Index Access Method indexam Index Access Method This chapter defines the interface between the core PostgreSQL system and index access methods, which manage individual index types. The core system knows nothing about indexes beyond what is specified here, so it is possible to develop entirely new index types by writing add-on code. All indexes in PostgreSQL are what are known technically as secondary indexes; that is, the index is physically separate from the table file that it describes. Each index is stored as its own physical relation and so is described by an entry in the pg_class catalog. The contents of an index are entirely under the control of its index access method. In practice, all index access methods divide indexes into standard-size pages so that they can use the regular storage manager and buffer manager to access the index contents. (All the existing index access methods furthermore use the standard page layout described in , and most use the same format for index tuple headers; but these decisions are not forced on an access method.) An index is effectively a mapping from some data key values to tuple identifiers, or TIDs, of row versions (tuples) in the index's parent table. A TID consists of a block number and an item number within that block (see ). This is sufficient information to fetch a particular row version from the table. Indexes are not directly aware that under MVCC, there might be multiple extant versions of the same logical row; to an index, each tuple is an independent object that needs its own index entry. Thus, an update of a row always creates all-new index entries for the row, even if the key values did not change. (HOT tuples are an exception to this statement; but indexes do not deal with those, either.) Index entries for dead tuples are reclaimed (by vacuuming) when the dead tuples themselves are reclaimed. Each index access method is described by a row in the pg_am system catalog. The pg_am entry specifies a name and a handler function for the index access method. These entries can be created and deleted using the and SQL commands. An index access method handler function must be declared to accept a single argument of type internal and to return the pseudo-type index_am_handler. The argument is a dummy value that simply serves to prevent handler functions from being called directly from SQL commands. The result of the function must be a palloc'd struct of type IndexAmRoutine, which contains everything that the core code needs to know to make use of the index access method. The IndexAmRoutine struct, also called the access method's API struct, includes fields specifying assorted fixed properties of the access method, such as whether it can support multicolumn indexes. More importantly, it contains pointers to support functions for the access method, which do all of the real work to access indexes. These support functions are plain C functions and are not visible or callable at the SQL level. The support functions are described in . The structure IndexAmRoutine is defined thus: typedef struct IndexAmRoutine { NodeTag type; /* * Total number of strategies (operators) by which we can traverse/search * this AM. Zero if AM does not have a fixed set of strategy assignments. */ uint16 amstrategies; /* total number of support functions that this AM uses */ uint16 amsupport; /* opclass options support function number or 0 */ uint16 amoptsprocnum; /* does AM support ORDER BY indexed column's value? */ bool amcanorder; /* does AM support ORDER BY result of an operator on indexed column? */ bool amcanorderbyop; /* does AM support backward scanning? */ bool amcanbackward; /* does AM support UNIQUE indexes? */ bool amcanunique; /* does AM support multi-column indexes? */ bool amcanmulticol; /* does AM require scans to have a constraint on the first index column? */ bool amoptionalkey; /* does AM handle ScalarArrayOpExpr quals? */ bool amsearcharray; /* does AM handle IS NULL/IS NOT NULL quals? */ bool amsearchnulls; /* can index storage data type differ from column data type? */ bool amstorage; /* can an index of this type be clustered on? */ bool amclusterable; /* does AM handle predicate locks? */ bool ampredlocks; /* does AM support parallel scan? */ bool amcanparallel; /* does AM support columns included with clause INCLUDE? */ bool amcaninclude; /* does AM use maintenance_work_mem? */ bool amusemaintenanceworkmem; /* does AM summarize tuples, with at least all tuples in the block * summarized in one summary */ bool amsummarizing; /* OR of parallel vacuum flags */ uint8 amparallelvacuumoptions; /* type of data stored in index, or InvalidOid if variable */ Oid amkeytype; /* interface functions */ ambuild_function ambuild; ambuildempty_function ambuildempty; aminsert_function aminsert; ambulkdelete_function ambulkdelete; amvacuumcleanup_function amvacuumcleanup; amcanreturn_function amcanreturn; /* can be NULL */ amcostestimate_function amcostestimate; amoptions_function amoptions; amproperty_function amproperty; /* can be NULL */ ambuildphasename_function ambuildphasename; /* can be NULL */ amvalidate_function amvalidate; amadjustmembers_function amadjustmembers; /* can be NULL */ ambeginscan_function ambeginscan; amrescan_function amrescan; amgettuple_function amgettuple; /* can be NULL */ amgetbitmap_function amgetbitmap; /* can be NULL */ amendscan_function amendscan; ammarkpos_function ammarkpos; /* can be NULL */ amrestrpos_function amrestrpos; /* can be NULL */ /* interface functions to support parallel index scans */ amestimateparallelscan_function amestimateparallelscan; /* can be NULL */ aminitparallelscan_function aminitparallelscan; /* can be NULL */ amparallelrescan_function amparallelrescan; /* can be NULL */ } IndexAmRoutine; To be useful, an index access method must also have one or more operator families and operator classes defined in pg_opfamily, pg_opclass, pg_amop, and pg_amproc. These entries allow the planner to determine what kinds of query qualifications can be used with indexes of this access method. Operator families and classes are described in , which is prerequisite material for reading this chapter. An individual index is defined by a pg_class entry that describes it as a physical relation, plus a pg_index entry that shows the logical content of the index — that is, the set of index columns it has and the semantics of those columns, as captured by the associated operator classes. The index columns (key values) can be either simple columns of the underlying table or expressions over the table rows. The index access method normally has no interest in where the index key values come from (it is always handed precomputed key values) but it will be very interested in the operator class information in pg_index. Both of these catalog entries can be accessed as part of the Relation data structure that is passed to all operations on the index. Some of the flag fields of IndexAmRoutine have nonobvious implications. The requirements of amcanunique are discussed in . The amcanmulticol flag asserts that the access method supports multi-key-column indexes, while amoptionalkey asserts that it allows scans where no indexable restriction clause is given for the first index column. When amcanmulticol is false, amoptionalkey essentially says whether the access method supports full-index scans without any restriction clause. Access methods that support multiple index columns must support scans that omit restrictions on any or all of the columns after the first; however they are permitted to require some restriction to appear for the first index column, and this is signaled by setting amoptionalkey false. One reason that an index AM might set amoptionalkey false is if it doesn't index null values. Since most indexable operators are strict and hence cannot return true for null inputs, it is at first sight attractive to not store index entries for null values: they could never be returned by an index scan anyway. However, this argument fails when an index scan has no restriction clause for a given index column. In practice this means that indexes that have amoptionalkey true must index nulls, since the planner might decide to use such an index with no scan keys at all. A related restriction is that an index access method that supports multiple index columns must support indexing null values in columns after the first, because the planner will assume the index can be used for queries that do not restrict these columns. For example, consider an index on (a,b) and a query with WHERE a = 4. The system will assume the index can be used to scan for rows with a = 4, which is wrong if the index omits rows where b is null. It is, however, OK to omit rows where the first indexed column is null. An index access method that does index nulls may also set amsearchnulls, indicating that it supports IS NULL and IS NOT NULL clauses as search conditions. The amcaninclude flag indicates whether the access method supports included columns, that is it can store (without processing) additional columns beyond the key column(s). The requirements of the preceding paragraph apply only to the key columns. In particular, the combination of amcanmulticol=false and amcaninclude=true is sensible: it means that there can only be one key column, but there can also be included column(s). Also, included columns must be allowed to be null, independently of amoptionalkey. The amsummarizing flag indicates whether the access method summarizes the indexed tuples, with summarizing granularity of at least per block. Access methods that do not point to individual tuples, but to block ranges (like BRIN), may allow the HOT optimization to continue. This does not apply to attributes referenced in index predicates, an update of such an attribute always disables HOT. The index construction and maintenance functions that an index access method must provide in IndexAmRoutine are: IndexBuildResult * ambuild (Relation heapRelation, Relation indexRelation, IndexInfo *indexInfo); Build a new index. The index relation has been physically created, but is empty. It must be filled in with whatever fixed data the access method requires, plus entries for all tuples already existing in the table. Ordinarily the ambuild function will call table_index_build_scan() to scan the table for existing tuples and compute the keys that need to be inserted into the index. The function must return a palloc'd struct containing statistics about the new index. void ambuildempty (Relation indexRelation); Build an empty index, and write it to the initialization fork (INIT_FORKNUM) of the given relation. This method is called only for unlogged indexes; the empty index written to the initialization fork will be copied over the main relation fork on each server restart. bool aminsert (Relation indexRelation, Datum *values, bool *isnull, ItemPointer heap_tid, Relation heapRelation, IndexUniqueCheck checkUnique, bool indexUnchanged, IndexInfo *indexInfo); Insert a new tuple into an existing index. The values and isnull arrays give the key values to be indexed, and heap_tid is the TID to be indexed. If the access method supports unique indexes (its amcanunique flag is true) then checkUnique indicates the type of uniqueness check to perform. This varies depending on whether the unique constraint is deferrable; see for details. Normally the access method only needs the heapRelation parameter when performing uniqueness checking (since then it will have to look into the heap to verify tuple liveness). The indexUnchanged Boolean value gives a hint about the nature of the tuple to be indexed. When it is true, the tuple is a duplicate of some existing tuple in the index. The new tuple is a logically unchanged successor MVCC tuple version. This happens when an UPDATE takes place that does not modify any columns covered by the index, but nevertheless requires a new version in the index. The index AM may use this hint to decide to apply bottom-up index deletion in parts of the index where many versions of the same logical row accumulate. Note that updating a non-key column or a column that only appears in a partial index predicate does not affect the value of indexUnchanged. The core code determines each tuple's indexUnchanged value using a low overhead approach that allows both false positives and false negatives. Index AMs must not treat indexUnchanged as an authoritative source of information about tuple visibility or versioning. The function's Boolean result value is significant only when checkUnique is UNIQUE_CHECK_PARTIAL. In this case a true result means the new entry is known unique, whereas false means it might be non-unique (and a deferred uniqueness check must be scheduled). For other cases a constant false result is recommended. Some indexes might not index all tuples. If the tuple is not to be indexed, aminsert should just return without doing anything. If the index AM wishes to cache data across successive index insertions within an SQL statement, it can allocate space in indexInfo->ii_Context and store a pointer to the data in indexInfo->ii_AmCache (which will be NULL initially). IndexBulkDeleteResult * ambulkdelete (IndexVacuumInfo *info, IndexBulkDeleteResult *stats, IndexBulkDeleteCallback callback, void *callback_state); Delete tuple(s) from the index. This is a bulk delete operation that is intended to be implemented by scanning the whole index and checking each entry to see if it should be deleted. The passed-in callback function must be called, in the style callback(TID, callback_state) returns bool, to determine whether any particular index entry, as identified by its referenced TID, is to be deleted. Must return either NULL or a palloc'd struct containing statistics about the effects of the deletion operation. It is OK to return NULL if no information needs to be passed on to amvacuumcleanup. Because of limited maintenance_work_mem, ambulkdelete might need to be called more than once when many tuples are to be deleted. The stats argument is the result of the previous call for this index (it is NULL for the first call within a VACUUM operation). This allows the AM to accumulate statistics across the whole operation. Typically, ambulkdelete will modify and return the same struct if the passed stats is not null. IndexBulkDeleteResult * amvacuumcleanup (IndexVacuumInfo *info, IndexBulkDeleteResult *stats); Clean up after a VACUUM operation (zero or more ambulkdelete calls). This does not have to do anything beyond returning index statistics, but it might perform bulk cleanup such as reclaiming empty index pages. stats is whatever the last ambulkdelete call returned, or NULL if ambulkdelete was not called because no tuples needed to be deleted. If the result is not NULL it must be a palloc'd struct. The statistics it contains will be used to update pg_class, and will be reported by VACUUM if VERBOSE is given. It is OK to return NULL if the index was not changed at all during the VACUUM operation, but otherwise correct stats should be returned. amvacuumcleanup will also be called at completion of an ANALYZE operation. In this case stats is always NULL and any return value will be ignored. This case can be distinguished by checking info->analyze_only. It is recommended that the access method do nothing except post-insert cleanup in such a call, and that only in an autovacuum worker process. bool amcanreturn (Relation indexRelation, int attno); Check whether the index can support index-only scans on the given column, by returning the column's original indexed value. The attribute number is 1-based, i.e., the first column's attno is 1. Returns true if supported, else false. This function should always return true for included columns (if those are supported), since there's little point in an included column that can't be retrieved. If the access method does not support index-only scans at all, the amcanreturn field in its IndexAmRoutine struct can be set to NULL. void amcostestimate (PlannerInfo *root, IndexPath *path, double loop_count, Cost *indexStartupCost, Cost *indexTotalCost, Selectivity *indexSelectivity, double *indexCorrelation, double *indexPages); Estimate the costs of an index scan. This function is described fully in , below. bytea * amoptions (ArrayType *reloptions, bool validate); Parse and validate the reloptions array for an index. This is called only when a non-null reloptions array exists for the index. reloptions is a text array containing entries of the form name=value. The function should construct a bytea value, which will be copied into the rd_options field of the index's relcache entry. The data contents of the bytea value are open for the access method to define; most of the standard access methods use struct StdRdOptions. When validate is true, the function should report a suitable error message if any of the options are unrecognized or have invalid values; when validate is false, invalid entries should be silently ignored. (validate is false when loading options already stored in pg_catalog; an invalid entry could only be found if the access method has changed its rules for options, and in that case ignoring obsolete entries is appropriate.) It is OK to return NULL if default behavior is wanted. bool amproperty (Oid index_oid, int attno, IndexAMProperty prop, const char *propname, bool *res, bool *isnull); The amproperty method allows index access methods to override the default behavior of pg_index_column_has_property and related functions. If the access method does not have any special behavior for index property inquiries, the amproperty field in its IndexAmRoutine struct can be set to NULL. Otherwise, the amproperty method will be called with index_oid and attno both zero for pg_indexam_has_property calls, or with index_oid valid and attno zero for pg_index_has_property calls, or with index_oid valid and attno greater than zero for pg_index_column_has_property calls. prop is an enum value identifying the property being tested, while propname is the original property name string. If the core code does not recognize the property name then prop is AMPROP_UNKNOWN. Access methods can define custom property names by checking propname for a match (use pg_strcasecmp to match, for consistency with the core code); for names known to the core code, it's better to inspect prop. If the amproperty method returns true then it has determined the property test result: it must set *res to the Boolean value to return, or set *isnull to true to return a NULL. (Both of the referenced variables are initialized to false before the call.) If the amproperty method returns false then the core code will proceed with its normal logic for determining the property test result. Access methods that support ordering operators should implement AMPROP_DISTANCE_ORDERABLE property testing, as the core code does not know how to do that and will return NULL. It may also be advantageous to implement AMPROP_RETURNABLE testing, if that can be done more cheaply than by opening the index and calling amcanreturn, which is the core code's default behavior. The default behavior should be satisfactory for all other standard properties. char * ambuildphasename (int64 phasenum); Return the textual name of the given build phase number. The phase numbers are those reported during an index build via the pgstat_progress_update_param interface. The phase names are then exposed in the pg_stat_progress_create_index view. bool amvalidate (Oid opclassoid); Validate the catalog entries for the specified operator class, so far as the access method can reasonably do that. For example, this might include testing that all required support functions are provided. The amvalidate function must return false if the opclass is invalid. Problems should be reported with ereport messages, typically at INFO level. void amadjustmembers (Oid opfamilyoid, Oid opclassoid, List *operators, List *functions); Validate proposed new operator and function members of an operator family, so far as the access method can reasonably do that, and set their dependency types if the default is not satisfactory. This is called during CREATE OPERATOR CLASS and during ALTER OPERATOR FAMILY ADD; in the latter case opclassoid is InvalidOid. The List arguments are lists of OpFamilyMember structs, as defined in amapi.h. Tests done by this function will typically be a subset of those performed by amvalidate, since amadjustmembers cannot assume that it is seeing a complete set of members. For example, it would be reasonable to check the signature of a support function, but not to check whether all required support functions are provided. Any problems can be reported by throwing an error. The dependency-related fields of the OpFamilyMember structs are initialized by the core code to create hard dependencies on the opclass if this is CREATE OPERATOR CLASS, or soft dependencies on the opfamily if this is ALTER OPERATOR FAMILY ADD. amadjustmembers can adjust these fields if some other behavior is more appropriate. For example, GIN, GiST, and SP-GiST always set operator members to have soft dependencies on the opfamily, since the connection between an operator and an opclass is relatively weak in these index types; so it is reasonable to allow operator members to be added and removed freely. Optional support functions are typically also given soft dependencies, so that they can be removed if necessary. The purpose of an index, of course, is to support scans for tuples matching an indexable WHERE condition, often called a qualifier or scan key. The semantics of index scanning are described more fully in , below. An index access method can support plain index scans, bitmap index scans, or both. The scan-related functions that an index access method must or may provide are: IndexScanDesc ambeginscan (Relation indexRelation, int nkeys, int norderbys); Prepare for an index scan. The nkeys and norderbys parameters indicate the number of quals and ordering operators that will be used in the scan; these may be useful for space allocation purposes. Note that the actual values of the scan keys aren't provided yet. The result must be a palloc'd struct. For implementation reasons the index access method must create this struct by calling RelationGetIndexScan(). In most cases ambeginscan does little beyond making that call and perhaps acquiring locks; the interesting parts of index-scan startup are in amrescan. void amrescan (IndexScanDesc scan, ScanKey keys, int nkeys, ScanKey orderbys, int norderbys); Start or restart an index scan, possibly with new scan keys. (To restart using previously-passed keys, NULL is passed for keys and/or orderbys.) Note that it is not allowed for the number of keys or order-by operators to be larger than what was passed to ambeginscan. In practice the restart feature is used when a new outer tuple is selected by a nested-loop join and so a new key comparison value is needed, but the scan key structure remains the same. bool amgettuple (IndexScanDesc scan, ScanDirection direction); Fetch the next tuple in the given scan, moving in the given direction (forward or backward in the index). Returns true if a tuple was obtained, false if no matching tuples remain. In the true case the tuple TID is stored into the scan structure. Note that success means only that the index contains an entry that matches the scan keys, not that the tuple necessarily still exists in the heap or will pass the caller's snapshot test. On success, amgettuple must also set scan->xs_recheck to true or false. False means it is certain that the index entry matches the scan keys. True means this is not certain, and the conditions represented by the scan keys must be rechecked against the heap tuple after fetching it. This provision supports lossy index operators. Note that rechecking will extend only to the scan conditions; a partial index predicate (if any) is never rechecked by amgettuple callers. If the index supports index-only scans (i.e., amcanreturn returns true for any of its columns), then on success the AM must also check scan->xs_want_itup, and if that is true it must return the originally indexed data for the index entry. Columns for which amcanreturn returns false can be returned as nulls. The data can be returned in the form of an IndexTuple pointer stored at scan->xs_itup, with tuple descriptor scan->xs_itupdesc; or in the form of a HeapTuple pointer stored at scan->xs_hitup, with tuple descriptor scan->xs_hitupdesc. (The latter format should be used when reconstructing data that might possibly not fit into an IndexTuple.) In either case, management of the data referenced by the pointer is the access method's responsibility. The data must remain good at least until the next amgettuple, amrescan, or amendscan call for the scan. The amgettuple function need only be provided if the access method supports plain index scans. If it doesn't, the amgettuple field in its IndexAmRoutine struct must be set to NULL. int64 amgetbitmap (IndexScanDesc scan, TIDBitmap *tbm); Fetch all tuples in the given scan and add them to the caller-supplied TIDBitmap (that is, OR the set of tuple IDs into whatever set is already in the bitmap). The number of tuples fetched is returned (this might be just an approximate count, for instance some AMs do not detect duplicates). While inserting tuple IDs into the bitmap, amgetbitmap can indicate that rechecking of the scan conditions is required for specific tuple IDs. This is analogous to the xs_recheck output parameter of amgettuple. Note: in the current implementation, support for this feature is conflated with support for lossy storage of the bitmap itself, and therefore callers recheck both the scan conditions and the partial index predicate (if any) for recheckable tuples. That might not always be true, however. amgetbitmap and amgettuple cannot be used in the same index scan; there are other restrictions too when using amgetbitmap, as explained in . The amgetbitmap function need only be provided if the access method supports bitmap index scans. If it doesn't, the amgetbitmap field in its IndexAmRoutine struct must be set to NULL. void amendscan (IndexScanDesc scan); End a scan and release resources. The scan struct itself should not be freed, but any locks or pins taken internally by the access method must be released, as well as any other memory allocated by ambeginscan and other scan-related functions. void ammarkpos (IndexScanDesc scan); Mark current scan position. The access method need only support one remembered scan position per scan. The ammarkpos function need only be provided if the access method supports ordered scans. If it doesn't, the ammarkpos field in its IndexAmRoutine struct may be set to NULL. void amrestrpos (IndexScanDesc scan); Restore the scan to the most recently marked position. The amrestrpos function need only be provided if the access method supports ordered scans. If it doesn't, the amrestrpos field in its IndexAmRoutine struct may be set to NULL. In addition to supporting ordinary index scans, some types of index may wish to support parallel index scans, which allow multiple backends to cooperate in performing an index scan. The index access method should arrange things so that each cooperating process returns a subset of the tuples that would be performed by an ordinary, non-parallel index scan, but in such a way that the union of those subsets is equal to the set of tuples that would be returned by an ordinary, non-parallel index scan. Furthermore, while there need not be any global ordering of tuples returned by a parallel scan, the ordering of that subset of tuples returned within each cooperating backend must match the requested ordering. The following functions may be implemented to support parallel index scans: Size amestimateparallelscan (void); Estimate and return the number of bytes of dynamic shared memory which the access method will be needed to perform a parallel scan. (This number is in addition to, not in lieu of, the amount of space needed for AM-independent data in ParallelIndexScanDescData.) It is not necessary to implement this function for access methods which do not support parallel scans or for which the number of additional bytes of storage required is zero. void aminitparallelscan (void *target); This function will be called to initialize dynamic shared memory at the beginning of a parallel scan. target will point to at least the number of bytes previously returned by amestimateparallelscan, and this function may use that amount of space to store whatever data it wishes. It is not necessary to implement this function for access methods which do not support parallel scans or in cases where the shared memory space required needs no initialization. void amparallelrescan (IndexScanDesc scan); This function, if implemented, will be called when a parallel index scan must be restarted. It should reset any shared state set up by aminitparallelscan such that the scan will be restarted from the beginning. In an index scan, the index access method is responsible for regurgitating the TIDs of all the tuples it has been told about that match the scan keys. The access method is not involved in actually fetching those tuples from the index's parent table, nor in determining whether they pass the scan's visibility test or other conditions. A scan key is the internal representation of a WHERE clause of the form index_key operator constant, where the index key is one of the columns of the index and the operator is one of the members of the operator family associated with that index column. An index scan has zero or more scan keys, which are implicitly ANDed — the returned tuples are expected to satisfy all the indicated conditions. The access method can report that the index is lossy, or requires rechecks, for a particular query. This implies that the index scan will return all the entries that pass the scan key, plus possibly additional entries that do not. The core system's index-scan machinery will then apply the index conditions again to the heap tuple to verify whether or not it really should be selected. If the recheck option is not specified, the index scan must return exactly the set of matching entries. Note that it is entirely up to the access method to ensure that it correctly finds all and only the entries passing all the given scan keys. Also, the core system will simply hand off all the WHERE clauses that match the index keys and operator families, without any semantic analysis to determine whether they are redundant or contradictory. As an example, given WHERE x > 4 AND x > 14 where x is a b-tree indexed column, it is left to the b-tree amrescan function to realize that the first scan key is redundant and can be discarded. The extent of preprocessing needed during amrescan will depend on the extent to which the index access method needs to reduce the scan keys to a normalized form. Some access methods return index entries in a well-defined order, others do not. There are actually two different ways that an access method can support sorted output: Access methods that always return entries in the natural ordering of their data (such as btree) should set amcanorder to true. Currently, such access methods must use btree-compatible strategy numbers for their equality and ordering operators. Access methods that support ordering operators should set amcanorderbyop to true. This indicates that the index is capable of returning entries in an order satisfying ORDER BY index_key operator constant. Scan modifiers of that form can be passed to amrescan as described previously. The amgettuple function has a direction argument, which can be either ForwardScanDirection (the normal case) or BackwardScanDirection. If the first call after amrescan specifies BackwardScanDirection, then the set of matching index entries is to be scanned back-to-front rather than in the normal front-to-back direction, so amgettuple must return the last matching tuple in the index, rather than the first one as it normally would. (This will only occur for access methods that set amcanorder to true.) After the first call, amgettuple must be prepared to advance the scan in either direction from the most recently returned entry. (But if amcanbackward is false, all subsequent calls will have the same direction as the first one.) Access methods that support ordered scans must support marking a position in a scan and later returning to the marked position. The same position might be restored multiple times. However, only one position need be remembered per scan; a new ammarkpos call overrides the previously marked position. An access method that does not support ordered scans need not provide ammarkpos and amrestrpos functions in IndexAmRoutine; set those pointers to NULL instead. Both the scan position and the mark position (if any) must be maintained consistently in the face of concurrent insertions or deletions in the index. It is OK if a freshly-inserted entry is not returned by a scan that would have found the entry if it had existed when the scan started, or for the scan to return such an entry upon rescanning or backing up even though it had not been returned the first time through. Similarly, a concurrent delete might or might not be reflected in the results of a scan. What is important is that insertions or deletions not cause the scan to miss or multiply return entries that were not themselves being inserted or deleted. If the index stores the original indexed data values (and not some lossy representation of them), it is useful to support index-only scans, in which the index returns the actual data not just the TID of the heap tuple. This will only avoid I/O if the visibility map shows that the TID is on an all-visible page; else the heap tuple must be visited anyway to check MVCC visibility. But that is no concern of the access method's. Instead of using amgettuple, an index scan can be done with amgetbitmap to fetch all tuples in one call. This can be noticeably more efficient than amgettuple because it allows avoiding lock/unlock cycles within the access method. In principle amgetbitmap should have the same effects as repeated amgettuple calls, but we impose several restrictions to simplify matters. First of all, amgetbitmap returns all tuples at once and marking or restoring scan positions isn't supported. Secondly, the tuples are returned in a bitmap which doesn't have any specific ordering, which is why amgetbitmap doesn't take a direction argument. (Ordering operators will never be supplied for such a scan, either.) Also, there is no provision for index-only scans with amgetbitmap, since there is no way to return the contents of index tuples. Finally, amgetbitmap does not guarantee any locking of the returned tuples, with implications spelled out in . Note that it is permitted for an access method to implement only amgetbitmap and not amgettuple, or vice versa, if its internal implementation is unsuited to one API or the other. Index access methods must handle concurrent updates of the index by multiple processes. The core PostgreSQL system obtains AccessShareLock on the index during an index scan, and RowExclusiveLock when updating the index (including plain VACUUM). Since these lock types do not conflict, the access method is responsible for handling any fine-grained locking it might need. An ACCESS EXCLUSIVE lock on the index as a whole will be taken only during index creation, destruction, or REINDEX (SHARE UPDATE EXCLUSIVE is taken instead with CONCURRENTLY). Building an index type that supports concurrent updates usually requires extensive and subtle analysis of the required behavior. For the b-tree and hash index types, you can read about the design decisions involved in src/backend/access/nbtree/README and src/backend/access/hash/README. Aside from the index's own internal consistency requirements, concurrent updates create issues about consistency between the parent table (the heap) and the index. Because PostgreSQL separates accesses and updates of the heap from those of the index, there are windows in which the index might be inconsistent with the heap. We handle this problem with the following rules: A new heap entry is made before making its index entries. (Therefore a concurrent index scan is likely to fail to see the heap entry. This is okay because the index reader would be uninterested in an uncommitted row anyway. But see .) When a heap entry is to be deleted (by VACUUM), all its index entries must be removed first. An index scan must maintain a pin on the index page holding the item last returned by amgettuple, and ambulkdelete cannot delete entries from pages that are pinned by other backends. The need for this rule is explained below. Without the third rule, it is possible for an index reader to see an index entry just before it is removed by VACUUM, and then to arrive at the corresponding heap entry after that was removed by VACUUM. This creates no serious problems if that item number is still unused when the reader reaches it, since an empty item slot will be ignored by heap_fetch(). But what if a third backend has already re-used the item slot for something else? When using an MVCC-compliant snapshot, there is no problem because the new occupant of the slot is certain to be too new to pass the snapshot test. However, with a non-MVCC-compliant snapshot (such as SnapshotAny), it would be possible to accept and return a row that does not in fact match the scan keys. We could defend against this scenario by requiring the scan keys to be rechecked against the heap row in all cases, but that is too expensive. Instead, we use a pin on an index page as a proxy to indicate that the reader might still be in flight from the index entry to the matching heap entry. Making ambulkdelete block on such a pin ensures that VACUUM cannot delete the heap entry before the reader is done with it. This solution costs little in run time, and adds blocking overhead only in the rare cases where there actually is a conflict. This solution requires that index scans be synchronous: we have to fetch each heap tuple immediately after scanning the corresponding index entry. This is expensive for a number of reasons. An asynchronous scan in which we collect many TIDs from the index, and only visit the heap tuples sometime later, requires much less index locking overhead and can allow a more efficient heap access pattern. Per the above analysis, we must use the synchronous approach for non-MVCC-compliant snapshots, but an asynchronous scan is workable for a query using an MVCC snapshot. In an amgetbitmap index scan, the access method does not keep an index pin on any of the returned tuples. Therefore it is only safe to use such scans with MVCC-compliant snapshots. When the ampredlocks flag is not set, any scan using that index access method within a serializable transaction will acquire a nonblocking predicate lock on the full index. This will generate a read-write conflict with the insert of any tuple into that index by a concurrent serializable transaction. If certain patterns of read-write conflicts are detected among a set of concurrent serializable transactions, one of those transactions may be canceled to protect data integrity. When the flag is set, it indicates that the index access method implements finer-grained predicate locking, which will tend to reduce the frequency of such transaction cancellations. PostgreSQL enforces SQL uniqueness constraints using unique indexes, which are indexes that disallow multiple entries with identical keys. An access method that supports this feature sets amcanunique true. (At present, only b-tree supports it.) Columns listed in the INCLUDE clause are not considered when enforcing uniqueness. Because of MVCC, it is always necessary to allow duplicate entries to exist physically in an index: the entries might refer to successive versions of a single logical row. The behavior we actually want to enforce is that no MVCC snapshot could include two rows with equal index keys. This breaks down into the following cases that must be checked when inserting a new row into a unique index: If a conflicting valid row has been deleted by the current transaction, it's okay. (In particular, since an UPDATE always deletes the old row version before inserting the new version, this will allow an UPDATE on a row without changing the key.) If a conflicting row has been inserted by an as-yet-uncommitted transaction, the would-be inserter must wait to see if that transaction commits. If it rolls back then there is no conflict. If it commits without deleting the conflicting row again, there is a uniqueness violation. (In practice we just wait for the other transaction to end and then redo the visibility check in toto.) Similarly, if a conflicting valid row has been deleted by an as-yet-uncommitted transaction, the would-be inserter must wait for that transaction to commit or abort, and then repeat the test. Furthermore, immediately before reporting a uniqueness violation according to the above rules, the access method must recheck the liveness of the row being inserted. If it is committed dead then no violation should be reported. (This case cannot occur during the ordinary scenario of inserting a row that's just been created by the current transaction. It can happen during CREATE UNIQUE INDEX CONCURRENTLY, however.) We require the index access method to apply these tests itself, which means that it must reach into the heap to check the commit status of any row that is shown to have a duplicate key according to the index contents. This is without a doubt ugly and non-modular, but it saves redundant work: if we did a separate probe then the index lookup for a conflicting row would be essentially repeated while finding the place to insert the new row's index entry. What's more, there is no obvious way to avoid race conditions unless the conflict check is an integral part of insertion of the new index entry. If the unique constraint is deferrable, there is additional complexity: we need to be able to insert an index entry for a new row, but defer any uniqueness-violation error until end of statement or even later. To avoid unnecessary repeat searches of the index, the index access method should do a preliminary uniqueness check during the initial insertion. If this shows that there is definitely no conflicting live tuple, we are done. Otherwise, we schedule a recheck to occur when it is time to enforce the constraint. If, at the time of the recheck, both the inserted tuple and some other tuple with the same key are live, then the error must be reported. (Note that for this purpose, live actually means any tuple in the index entry's HOT chain is live.) To implement this, the aminsert function is passed a checkUnique parameter having one of the following values: UNIQUE_CHECK_NO indicates that no uniqueness checking should be done (this is not a unique index). UNIQUE_CHECK_YES indicates that this is a non-deferrable unique index, and the uniqueness check must be done immediately, as described above. UNIQUE_CHECK_PARTIAL indicates that the unique constraint is deferrable. PostgreSQL will use this mode to insert each row's index entry. The access method must allow duplicate entries into the index, and report any potential duplicates by returning false from aminsert. For each row for which false is returned, a deferred recheck will be scheduled. The access method must identify any rows which might violate the unique constraint, but it is not an error for it to report false positives. This allows the check to be done without waiting for other transactions to finish; conflicts reported here are not treated as errors and will be rechecked later, by which time they may no longer be conflicts. UNIQUE_CHECK_EXISTING indicates that this is a deferred recheck of a row that was reported as a potential uniqueness violation. Although this is implemented by calling aminsert, the access method must not insert a new index entry in this case. The index entry is already present. Rather, the access method must check to see if there is another live index entry. If so, and if the target row is also still live, report error. It is recommended that in a UNIQUE_CHECK_EXISTING call, the access method further verify that the target row actually does have an existing entry in the index, and report error if not. This is a good idea because the index tuple values passed to aminsert will have been recomputed. If the index definition involves functions that are not really immutable, we might be checking the wrong area of the index. Checking that the target row is found in the recheck verifies that we are scanning for the same tuple values as were used in the original insertion. The amcostestimate function is given information describing a possible index scan, including lists of WHERE and ORDER BY clauses that have been determined to be usable with the index. It must return estimates of the cost of accessing the index and the selectivity of the WHERE clauses (that is, the fraction of parent-table rows that will be retrieved during the index scan). For simple cases, nearly all the work of the cost estimator can be done by calling standard routines in the optimizer; the point of having an amcostestimate function is to allow index access methods to provide index-type-specific knowledge, in case it is possible to improve on the standard estimates. Each amcostestimate function must have the signature: void amcostestimate (PlannerInfo *root, IndexPath *path, double loop_count, Cost *indexStartupCost, Cost *indexTotalCost, Selectivity *indexSelectivity, double *indexCorrelation, double *indexPages); The first three parameters are inputs: root The planner's information about the query being processed. path The index access path being considered. All fields except cost and selectivity values are valid. loop_count The number of repetitions of the index scan that should be factored into the cost estimates. This will typically be greater than one when considering a parameterized scan for use in the inside of a nestloop join. Note that the cost estimates should still be for just one scan; a larger loop_count means that it may be appropriate to allow for some caching effects across multiple scans. The last five parameters are pass-by-reference outputs: *indexStartupCost Set to cost of index start-up processing *indexTotalCost Set to total cost of index processing *indexSelectivity Set to index selectivity *indexCorrelation Set to correlation coefficient between index scan order and underlying table's order *indexPages Set to number of index leaf pages Note that cost estimate functions must be written in C, not in SQL or any available procedural language, because they must access internal data structures of the planner/optimizer. The index access costs should be computed using the parameters used by src/backend/optimizer/path/costsize.c: a sequential disk block fetch has cost seq_page_cost, a nonsequential fetch has cost random_page_cost, and the cost of processing one index row should usually be taken as cpu_index_tuple_cost. In addition, an appropriate multiple of cpu_operator_cost should be charged for any comparison operators invoked during index processing (especially evaluation of the indexquals themselves). The access costs should include all disk and CPU costs associated with scanning the index itself, but not the costs of retrieving or processing the parent-table rows that are identified by the index. The start-up cost is the part of the total scan cost that must be expended before we can begin to fetch the first row. For most indexes this can be taken as zero, but an index type with a high start-up cost might want to set it nonzero. The indexSelectivity should be set to the estimated fraction of the parent table rows that will be retrieved during the index scan. In the case of a lossy query, this will typically be higher than the fraction of rows that actually pass the given qual conditions. The indexCorrelation should be set to the correlation (ranging between -1.0 and 1.0) between the index order and the table order. This is used to adjust the estimate for the cost of fetching rows from the parent table. The indexPages should be set to the number of leaf pages. This is used to estimate the number of workers for parallel index scan. When loop_count is greater than one, the returned numbers should be averages expected for any one scan of the index. A typical cost estimator will proceed as follows: Estimate and return the fraction of parent-table rows that will be visited based on the given qual conditions. In the absence of any index-type-specific knowledge, use the standard optimizer function clauselist_selectivity(): *indexSelectivity = clauselist_selectivity(root, path->indexquals, path->indexinfo->rel->relid, JOIN_INNER, NULL); Estimate the number of index rows that will be visited during the scan. For many index types this is the same as indexSelectivity times the number of rows in the index, but it might be more. (Note that the index's size in pages and rows is available from the path->indexinfo struct.) Estimate the number of index pages that will be retrieved during the scan. This might be just indexSelectivity times the index's size in pages. Compute the index access cost. A generic estimator might do this: /* * Our generic assumption is that the index pages will be read * sequentially, so they cost seq_page_cost each, not random_page_cost. * Also, we charge for evaluation of the indexquals at each index row. * All the costs are assumed to be paid incrementally during the scan. */ cost_qual_eval(&index_qual_cost, path->indexquals, root); *indexStartupCost = index_qual_cost.startup; *indexTotalCost = seq_page_cost * numIndexPages + (cpu_index_tuple_cost + index_qual_cost.per_tuple) * numIndexTuples; However, the above does not account for amortization of index reads across repeated index scans. Estimate the index correlation. For a simple ordered index on a single field, this can be retrieved from pg_statistic. If the correlation is not known, the conservative estimate is zero (no correlation). Examples of cost estimator functions can be found in src/backend/utils/adt/selfuncs.c.