<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<HTML
><HEAD
><TITLE
>Comparison of Different Solutions</TITLE
><META
NAME="GENERATOR"
CONTENT="Modular DocBook HTML Stylesheet Version 1.79"><LINK
REV="MADE"
HREF="mailto:pgsql-docs@postgresql.org"><LINK
REL="HOME"
TITLE="PostgreSQL 9.6.15 Documentation"
HREF="index.html"><LINK
REL="UP"
TITLE="High Availability, Load Balancing, and Replication"
HREF="high-availability.html"><LINK
REL="PREVIOUS"
TITLE="High Availability, Load Balancing, and Replication"
HREF="high-availability.html"><LINK
REL="NEXT"
TITLE="Log-Shipping Standby Servers"
HREF="warm-standby.html"><LINK
REL="STYLESHEET"
TYPE="text/css"
HREF="stylesheet.css"><META
HTTP-EQUIV="Content-Type"
CONTENT="text/html; charset=ISO-8859-1"><META
NAME="creation"
CONTENT="2019-08-11T16:06:51"></HEAD
><BODY
CLASS="SECT1"
><DIV
CLASS="NAVHEADER"
><TABLE
SUMMARY="Header navigation table"
WIDTH="100%"
BORDER="0"
CELLPADDING="0"
CELLSPACING="0"
><TR
><TH
COLSPAN="4"
ALIGN="center"
VALIGN="bottom"
><A
HREF="index.html"
>PostgreSQL 9.6.15 Documentation</A
></TH
></TR
><TR
><TD
WIDTH="10%"
ALIGN="left"
VALIGN="top"
><A
TITLE="High Availability, Load Balancing, and Replication"
HREF="high-availability.html"
ACCESSKEY="P"
>Prev</A
></TD
><TD
WIDTH="10%"
ALIGN="left"
VALIGN="top"
><A
HREF="high-availability.html"
ACCESSKEY="U"
>Up</A
></TD
><TD
WIDTH="60%"
ALIGN="center"
VALIGN="bottom"
>Chapter 26. High Availability, Load Balancing, and Replication</TD
><TD
WIDTH="20%"
ALIGN="right"
VALIGN="top"
><A
TITLE="Log-Shipping Standby Servers"
HREF="warm-standby.html"
ACCESSKEY="N"
>Next</A
></TD
></TR
></TABLE
><HR
ALIGN="LEFT"
WIDTH="100%"></DIV
><DIV
CLASS="SECT1"
><H1
CLASS="SECT1"
><A
NAME="DIFFERENT-REPLICATION-SOLUTIONS"
>26.1. Comparison of Different Solutions</A
></H1
><P
></P
><DIV
CLASS="VARIABLELIST"
><DL
><DT
>Shared Disk Failover</DT
><DD
><P
> Shared disk failover avoids synchronization overhead by having only one
copy of the database. It uses a single disk array that is shared by
multiple servers. If the main database server fails, the standby server
is able to mount and start the database as though it were recovering from
a database crash. This allows rapid failover with no data loss.
</P
><P
> Shared hardware functionality is common in network storage devices.
Using a network file system is also possible, though care must be
taken that the file system has full <ACRONYM
CLASS="ACRONYM"
>POSIX</ACRONYM
> behavior (see <A
HREF="creating-cluster.html#CREATING-CLUSTER-NFS"
>Section 18.2.2</A
>). One significant limitation of this
method is that if the shared disk array fails or becomes corrupt, the
primary and standby servers are both nonfunctional. Another issue is
that the standby server should never access the shared storage while
the primary server is running.
</P
></DD
><DT
>File System (Block-Device) Replication</DT
><DD
><P
> A modified version of shared hardware functionality is file system
replication, where all changes to a file system are mirrored to a file
system residing on another computer. The only restriction is that
the mirroring must be done in a way that ensures the standby server
has a consistent copy of the file system — specifically, writes
to the standby must be done in the same order as those on the master.
<SPAN
CLASS="PRODUCTNAME"
>DRBD</SPAN
> is a popular file system replication solution
for Linux.
</P
></DD
><DT
>Transaction Log Shipping</DT
><DD
><P
> Warm and hot standby servers can be kept current by reading a
stream of write-ahead log (<ACRONYM
CLASS="ACRONYM"
>WAL</ACRONYM
>)
records. If the main server fails, the standby contains
almost all of the data of the main server, and can be quickly
made the new master database server. This can be synchronous or
asynchronous and can only be done for the entire database server.
</P
><P
> A standby server can be implemented using file-based log shipping
(<A
HREF="warm-standby.html"
>Section 26.2</A
>) or streaming replication (see
<A
HREF="warm-standby.html#STREAMING-REPLICATION"
>Section 26.2.5</A
>), or a combination of both. For
information on hot standby, see <A
HREF="hot-standby.html"
>Section 26.5</A
>.
</P
></DD
><DT
>Trigger-Based Master-Standby Replication</DT
><DD
><P
> A master-standby replication setup sends all data modification
queries to the master server. The master server asynchronously
sends data changes to the standby server. The standby can answer
read-only queries while the master server is running. The
standby server is ideal for data warehouse queries.
</P
><P
> <SPAN
CLASS="PRODUCTNAME"
>Slony-I</SPAN
> is an example of this type of replication, with per-table
granularity, and support for multiple standby servers. Because it
updates the standby server asynchronously (in batches), there is
possible data loss during fail over.
</P
></DD
><DT
>Statement-Based Replication Middleware</DT
><DD
><P
> With statement-based replication middleware, a program intercepts
every SQL query and sends it to one or all servers. Each server
operates independently. Read-write queries must be sent to all servers,
so that every server receives any changes. But read-only queries can be
sent to just one server, allowing the read workload to be distributed
among them.
</P
><P
> If queries are simply broadcast unmodified, functions like
<CODE
CLASS="FUNCTION"
>random()</CODE
>, <CODE
CLASS="FUNCTION"
>CURRENT_TIMESTAMP</CODE
>, and
sequences can have different values on different servers.
This is because each server operates independently, and because
SQL queries are broadcast (and not actual modified rows). If
this is unacceptable, either the middleware or the application
must query such values from a single server and then use those
values in write queries. Another option is to use this replication
option with a traditional master-standby setup, i.e. data modification
queries are sent only to the master and are propagated to the
standby servers via master-standby replication, not by the replication
middleware. Care must also be taken that all
transactions either commit or abort on all servers, perhaps
using two-phase commit (<A
HREF="sql-prepare-transaction.html"
>PREPARE TRANSACTION</A
>
and <A
HREF="sql-commit-prepared.html"
>COMMIT PREPARED</A
>).
<SPAN
CLASS="PRODUCTNAME"
>Pgpool-II</SPAN
> and <SPAN
CLASS="PRODUCTNAME"
>Continuent Tungsten</SPAN
>
are examples of this type of replication.
</P
></DD
><DT
>Asynchronous Multimaster Replication</DT
><DD
><P
> For servers that are not regularly connected, like laptops or
remote servers, keeping data consistent among servers is a
challenge. Using asynchronous multimaster replication, each
server works independently, and periodically communicates with
the other servers to identify conflicting transactions. The
conflicts can be resolved by users or conflict resolution rules.
Bucardo is an example of this type of replication.
</P
></DD
><DT
>Synchronous Multimaster Replication</DT
><DD
><P
> In synchronous multimaster replication, each server can accept
write requests, and modified data is transmitted from the
original server to every other server before each transaction
commits. Heavy write activity can cause excessive locking,
leading to poor performance. In fact, write performance is
often worse than that of a single server. Read requests can
be sent to any server. Some implementations use shared disk
to reduce the communication overhead. Synchronous multimaster
replication is best for mostly read workloads, though its big
advantage is that any server can accept write requests —
there is no need to partition workloads between master and
standby servers, and because the data changes are sent from one
server to another, there is no problem with non-deterministic
functions like <CODE
CLASS="FUNCTION"
>random()</CODE
>.
</P
><P
> <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
> does not offer this type of replication,
though <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
> two-phase commit (<A
HREF="sql-prepare-transaction.html"
>PREPARE TRANSACTION</A
> and <A
HREF="sql-commit-prepared.html"
>COMMIT PREPARED</A
>)
can be used to implement this in application code or middleware.
</P
></DD
><DT
>Commercial Solutions</DT
><DD
><P
> Because <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
> is open source and easily
extended, a number of companies have taken <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
>
and created commercial closed-source solutions with unique
failover, replication, and load balancing capabilities.
</P
></DD
></DL
></DIV
><P
> <A
HREF="different-replication-solutions.html#HIGH-AVAILABILITY-MATRIX"
>Table 26-1</A
> summarizes
the capabilities of the various solutions listed above.
</P
><DIV
CLASS="TABLE"
><A
NAME="HIGH-AVAILABILITY-MATRIX"
></A
><P
><B
>Table 26-1. High Availability, Load Balancing, and Replication Feature Matrix</B
></P
><TABLE
BORDER="1"
CLASS="CALSTABLE"
><COL><COL><COL><COL><COL><COL><COL><COL><THEAD
><TR
><TH
>Feature</TH
><TH
>Shared Disk Failover</TH
><TH
>File System Replication</TH
><TH
>Transaction Log Shipping</TH
><TH
>Trigger-Based Master-Standby Replication</TH
><TH
>Statement-Based Replication Middleware</TH
><TH
>Asynchronous Multimaster Replication</TH
><TH
>Synchronous Multimaster Replication</TH
></TR
></THEAD
><TBODY
><TR
><TD
>Most Common Implementation</TD
><TD
ALIGN="CENTER"
>NAS</TD
><TD
ALIGN="CENTER"
>DRBD</TD
><TD
ALIGN="CENTER"
>Streaming Repl.</TD
><TD
ALIGN="CENTER"
>Slony</TD
><TD
ALIGN="CENTER"
>pgpool-II</TD
><TD
ALIGN="CENTER"
>Bucardo</TD
><TD
ALIGN="CENTER"
> </TD
></TR
><TR
><TD
>Communication Method</TD
><TD
ALIGN="CENTER"
>shared disk</TD
><TD
ALIGN="CENTER"
>disk blocks</TD
><TD
ALIGN="CENTER"
>WAL</TD
><TD
ALIGN="CENTER"
>table rows</TD
><TD
ALIGN="CENTER"
>SQL</TD
><TD
ALIGN="CENTER"
>table rows</TD
><TD
ALIGN="CENTER"
>table rows and row locks</TD
></TR
><TR
><TD
>No special hardware required</TD
><TD
ALIGN="CENTER"
> </TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
>•</TD
></TR
><TR
><TD
>Allows multiple master servers</TD
><TD
ALIGN="CENTER"
> </TD
><TD
ALIGN="CENTER"
> </TD
><TD
ALIGN="CENTER"
> </TD
><TD
ALIGN="CENTER"
> </TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
>•</TD
></TR
><TR
><TD
>No master server overhead</TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
> </TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
> </TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
> </TD
><TD
ALIGN="CENTER"
> </TD
></TR
><TR
><TD
>No waiting for multiple servers</TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
> </TD
><TD
ALIGN="CENTER"
>with sync off</TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
> </TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
> </TD
></TR
><TR
><TD
>Master failure will never lose data</TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
>with sync on</TD
><TD
ALIGN="CENTER"
> </TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
> </TD
><TD
ALIGN="CENTER"
>•</TD
></TR
><TR
><TD
>Standby accept read-only queries</TD
><TD
ALIGN="CENTER"
> </TD
><TD
ALIGN="CENTER"
> </TD
><TD
ALIGN="CENTER"
>with hot</TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
>•</TD
></TR
><TR
><TD
>Per-table granularity</TD
><TD
ALIGN="CENTER"
> </TD
><TD
ALIGN="CENTER"
> </TD
><TD
ALIGN="CENTER"
> </TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
> </TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
>•</TD
></TR
><TR
><TD
>No conflict resolution necessary</TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
>•</TD
><TD
ALIGN="CENTER"
> </TD
><TD
ALIGN="CENTER"
> </TD
><TD
ALIGN="CENTER"
>•</TD
></TR
></TBODY
></TABLE
></DIV
><P
> There are a few solutions that do not fit into the above categories:
</P
><P
></P
><DIV
CLASS="VARIABLELIST"
><DL
><DT
>Data Partitioning</DT
><DD
><P
> Data partitioning splits tables into data sets. Each set can
be modified by only one server. For example, data can be
partitioned by offices, e.g., London and Paris, with a server
in each office. If queries combining London and Paris data
are necessary, an application can query both servers, or
master/standby replication can be used to keep a read-only copy
of the other office's data on each server.
</P
></DD
><DT
>Multiple-Server Parallel Query Execution</DT
><DD
><P
> Many of the above solutions allow multiple servers to handle multiple
queries, but none allow a single query to use multiple servers to
complete faster. This solution allows multiple servers to work
concurrently on a single query. It is usually accomplished by
splitting the data among servers and having each server execute its
part of the query and return results to a central server where they
are combined and returned to the user. <SPAN
CLASS="PRODUCTNAME"
>Pgpool-II</SPAN
>
has this capability. Also, this can be implemented using the
<SPAN
CLASS="PRODUCTNAME"
>PL/Proxy</SPAN
> tool set.
</P
></DD
></DL
></DIV
></DIV
><DIV
CLASS="NAVFOOTER"
><HR
ALIGN="LEFT"
WIDTH="100%"><TABLE
SUMMARY="Footer navigation table"
WIDTH="100%"
BORDER="0"
CELLPADDING="0"
CELLSPACING="0"
><TR
><TD
WIDTH="33%"
ALIGN="left"
VALIGN="top"
><A
HREF="high-availability.html"
ACCESSKEY="P"
>Prev</A
></TD
><TD
WIDTH="34%"
ALIGN="center"
VALIGN="top"
><A
HREF="index.html"
ACCESSKEY="H"
>Home</A
></TD
><TD
WIDTH="33%"
ALIGN="right"
VALIGN="top"
><A
HREF="warm-standby.html"
ACCESSKEY="N"
>Next</A
></TD
></TR
><TR
><TD
WIDTH="33%"
ALIGN="left"
VALIGN="top"
>High Availability, Load Balancing, and Replication</TD
><TD
WIDTH="34%"
ALIGN="center"
VALIGN="top"
><A
HREF="high-availability.html"
ACCESSKEY="U"
>Up</A
></TD
><TD
WIDTH="33%"
ALIGN="right"
VALIGN="top"
>Log-Shipping Standby Servers</TD
></TR
></TABLE
></DIV
></BODY
></HTML
>
distrib
>
Mageia
>
6
>
x86_64
>
media
>
core-updates
>
by-pkgid
>
28c251aa384f373fb037135d1f7f0b9e
>
files
>
200
