustre File System - Overview

hth,
DN2