Re: Implementing Incremental View Maintenance

From: Yugo NAGATA <nagata(at)sraoss(dot)co(dot)jp>
To: nuko yokohama <nuko(dot)yokohama(at)gmail(dot)com>
Cc: pgsql-hackers(at)postgresql(dot)org, Kevin Grittner <kgrittn(at)gmail(dot)com>
Subject: Re: Implementing Incremental View Maintenance
Date: 2020-01-17 08:11:53
Message-ID: 20200117171153.1aaf4fe24167732bb1941151@sraoss.co.jp
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Thu, 16 Jan 2020 12:59:11 +0900
nuko yokohama <nuko(dot)yokohama(at)gmail(dot)com> wrote:

> Aggregate operation of user-defined type cannot be specified
> (commit e150d964df7e3aeb768e4bae35d15764f8abd284)
>
> A SELECT statement using the MIN() and MAX() functions can be executed on a
> user-defined type column that implements the aggregate functions MIN () and
> MAX ().
> However, if the same SELECT statement is specified in the AS clause of
> CREATE INCREMENTAL MATERIALIZED VIEW, the following error will occur.
>
> ```
> SELECT MIN(data) data_min, MAX(data) data_max FROM foo;
> data_min | data_max
> ----------+----------
> 1/3 | 2/3
> (1 row)
>
> CREATE INCREMENTAL MATERIALIZED VIEW foo_min_imv AS SELECT MIN(data)
> data_min FROM foo;
> psql:extension-agg.sql:14: ERROR: aggregate function min is not supported
> CREATE INCREMENTAL MATERIALIZED VIEW foo_max_imv AS SELECT MAX(data)
> data_max FROM foo;
> psql:extension-agg.sql:15: ERROR: aggregate function max is not supported
> ```
>
> Does query including user-defined type aggregate operation not supported by
> INCREMENTAL MATERIALIZED VIEW?

The current implementation supports only built-in aggregate functions, so
user-defined aggregates are not supported, although it is allowed before.
This is because we can not know how user-defined aggregates behave and if
it can work safely with IVM. Min/Max on your fraction type may work well,
but it is possible that some user-defined aggregate functions named min
or max behave in totally different way than we expected.

In future, maybe it is possible support user-defined aggregates are supported
by extending pg_aggregate and adding support functions for IVM, but there is
not still a concrete plan for now.

BTW, the following error message doesn't look good because built-in min is
supported, so I will improve it.

ERROR: aggregate function min is not supported

Regards,
Yugo Nagata

>
> An execution example is shown below.
>
> ```
> [ec2-user(at)ip-10-0-1-10 ivm]$ cat extension-agg.sql
> --
> -- pg_fraction: https://github.com/nuko-yokohama/pg_fraction
> --
> DROP EXTENSION IF EXISTS pg_fraction CASCADE;
> DROP TABLE IF EXISTS foo CASCADE;
>
> CREATE EXTENSION IF NOT EXISTS pg_fraction;
> \dx
> \dT+ fraction
>
> CREATE TABLE foo (id int, data fraction);
> INSERT INTO foo (id, data) VALUES (1,'2/3'),(2,'1/3'),(3,'1/2');
> SELECT MIN(data) data_min, MAX(data) data_max FROM foo;
> CREATE INCREMENTAL MATERIALIZED VIEW foo_min_imv AS SELECT MIN(data)
> data_min FROM foo;
> CREATE INCREMENTAL MATERIALIZED VIEW foo_max_imv AS SELECT MAX(data)
> data_max FROM foo;
>
> SELECT MIN(id) id_min, MAX(id) id_max FROM foo;
> CREATE INCREMENTAL MATERIALIZED VIEW foo_id_imv AS SELECT MIN(id) id_min,
> MAX(id) id_max FROM foo;
> ```
>
> Best regards.
>
> 2018年12月27日(木) 21:57 Yugo Nagata <nagata(at)sraoss(dot)co(dot)jp>:
>
> > Hi,
> >
> > I would like to implement Incremental View Maintenance (IVM) on
> > PostgreSQL.
> > IVM is a technique to maintain materialized views which computes and
> > applies
> > only the incremental changes to the materialized views rather than
> > recomputate the contents as the current REFRESH command does.
> >
> > I had a presentation on our PoC implementation of IVM at PGConf.eu 2018
> > [1].
> > Our implementation uses row OIDs to compute deltas for materialized
> > views.
> > The basic idea is that if we have information about which rows in base
> > tables
> > are contributing to generate a certain row in a matview then we can
> > identify
> > the affected rows when a base table is updated. This is based on an idea of
> > Dr. Masunaga [2] who is a member of our group and inspired from ID-based
> > approach[3].
> >
> > In our implementation, the mapping of the row OIDs of the materialized view
> > and the base tables are stored in "OID map". When a base relation is
> > modified,
> > AFTER trigger is executed and the delta is recorded in delta tables using
> > the transition table feature. The accual udpate of the matview is triggerd
> > by REFRESH command with INCREMENTALLY option.
> >
> > However, we realize problems of our implementation. First, WITH OIDS will
> > be removed since PG12, so OIDs are no longer available. Besides this, it
> > would
> > be hard to implement this since it needs many changes of executor nodes to
> > collect base tables's OIDs during execuing a query. Also, the cost of
> > maintaining
> > OID map would be high.
> >
> > For these reasons, we started to think to implement IVM without relying on
> > OIDs
> > and made a bit more surveys.
> >
> > We also looked at Kevin Grittner's discussion [4] on incremental matview
> > maintenance. In this discussion, Kevin proposed to use counting algorithm
> > [5]
> > to handle projection views (using DISTNICT) properly. This algorithm need
> > an
> > additional system column, count_t, in materialized views and delta tables
> > of
> > base tables.
> >
> > However, the discussion about IVM is now stoped, so we would like to
> > restart and
> > progress this.
> >
> >
> > Through our PoC inplementation and surveys, I think we need to think at
> > least
> > the followings for implementing IVM.
> >
> > 1. How to extract changes on base tables
> >
> > I think there would be at least two approaches for it.
> >
> > - Using transition table in AFTER triggers
> > - Extracting changes from WAL using logical decoding
> >
> > In our PoC implementation, we used AFTER trigger and transition tables,
> > but using
> > logical decoding might be better from the point of performance of base
> > table
> > modification.
> >
> > If we can represent a change of UPDATE on a base table as query-like
> > rather than
> > OLD and NEW, it may be possible to update the materialized view directly
> > instead
> > of performing delete & insert.
> >
> >
> > 2. How to compute the delta to be applied to materialized views
> >
> > Essentially, IVM is based on relational algebra. Theorically, changes on
> > base
> > tables are represented as deltas on this, like "R <- R + dR", and the
> > delta on
> > the materialized view is computed using base table deltas based on "change
> > propagation equations". For implementation, we have to derive the
> > equation from
> > the view definition query (Query tree, or Plan tree?) and describe this as
> > SQL
> > query to compulte delta to be applied to the materialized view.
> >
> > There could be several operations for view definition: selection,
> > projection,
> > join, aggregation, union, difference, intersection, etc. If we can
> > prepare a
> > module for each operation, it makes IVM extensable, so we can start a
> > simple
> > view definition, and then support more complex views.
> >
> >
> > 3. How to identify rows to be modifed in materialized views
> >
> > When applying the delta to the materialized view, we have to identify
> > which row
> > in the matview is corresponding to a row in the delta. A naive method is
> > matching
> > by using all columns in a tuple, but clearly this is unefficient. If
> > thematerialized
> > view has unique index, we can use this. Maybe, we have to force
> > materialized views
> > to have all primary key colums in their base tables. In our PoC
> > implementation, we
> > used OID to identify rows, but this will be no longer available as said
> > above.
> >
> >
> > 4. When to maintain materialized views
> >
> > There are two candidates of the timing of maintenance, immediate (eager)
> > or deferred.
> >
> > In eager maintenance, the materialized view is updated in the same
> > transaction
> > where the base table is updated. In deferred maintenance, this is done
> > after the
> > transaction is commited, for example, when view is accessed, as a response
> > to user
> > request, etc.
> >
> > In the previous discussion[4], it is planned to start from "eager"
> > approach. In our PoC
> > implementaion, we used the other aproach, that is, using REFRESH command
> > to perform IVM.
> > I am not sure which is better as a start point, but I begin to think that
> > the eager
> > approach may be more simple since we don't have to maintain base table
> > changes in other
> > past transactions.
> >
> > In the eager maintenance approache, we have to consider a race condition
> > where two
> > different transactions change base tables simultaneously as discussed in
> > [4].
> >
> >
> > [1]
> > https://www.postgresql.eu/events/pgconfeu2018/schedule/session/2195-implementing-incremental-view-maintenance-on-postgresql/
> > [2]
> > https://ipsj.ixsq.nii.ac.jp/ej/index.php?active_action=repository_view_main_item_detail&page_id=13&block_id=8&item_id=191254&item_no=1
> > (Japanese only)
> > [3] https://dl.acm.org/citation.cfm?id=2750546
> > [4]
> > https://www.postgresql.org/message-id/flat/1368561126.64093.YahooMailNeo%40web162904.mail.bf1.yahoo.com
> > [5] https://dl.acm.org/citation.cfm?id=170066
> >
> > Regards,
> > --
> > Yugo Nagata <nagata(at)sraoss(dot)co(dot)jp>
> >
> >

--
Yugo NAGATA <nagata(at)sraoss(dot)co(dot)jp>

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Vik Fearing 2020-01-17 08:13:06 Re: FETCH FIRST clause PERCENT option
Previous Message Andreas Joseph Krogh 2020-01-17 07:49:49 Re: Minimal logical decoding on standbys