* Object rotation by Euler angles

* Indexing of spherical data types

This is an R-Tree implementation using GiST for spherical objects like

This is an R-tree implementation using GiST for spherical objects like

spherical points and spherical circles with useful functions and operators.

It alsosupport the Block Range INdexing (BRIN) for large datasets.

It alsosupports the Block Range INdexing (BRIN) for large datasets.

NOTICE:

This version will work only with PostgreSQL version 10 and above.

<application>pgSphere</application> uses <literal>GiST</literal>

and Block Range INdexing (<literal>BRIN</literal>) algorithms to create

spherical indices.

<literal>GiST</literal>index represents the R-Tree implementation for

spherical objects, while <literal>BRINs</literal> are based on "summarization"

<literal>GiST</literal>indexes utilize an R-tree implementation for

spherical objects, while <literal>BRIN</literal>indexesare based on the "summarization"

of data blocks (<literal>pages</literal>) on physical storage in order to

organize data searches on ranges of summarized data that can be easily skipped

on the base of search filters (see <ulink

url="https://www.postgresql.org/docs/9.5/static/brin-intro.html"><citetitle>

PostgreSQL documentation</citetitle></ulink> for further details onBRINs).

As a consequence,BRINs result to be reallysmall indexes (up to 1000 times

than GiST ones), generally with loweraperformance compared with a GiST one,

url="https://www.postgresql.org/docs/current/brin-intro.html"><citetitle>

PostgreSQL documentation</citetitle></ulink> for further details onBRIN indexes).

As a consequence,BRIN indexes are verysmall indexes (up to 1000 times smaller

than GiST ones), generally with lower performance compared with a GiST one,

but up to 100 times faster than a full sequential scan of a table performed

without any index. SoBRINs are particularly suitable in a big data context.

without any index. SoBRIN indexes are particularly suitable in a big data context.

An index speeds up the execution time of searches based on operators <link

linkend="op.over"><literal>&lt;@</literal></link>, <link

linkend="op.over"><literal>@</literal></link>, <link

<![CDATA[VACUUM ANALYZE test;]]>

</programlisting>

<para>

BRINs can be created through the following syntax:

BRIN index can be created through the following syntax:

</para>

<programlisting>

<![CDATA[CREATE INDEX test_pos_idx USING BRIN ON test (pos);]]>

</programlisting>

<para>

By default, BRINs summarize block of 128 pages. The lower numbers

of pages are specified, the higher granularity is reached during

the searches, and performance's gap between GiST indexes and BRINs

is lower (consider that BRINs size increases as well). Different

summarizations can be used with the following command:

By default, BRIN indexes summarize blocks of 128 pages. The smaller the

number of pages specified, the higher the granularity in searches,

and the gap in performance between GiST indexes and BRIN indexes will be

decreased. Note that the size of the BRIN indexes increases as well.

Different summarizations can be specified with the following

command:

</para>

<programlisting>

<![CDATA[CREATE INDEX test_pos_idx USING BRIN ON test (pos) WITH (pages_per_range = 16);]]>