Foreign Tables

Foreign tables can be used to run queries directly on data without inserting it into SQream DB first.

SQream DB supports read only foreign tables, so you can query from foreign tables, but you cannot insert to them, or run deletes or updates on them.

Running queries directly on foreign (external) data is most effectively used for things like one off querying. If you will be repeatedly querying data, the performance will usually be better if you insert the data into SQream DB first.

Although foreign tables can be used without inserting data into SQream DB, one of their main use cases is to help with the insertion process. An insert select statement on an foreign table can be used to insert data into SQream using the full power of the query engine to perform ETL.

What kind of data is supported?

SQream DB uses foreign data wrappers (FDW) to abstract external sources.

SQream DB supports these FDWs:

  • text files (e.g. CSV, PSV, TSV) via the csv_fdw
  • ORC via the orc_fdw
  • Parquet via the parquet_fdw

What kind of data staging is supported?

SQream DB can stage data from:

  • a local filesystem (e.g. /mnt/storage/....)
  • S3 buckets (e.g. s3://pp-secret-bucket/users/*.parquet)
  • HDFS (e.g. hdfs://hadoop-nn.piedpiper.com/rhendricks/*.csv)

Using foreign tables - a practical example

Use an foreign table to stage data before loading from CSV, Parquet or ORC files.

Planning for data staging

For the following examples, we will want to interact with a CSV file. Here’s a peek at the table contents:

nba.csv
Name Team Number Position Age Height Weight College Salary
Avery Bradley Boston Celtics 0.0 PG 25.0 6-2 180.0 Texas 7730337.0
Jae Crowder Boston Celtics 99.0 SF 25.0 6-6 235.0 Marquette 6796117.0
John Holland Boston Celtics 30.0 SG 27.0 6-5 205.0 Boston University  
R.J. Hunter Boston Celtics 28.0 SG 22.0 6-5 185.0 Georgia State 1148640.0
Jonas Jerebko Boston Celtics 8.0 PF 29.0 6-10 231.0   5000000.0
Amir Johnson Boston Celtics 90.0 PF 29.0 6-9 240.0   12000000.0
Jordan Mickey Boston Celtics 55.0 PF 21.0 6-8 235.0 LSU 1170960.0
Kelly Olynyk Boston Celtics 41.0 C 25.0 7-0 238.0 Gonzaga 2165160.0
Terry Rozier Boston Celtics 12.0 PG 22.0 6-2 190.0 Louisville 1824360.0

The file is stored on S3, at s3://sqream-demo-data/nba_players.csv.

We will make note of the file structure, to create a matching CREATE_EXTERNAL_TABLE statement.

Creating the foreign table

Based on the source file structure, we we create a foreign table with the appropriate structure, and point it to the file.

The file format in this case is CSV, with a DOS newline (\r\n).

Here’s how the table would be created from S3 and HDFS:

From S3

CREATE FOREIGN TABLE nba
(
   Name varchar(40),
   Team varchar(40),
   Number tinyint,
   Position varchar(2),
   Age tinyint,
   Height varchar(4),
   Weight real,
   College varchar(40),
   Salary float
 )
   WRAPPER csv_fdw -- Text file
   OPTIONS
     ( LOCATION = 's3://sqream-demo-data/nba_players.csv',
       RECORD_DELIMITER = '\r\n'; -- DOS delimited file
     )
 ;

From HDFS

CREATE FOREIGN TABLE nba
(
   Name varchar(40),
   Team varchar(40),
   Number tinyint,
   Position varchar(2),
   Age tinyint,
   Height varchar(4),
   Weight real,
   College varchar(40),
   Salary float
 )
   WRAPPER csv_fdw -- Text file
   OPTIONS
     (
       LOCATION = 'hdfs://hadoop-nn.piedpiper.com:8020/demo-data/nba_players.csv',
       RECORD_DELIMITER = '\r\n'; -- DOS delimited file
     )
;

Querying foreign tables

Let’s peek at the data from the foreign table:

t=> SELECT * FROM nba LIMIT 10;
name          | team           | number | position | age | height | weight | college           | salary
--------------+----------------+--------+----------+-----+--------+--------+-------------------+---------
Avery Bradley | Boston Celtics |      0 | PG       |  25 | 6-2    |    180 | Texas             |  7730337
Jae Crowder   | Boston Celtics |     99 | SF       |  25 | 6-6    |    235 | Marquette         |  6796117
John Holland  | Boston Celtics |     30 | SG       |  27 | 6-5    |    205 | Boston University |
R.J. Hunter   | Boston Celtics |     28 | SG       |  22 | 6-5    |    185 | Georgia State     |  1148640
Jonas Jerebko | Boston Celtics |      8 | PF       |  29 | 6-10   |    231 |                   |  5000000
Amir Johnson  | Boston Celtics |     90 | PF       |  29 | 6-9    |    240 |                   | 12000000
Jordan Mickey | Boston Celtics |     55 | PF       |  21 | 6-8    |    235 | LSU               |  1170960
Kelly Olynyk  | Boston Celtics |     41 | C        |  25 | 7-0    |    238 | Gonzaga           |  2165160
Terry Rozier  | Boston Celtics |     12 | PG       |  22 | 6-2    |    190 | Louisville        |  1824360
Marcus Smart  | Boston Celtics |     36 | PG       |  22 | 6-4    |    220 | Oklahoma State    |  3431040

Modifying data from staging

One of the main reasons for staging data is to examine the contents and modify them before loading them.

Assume we are unhappy with weight being in pounds, because we want to use kilograms instead. We can apply the transformation as part of a query:

t=> SELECT name, team, number, position, age, height, (weight / 2.205) as weight, college, salary
.          FROM nba
.          ORDER BY weight;
name                     | team                   | number | position | age | height | weight   | college               | salary
-------------------------+------------------------+--------+----------+-----+--------+----------+-----------------------+---------
Nikola Pekovic           | Minnesota Timberwolves |     14 | C        |  30 | 6-11   |  139.229 |                       | 12100000
Boban Marjanovic         | San Antonio Spurs      |     40 | C        |  27 | 7-3    | 131.5193 |                       |  1200000
Al Jefferson             | Charlotte Hornets      |     25 | C        |  31 | 6-10   | 131.0658 |                       | 13500000
Jusuf Nurkic             | Denver Nuggets         |     23 | C        |  21 | 7-0    | 126.9841 |                       |  1842000
Andre Drummond           | Detroit Pistons        |      0 | C        |  22 | 6-11   | 126.5306 | Connecticut           |  3272091
Kevin Seraphin           | New York Knicks        |      1 | C        |  26 | 6-10   | 126.0771 |                       |  2814000
Brook Lopez              | Brooklyn Nets          |     11 | C        |  28 | 7-0    | 124.7166 | Stanford              | 19689000
Jahlil Okafor            | Philadelphia 76ers     |      8 | C        |  20 | 6-11   | 124.7166 | Duke                  |  4582680
Cristiano Felicio        | Chicago Bulls          |      6 | PF       |  23 | 6-10   | 124.7166 |                       |   525093
[...]

Now, if we’re happy with the results, we can convert the staged foreign table to a standard table

Converting a foreign table to a standard database table

CREATE TABLE AS can be used to materialize a foreign table into a regular table.

Tip

If you intend to use the table multiple times, convert the foreign table to a standard table.

t=> CREATE TABLE real_nba AS
.    SELECT name, team, number, position, age, height, (weight / 2.205) as weight, college, salary
.            FROM nba
.            ORDER BY weight;
executed
t=> SELECT * FROM real_nba LIMIT 5;
name             | team                   | number | position | age | height | weight   | college     | salary
-----------------+------------------------+--------+----------+-----+--------+----------+-------------+---------
Nikola Pekovic   | Minnesota Timberwolves |     14 | C        |  30 | 6-11   |  139.229 |             | 12100000
Boban Marjanovic | San Antonio Spurs      |     40 | C        |  27 | 7-3    | 131.5193 |             |  1200000
Al Jefferson     | Charlotte Hornets      |     25 | C        |  31 | 6-10   | 131.0658 |             | 13500000
Jusuf Nurkic     | Denver Nuggets         |     23 | C        |  21 | 7-0    | 126.9841 |             |  1842000
Andre Drummond   | Detroit Pistons        |      0 | C        |  22 | 6-11   | 126.5306 | Connecticut |  3272091

Error handling and limitations

  • Error handling in foreign tables is limited. Any error that occurs during source data parsing will result in the statement aborting.

  • Foreign tables are logical and do not contain any data, their structure is not verified or enforced until a query uses the table.

    For example, a CSV with the wrong delimiter may cause a query to fail, even though the table has been created successfully:

    t=> SELECT * FROM nba;
    master=> select * from nba;
    Record delimiter mismatch during CSV parsing. User defined line delimiter \n does not match the first delimiter \r\n found in s3://sqream-demo-data/nba.csv
    
  • Since the data for a foreign table is not stored in SQream DB, it can be changed or removed at any time by an external process. As a result, the same query can return different results each time it runs against a foreign table. Similarly, a query might fail if the external data is moved, removed, or has changed structure.