Re: Re: xReader, double-effort (was: Temporary tables under hot standby)

<div class="gmail_extra">Hello Tom,</div><div class="gmail_extra"><br /></div><span style="style">> I'm not sure
I</span><brstyle="style" /><span style="style">> get the point of logical replication that requires a physical
replicaas</span><br style="style" /><div class="gmail_extra"><span style="style">> a prerequisite.</span></div><div
class="gmail_extra"><br/></div><div class="gmail_extra">> It would be enormously<br />> more performant for the
masterto be emitting logical replication<br /> > records to start with, since it already has all the right names
etc<br/>> at hand at basically no cost.<br /></div><div class="gmail_extra"><br /></div><div class="gmail_extra">I
wantto emphasize that any system which uses the logical reader setup, including a replication system, tends to be very
computationallyintensive on the database which it queries all so often. In most of the environments, the source
databaseis the main database, and as such, any performance degradation on this database is very bad.</div><div
class="gmail_extra"><br/></div><div class="gmail_extra">On the other hand, if we offload almost all the work to the
physicalreplica, our source database, which is the main database, still functions at the same throughput.</div><div
class="gmail_extra"><br/></div><div class="gmail_extra">Thus, at the cost of having the replication system as a whole
runa little slower than it could if it were using the main database, we have made sure that our performance critical
mainsource database is not affected in performance at all.</div>