Understanding the Limitations of Causally and Totally Ordered Communication

Understanding the Limitations of Causally and Totally Ordered Communication

Abstract
Causally and totally ordered communication support (CATOCS) has been proposed as important to provide as part of the basic building blocks for constructing reliable distributed systems. In this paper, we identify four major limitations to CATOCS, investigate the applicability of CATOCS to several classes of distributed applications in light of these limitations, and the potential impact of these facilities on communication scalability and robustness. From this investigation, we find limited merit and several potential problems in using CATOCS. The fundamental difficulty with the CN.OCS is that it attempts to solve state problems at the communication level in violation of the well-known “end-to-end” argument.

"The basic diftkulty with CATOCS can be best explained in terms of the end-to-end argument [25]. CATOCS is at the communication level, but consistency requirementc are typically expressed in terms of the application’s state."

"... incidental ordering ..."
"... semantic ordering ..."
"Consequently, many systems use or provide what we call prescriptive ordering where message delivery order is effectively based on ordering constraints explicitly specified or prescribed by a process at the time it sends a message. As described in the following sections, prescriptive ordering using state-level clocks, temporal or logical, offers an attractive alternative to CATOCS in many ml applications."

"We informally summarize these limitations of CATOCS as: 1) it can’t say "for sure", 2) it can’t say the "whole story", 3) it can’t say "together" and 4) it can’t say "efficiently".

I get it from Werner Vogels blog here.