9.19. GIN Indexes

9.19.1. Introduction

GIN stands for Generalized Inverted Index. GIN is designed for handling cases where the items to be indexed are composite values, and the queries to be handled by the index need to search for element values that appear within the composite items. For example, the items could be documents, and the queries could be searches for documents containing specific words.

We use the word item to refer to a composite value that is to be indexed, and the word key to refer to an element value. GIN always stores and searches for keys, not item values per se.

A GIN index stores a set of (key, posting list) pairs, where a posting list is a set of row IDs in which the key occurs. The same row ID can appear in multiple posting lists, since an item can contain more than one key. Each key value is stored only once, so a GIN index is very compact for cases where the same key appears many times.

GIN is generalized in the sense that the GIN access method code does not need to know the specific operations that it accelerates. Instead, it uses custom strategies defined for particular data types. The strategy defines how keys are extracted from indexed items and query conditions, and how to determine whether a row that contains some of the key values in a query actually satisfies the query.

One advantage of GIN is that it allows the development of custom data types with the appropriate access methods, by an expert in the domain of the data type, rather than a database expert. This is much the same advantage as using GiST.

The GIN implementation in PostgreSQL is primarily maintained by Teodor Sigaev and Oleg Bartunov. There is more information about GIN on their website.

9.19.2. Built-in Operator Classes

The core PostgreSQL distribution includes the GIN operator classes shown in gin-builtin-opclasses-table. (Some of the optional modules described in contrib provide additional GIN operator classes.)

Built-in GIN Operator Classes

NameIndexable Operator :widths: 10, 10 **array_ops****&& (anyarray,anyarray)* **@> (anyarray,anyarray)* **<@ (anyarray,anyarray)* **= (anyarray,anyarray)* **jsonb_ops****@> (jsonb,jsonb)* **@? (jsonb,jsonpath)* **@@ (jsonb,jsonpath)* **? (jsonb,text)* **?| (jsonb,text[])* **?& (jsonb,text[])* **jsonb_path_ops****@> (jsonb,jsonb)* **@? (jsonb,jsonpath)* **@@ (jsonb,jsonpath)* **tsvector_ops****@@ (tsvector,tsquery)* **@@@ (tsvector,tsquery)*

Of the two operator classes for type jsonb, jsonb_ops is the default. jsonb_path_ops supports fewer operators but offers better performance for those operators. See json-indexing for details.

9.19.3. Extensibility

The GIN interface has a high level of abstraction, requiring the access method implementer only to implement the semantics of the data type being accessed. The GIN layer itself takes care of concurrency, logging and searching the tree structure.

All it takes to get a GIN access method working is to implement a few user-defined methods, which define the behavior of keys in the tree and the relationships between keys, indexed items, and indexable queries. In short, GIN combines extensibility with generality, code reuse, and a clean interface.

There are two methods that an operator class for GIN must provide:

1. Returns a palloc’d array of keys given an item to be indexed. The number of returned keys must be stored into *nkeys. If any of the keys can be null, also palloc an array of *nkeys bool fields, store its address at *nullFlags, and set these null flags as needed. *nullFlags can be left NULL (its initial value) if all keys are non-null. The return value can be NULL if the item contains no keys.

2. Returns a palloc’d array of keys given a value to be queried; that is, query is the value on the right-hand side of an indexable operator whose left-hand side is the indexed column. n is the strategy number of the operator within the operator class (see xindex-strategies). Often, extractQuery will need to consult n to determine the data type of query and the method it should use to extract key values. The number of returned keys must be stored into *nkeys. If any of the keys can be null, also palloc an array of *nkeys bool fields, store its address at *nullFlags, and set these null flags as needed. *nullFlags can be left NULL (its initial value) if all keys are non-null. The return value can be NULL if the query contains no keys.

   searchMode is an output argument that allows extractQuery to specify details about how the search will be done. If *searchMode is set to GIN_SEARCH_MODE_DEFAULT (which is the value it is initialized to before call), only items that match at least one of the returned keys are considered candidate matches. If *searchMode is set to GIN_SEARCH_MODE_INCLUDE_EMPTY, then in addition to items containing at least one matching key, items that contain no keys at all are considered candidate matches. (This mode is useful for implementing is-subset-of operators, for example.) If *searchMode is set to GIN_SEARCH_MODE_ALL, then all non-null items in the index are considered candidate matches, whether they match any of the returned keys or not. (This mode is much slower than the other two choices, since it requires scanning essentially the entire index, but it may be necessary to implement corner cases correctly. An operator that needs this mode in most cases is probably not a good candidate for a GIN operator class.) The symbols to use for setting this mode are defined in access/gin.h.

   pmatch is an output argument for use when partial match is supported. To use it, extractQuery must allocate an array of *nkeys bool**s and store its address at ***pmatch. Each element of the array should be set to true if the corresponding key requires partial match, false if not. If *pmatch is set to NULL then GIN assumes partial match is not required. The variable is initialized to NULL before call, so this argument can simply be ignored by operator classes that do not support partial match.

   extra_data is an output argument that allows extractQuery to pass additional data to the consistent and comparePartial methods. To use it, extractQuery must allocate an array of *nkeys pointers and store its address at *extra_data, then store whatever it wants to into the individual pointers. The variable is initialized to NULL before call, so this argument can simply be ignored by operator classes that do not require extra data. If *extra_data is set, the whole array is passed to the consistent method, and the appropriate element to the comparePartial method.

An operator class must also provide a function to check if an indexed item matches the query. It comes in two flavors, a Boolean consistent function, and a ternary triConsistent function. triConsistent covers the functionality of both, so providing triConsistent alone is sufficient. However, if the Boolean variant is significantly cheaper to calculate, it can be advantageous to provide both. If only the Boolean variant is provided, some optimizations that depend on refuting index items before fetching all the keys are disabled.

1. Returns true if an indexed item satisfies the query operator with strategy number n (or might satisfy it, if the recheck indication is returned). This function does not have direct access to the indexed item’s value, since GIN does not store items explicitly. Rather, what is available is knowledge about which key values extracted from the query appear in a given indexed item. The check array has length nkeys, which is the same as the number of keys previously returned by extractQuery for this query datum. Each element of the check array is true if the indexed item contains the corresponding query key, i.e., if (check[i] == true) the i-th key of the extractQuery result array is present in the indexed item. The original query datum is passed in case the consistent method needs to consult it, and so are the queryKeys[] and nullFlags[] arrays previously returned by extractQuery. extra_data is the extra-data array returned by extractQuery, or NULL if none.

   When extractQuery returns a null key in queryKeys[], the corresponding check[] element is true if the indexed item contains a null key; that is, the semantics of check[] are like IS NOT DISTINCT FROM. The consistent function can examine the corresponding nullFlags[] element if it needs to tell the difference between a regular value match and a null match.

   On success, *recheck should be set to true if the heap tuple needs to be rechecked against the query operator, or false if the index test is exact. That is, a false return value guarantees that the heap tuple does not match the query; a true return value with *recheck set to false guarantees that the heap tuple does match the query; and a true return value with *recheck set to true means that the heap tuple might match the query, so it needs to be fetched and rechecked by evaluating the query operator directly against the originally indexed item.

2. triConsistent is similar to consistent, but instead of Booleans in the check vector, there are three possible values for each key: GIN_TRUE, GIN_FALSE and GIN_MAYBE. GIN_FALSE and GIN_TRUE have the same meaning as regular Boolean values, while GIN_MAYBE means that the presence of that key is not known. When GIN_MAYBE values are present, the function should only return GIN_TRUE if the item certainly matches whether or not the index item contains the corresponding query keys. Likewise, the function must return GIN_FALSE only if the item certainly does not match, whether or not it contains the GIN_MAYBE keys. If the result depends on the GIN_MAYBE entries, i.e., the match cannot be confirmed or refuted based on the known query keys, the function must return GIN_MAYBE.

   When there are no GIN_MAYBE values in the check vector, a GIN_MAYBE return value is the equivalent of setting the recheck flag in the Boolean consistent function.

In addition, GIN must have a way to sort the key values stored in the index. The operator class can define the sort ordering by specifying a comparison method:

1. Compares two keys (not indexed items!) and returns an integer less than zero, zero, or greater than zero, indicating whether the first key is less than, equal to, or greater than the second. Null keys are never passed to this function.

Alternatively, if the operator class does not provide a compare method, GIN will look up the default btree operator class for the index key data type, and use its comparison function. It is recommended to specify the comparison function in a GIN operator class that is meant for just one data type, as looking up the btree operator class costs a few cycles. However, polymorphic GIN operator classes (such as array_ops) typically cannot specify a single comparison function.

An operator class for GIN can optionally supply the following methods:

1. Compare a partial-match query key to an index key. Returns an integer whose sign indicates the result: less than zero means the index key does not match the query, but the index scan should continue; zero means that the index key does match the query; greater than zero indicates that the index scan should stop because no more matches are possible. The strategy number n of the operator that generated the partial match query is provided, in case its semantics are needed to determine when to end the scan. Also, extra_data is the corresponding element of the extra-data array made by extractQuery, or NULL if none. Null keys are never passed to this function.

2. Defines a set of user-visible parameters that control operator class behavior.

   The options function is passed a pointer to a local_relopts struct, which needs to be filled with a set of operator class specific options. The options can be accessed from other support functions using the PG_HAS_OPCLASS_OPTIONS() and PG_GET_OPCLASS_OPTIONS() macros.

   Since both key extraction of indexed values and representation of the key in GIN are flexible, they may depend on user-specified parameters.

To support partial match queries, an operator class must provide the comparePartial method, and its extractQuery method must set the pmatch parameter when a partial-match query is encountered. See gin-partial-match for details.

The actual data types of the various Datum values mentioned above vary depending on the operator class. The item values passed to extractValue are always of the operator class’s input type, and all key values must be of the class’s STORAGE type. The type of the query argument passed to extractQuery, consistent and triConsistent is whatever is the right-hand input type of the class member operator identified by the strategy number. This need not be the same as the indexed type, so long as key values of the correct type can be extracted from it. However, it is recommended that the SQL declarations of these three support functions use the opclass’s indexed data type for the query argument, even though the actual type might be something else depending on the operator.

9.19.4. Implementation

Internally, a GIN index contains a B-tree index constructed over keys, where each key is an element of one or more indexed items (a member of an array, for example) and where each tuple in a leaf page contains either a pointer to a B-tree of heap pointers (a posting tree), or a simple list of heap pointers (a posting list) when the list is small enough to fit into a single index tuple along with the key value. gin-internals-figure illustrates these components of a GIN index.

As of PostgreSQL 9.1, null key values can be included in the index. Also, placeholder nulls are included in the index for indexed items that are null or contain no keys according to extractValue. This allows searches that should find empty items to do so.

Multicolumn GIN indexes are implemented by building a single B-tree over composite values (column number, key value). The key values for different columns can be of different types.

GIN Internals

fileref=»images/gin.svg» format=»SVG» width=»100%»

9.19.4.1. GIN Fast Update Technique

Updating a GIN index tends to be slow because of the intrinsic nature of inverted indexes: inserting or updating one heap row can cause many inserts into the index (one for each key extracted from the indexed item). GIN is capable of postponing much of this work by inserting new tuples into a temporary, unsorted list of pending entries. When the table is vacuumed or autoanalyzed, or when gin_clean_pending_list function is called, or if the pending list becomes larger than guc-gin-pending-list-limit, the entries are moved to the main GIN data structure using the same bulk insert techniques used during initial index creation. This greatly improves GIN index update speed, even counting the additional vacuum overhead. Moreover the overhead work can be done by a background process instead of in foreground query processing.

The main disadvantage of this approach is that searches must scan the list of pending entries in addition to searching the regular index, and so a large list of pending entries will slow searches significantly. Another disadvantage is that, while most updates are fast, an update that causes the pending list to become too large will incur an immediate cleanup cycle and thus be much slower than other updates. Proper use of autovacuum can minimize both of these problems.

If consistent response time is more important than update speed, use of pending entries can be disabled by turning off the fastupdate storage parameter for a GIN index. See sql-createindex for details.

9.19.4.2. Partial Match Algorithm

GIN can support partial match queries, in which the query does not determine an exact match for one or more keys, but the possible matches fall within a reasonably narrow range of key values (within the key sorting order determined by the compare support method). The extractQuery method, instead of returning a key value to be matched exactly, returns a key value that is the lower bound of the range to be searched, and sets the pmatch flag true. The key range is then scanned using the comparePartial method. comparePartial must return zero for a matching index key, less than zero for a non-match that is still within the range to be searched, or greater than zero if the index key is past the range that could match.

9.19.5. GIN Tips and Tricks

1. Insertion into a GIN index can be slow due to the likelihood of many keys being inserted for each item. So, for bulk insertions into a table it is advisable to drop the GIN index and recreate it after finishing bulk insertion.

   When fastupdate is enabled for GIN (see gin-fast-update for details), the penalty is less than when it is not. But for very large updates it may still be best to drop and recreate the index.

2. Build time for a GIN index is very sensitive to the maintenance_work_mem setting; it doesn’t pay to skimp on work memory during index creation.

3. During a series of insertions into an existing GIN index that has fastupdate enabled, the system will clean up the pending-entry list whenever the list grows larger than gin_pending_list_limit. To avoid fluctuations in observed response time, it’s desirable to have pending-list cleanup occur in the background (i.e., via autovacuum). Foreground cleanup operations can be avoided by increasing gin_pending_list_limit or making autovacuum more aggressive. However, enlarging the threshold of the cleanup operation means that if a foreground cleanup does occur, it will take even longer.

   gin_pending_list_limit can be overridden for individual GIN indexes by changing storage parameters, which allows each GIN index to have its own cleanup threshold. For example, it’s possible to increase the threshold only for the GIN index which can be updated heavily, and decrease it otherwise.

4. The primary goal of developing GIN indexes was to create support for highly scalable full-text search in PostgreSQL, and there are often situations when a full-text search returns a very large set of results. Moreover, this often happens when the query contains very frequent words, so that the large result set is not even useful. Since reading many tuples from the disk and sorting them could take a lot of time, this is unacceptable for production. (Note that the index search itself is very fast.)

   To facilitate controlled execution of such queries, GIN has a configurable soft upper limit on the number of rows returned: the gin_fuzzy_search_limit configuration parameter. It is set to 0 (meaning no limit) by default. If a non-zero limit is set, then the returned set is a subset of the whole result set, chosen at random.

   Soft means that the actual number of returned results could differ somewhat from the specified limit, depending on the query and the quality of the system’s random number generator.

   From experience, values in the thousands (e.g., 5000 — 20000) work well.

9.19.6. Limitations

GIN assumes that indexable operators are strict. This means that extractValue will not be called at all on a null item value (instead, a placeholder index entry is created automatically), and extractQuery will not be called on a null query value either (instead, the query is presumed to be unsatisfiable). Note however that null key values contained within a non-null composite item or query value are supported.

9.19.7. Examples

The core PostgreSQL distribution includes the GIN operator classes previously shown in gin-builtin-opclasses-table. The following contrib modules also contain GIN operator classes:

1. B-tree equivalent functionality for several data types

2. Module for storing (key, value) pairs

3. Enhanced support for int[]

4. Text similarity using trigram matching