Prior to Adaptive Server version 12.0, local transactions that executed RPCs or updated data via CIS could not roll back the work done in remote Adaptive Servers. Moreover, the client executing the local transaction could not ensure that the remote work was actually committed if, for example, the remote server experienced a system failure.
Adaptive Server provides services to propagate transactions to remote servers and coordinate the work of all servers, ensuring that all work is either committed or rolled back as a logical unit. With these transaction coordination services, Adaptive Server itself can act as a distributed transaction manager for transactions that update data in multiple Adaptive Servers.
