Method and apparatus for database query decomposition

Abstract

An improved system for database query processing by means of “query decomposition” intercepts database queries prior to processing by a database management system (“DBMS”). The system decomposes at least selected queries to generate multiple subqueries for application, in parallel, to the DBMS, in lieu of the intercepted query. Responses by the DBMS to the subqueries are assembled by the system to generate a final response. The system also provides improved methods and apparatus for storage and retrieval of records from a database utilizing the DBMS's cluster storage and index retrieval facilitates, in combination with a smaller-than-usual hash bucket size.

Description

REFERENCE TO APPENDICES

The disclosure of this patent document contains material which is subject to copyright protection. The owner thereof has no objection to facsimile reproduction by anyone of the patent document or the patent disclosure, as it appears in the U.S. Patent and Trademark Office patent file or records, but otherwise reserves all copyright whatsoever.

BACKGROUND OF THE INVENTION

This invention relates to digital data processing and, more particularly, to methods and apparatus for database management systems on multiprocessor digital data processing systems.

In addition to performing calculations, computers have traditionally been used to store and retrieve large amounts of data. Early computer systems were typically programmed for this on an ad hoc basis. For example, to track a company's employees, a program was typically written to handle all steps necessary to input, sort and store employee data in a computer file and, as necessary, to retrieve and collate it to generate reports. Special-purpose software packages, referred to as database management systems (or “DBMS's”), were later developed to handle all but the highest-level of these tasks.

Among the most widely used database management systems are the so-called relational systems. From an operator's perspective, these store data in two-dimensional tables. For example, each row (or record) of an employee data table might include the following columns (or fields) of information: name of an employee, his or her identification number, address, and department number.

	.	.	.	.
	.	.	.	.
	.	.	.	.
	Smith	1056	5 Oak Avenue	10
	James	1058	3 State Street	41
	Wright	1059	15 Main Street	25
	.	.	.	.
	.	.	.	.
	.	.	.	.

One or more indexes on large tables are generally provided to facilitate the most common data accesses, e.g., look-ups based on employee name.

In relational systems, corresponding rows in two or more tables are identified by matching data values in one or more columns. For example, the department name corresponding to a given employee may be identified by matching his or her department number to a row in a department data table that gives department numbers and department names. This is in contrast to hierarchical, network, and other DBMS's that use pointers instead of data values to indicate corresponding rows when tables are combined, or “joined.”

Relational DBMS's typically permit the operator to access information in the database via a query. This is a command that specifies which data fields (columns) are to be retrieved from a database table and which records (rows) those fields are to be selected from. For example, a query for the names of all employees in department 10 might be fashioned as follows:

SELECT name, department_number

FROM employee

WHERE department_number=10

There is no particular ordering of the resulting rows retrieved by the DBMS, unless the query specifies an ordering (e.g., ORDER BY name).

A query may also involve multiple tables. For example, to retrieve department names instead of numbers, the above query might be refashioned as follows:

SELECT name, department_name

FROM employee, department

WHERE department_number=10

AND employee.department_number=department.department_number

A particular relational data table need not be stored in a single computer file but, rather, can be partitioned among many files. This makes such tables particularly suited for use on multiprocessor computer systems, i.e., computer systems having multiple processors and multiple disk drives (or other storage devices) of the type disclosed in U.S. Pat. No. 5,055,999. Unfortunately, prior art DBMS's have not proven capable of taking full advantage of the power of such multiprocessing systems and, particularly, their power to simultaneously process data (in parallel) from multiple partitions on multiple storage devices with multiple central processing units.

In view of the foregoing, an object of the invention is to provide improved methods and apparatus for database management and, particularly, improved methods and apparatus for data base management capable of operating on multiprocessor systems.

A further object of the invention is to provide improved systems for database management capable of effectively accessing a relational database contained in multiple tables and multiple partitions.

A still further object is to provide improved methods and apparatus for storing and retrieving data for access by a DBMS.

These and other objects are evident in the attached drawings and the description which follows.

SUMMARY OF THE INVENTION

The foregoing and other objects are attained by the invention which provides, in one aspect, improvements to digital data processors of the type having a database management system (DBMS) that accesses data records stored in a database table contained among plural independently accessible partitions (e.g., data partitions contained on separate disk drives), where that DBMS has a standard interface for processing queries to access those data records.

The improvement is characterized by a parallel interface that intercepts selected queries prior to substantive processing by the standard interface. The standard interface is often called the “server” interface; it is accessed by clients that are the source of queries. A decomposition element within the parallel interface generates multiple subqueries from the intercepted query. Those subqueries, each representing a request for access to data stored in a respective partition of the table, are applied in parallel to the standard interface in lieu of the intercepted query. Responses by the DBMS to the subqueries are reassembled to generate a final response representing the response the DBMS would have generated to the intercepted query signal itself. Such reassembly can include interleaving the data contained in the responses (e.g., to create a single sorted list) or applying an aggregate function (e.g., sum or average) to that data.

According to a further aspect of the invention, the decomposition element generates the subqueries to be substantially identical to the intercepted signal but including an “intersecting predicate” (i.e., additional query conditions) that evaluates true for all data records in respective partitions of said database table and false for all others. This can be, for example, a logically AND'ed condition that evaluates true for records in the respective partition. Continuing the first example above, assuming that the employee database is partitioned randomly across multiple partitions, a subquery for the first partition could be generated as follows (where rowid has three parts, the last of which indicates the partition number):

SELECT name, department_number

FROM employee

WHERE department_number=10 AND

employee.rowid>=0.0.1 AND

employee.rowid<0.0.2

In another aspect, the invention contemplates a further improvement to a digital data processing system of the type described above, wherein the DBMS responds to selected queries for accessing data records joined from one or more of database tables, and wherein the DBMS includes an optimizer for determining an optimal strategy for applying such queries to the tables. The improvement of this aspect is characterized by an element for identifying, from output of the optimizer, a driving table whose partitions will be targeted by subqueries generated in responding to an intercepted query. The improvement is further characterized by generating the subqueries to include, in addition to the predicate list of the intercepted query, an intersecting predicate for all data records in respective partitions of the driving database table. Those skilled in the art will appreciate that tables referenced in the query other than the driving table need not be identically partitioned to the driving table, nor co-located with its partitions on storage devices. Tables may be accessed through either full-table scans or indexed scans, i.e., whether the DBMS searches all blocks of the relevant partition or only those indicated by a relevant index.

According to another aspect, the invention provides an improvement to a digital data processing system of the type described, wherein the DBMS's standard interface is invoked by a procedure or function call. The improvement is characterized by functionality for invoking the parallel interface in lieu of the client-side portion of the standard interface in response to such a procedure/function call. And, by responding to a query for generating plural subqueries in the form of further procedures/functions to the standard server interface. The parallel interface can form part of an object code library for linking with a computer program including procedures/function calls for invoking the DBMS.

In still another aspect, the invention contemplates an improvement to a digital data processing system as described above, wherein the standard interface normally responds to insert/select queries by placing requested data from the database table means in a further database table (i.e., as opposed to merely printing the requested data or otherwise outputting it in text form or merely returning the data to the requesting program). The improvement of this aspect is characterized by generating the plural subqueries so as to cause the DBMS to place the data requested from each respective partition in the designated database table.

In yet another aspect of the invention, a digital data processing system as described above can include functionality for executing multiple threads, or “lightweight processes,” each for applying a respective subquery signal to the DBMS's interface element. Those threads can be executed in parallel on multiple central processing units, and can be serviced by multiple server processes within the DBMS that also execute in parallel.

Further aspects of the invention provide improvements to a digital data processing system of the type having a storage element (e.g., a disk drive or other random-access media) for storing and retrieving data records, as well as a DBMS having (i) a hashing element to effect storage of data records in “hash bucket” regions in the storage element, where each record is stored in a root hash bucket region corresponding to a hash function of a selected value of the data record or, alternatively, to effect storage of data records in an overflow hash bucket region associated with that root hash bucket region; and (2) an indexing element to index each stored data record for direct access in accord with a respective value of that data record.

The improvement is characterized by a scatter cluster retrieval element that responds to a request for accessing a data record previously stored via the hashing element, by invoking the indexing element to retrieve that record in accord with the index value thereof, where stored records have previously been indexed by the indexing element with respect to the same fields (columns) used by the hashing element. In a related aspect of the invention, the hashing element stores the data records in hash bucket regions that are sized so as to create at least one overflow hash bucket region per root bucket region, and such that overflow bucket regions for a given root bucket region are distributed roughly evenly across different storage partitions.

Another aspect of the invention provides a digital data processing system of the type described above, in which plural subcursor buffers are associated with each subquery signal for storing results generated by the DBMS's standard interface means in response to that subquery signal. To assemble all results of those subqueries, a root buffer stores a then-current result, while a fetching element simultaneously assembles a final result signal based upon those results currently stored in selected subcursor buffers. As results are taken from each of those buffers, they are emptied. For each such emptied buffer, a subquery is applied to the standard interface asynchronously with respect to demand for that buffer's contents in assembling the final result. In the case of queries involving aggregates, the root buffer stores then-current results in a temporary table to be queried later by an aggregate query generated by the decomposition element.

In still other aspects, the invention provides a method for digital data processing paralleling the operation of the digital data processing system described above; i.e., “transparent” to the DBMS client other than by improved performance.

BRIEF DESCRIPTION OF THE DRAWINGS

A better appreciation of the invention may be attained by reference to the drawings, in which

FIG. 1 depicts a preferred multiprocessing system used to practice the invention.

FIG. 2 illustrates in greater detail processing cells and their interconnection within the processing system of FIG. 1 .

FIG. 3A depicts a standard arrangement of processes and software modules utilized in digital data processor 10 without query-decomposition and data access according to the invention.

FIG. 3B depicts a preferred arrangement of threads, processes and software modules utilized in digital data processor 10 for query decomposition and data access according to the invention.

FIG. 4 shows the operation of assembler 74 B on results generated by the DBMS 76 and threads 78 A, 78 B, 78 C in response to the subquery signals.

FIG. 5 depicts a preferred mechanism, referred to as “scatter clustering,” for storing and retrieving data from database 72 .

FIGS. 6-7 are used in connection with the discussion of the operation and use of a preferred query decomposition system according to the invention.

FIGS. 8-10 are used in connection with the discussion of design provided in Database Note # 26 .

FIGS. 11-13 are used in connection with the discussion of query decomposition for applications running on client workstations in Database Note # 61 .

FIGS. 14-16 are used in connection with the discussion of the framework of rules for automating query decomposition in Database Note # 32 .

FIGS. 17-23 are used in connection with the discussion of parallel cursor building blocks in Database Note # 36 .

FIGS. 24-25 are used in connection with the discussion of parse tree requirements for query decomposition in Database Note # 37 .

FIGS. 26-27 are used in connection with the discussion of query decomposition control structures in Database Notes # 41 .

FIGS. 28-30 used in connection with the discussion of upper tree parallelism in parallel cursors in Database Note # 42 .

DETAILED DESCRIPTION OF THE ILLUSTRATED EMBODIMENT

FIG. 1 depicts a preferred multiprocessing system used to practice the invention. The illustrated system 10 includes three information transfer levels: level: 0 , level: 1 , and level: 2 . Each information transfer level includes one or more level segments, characterized by a bus element and a plurality of interface elements. Particularly, level: 0 of the illustrated system 10 includes six segments, designated 12 A, 12 B, 12 C, 12 D, 12 E and 12 F, respectively. Similarly, level: 1 includes segments 14 A and 14 B, while level: 2 includes segment 16 .

Each segment of level: 0 , i.e., segments 12 A, 12 B,... 12 F, comprise a plurality of processing cells. For example, segment 12 A includes cells 18 A, 18 B and 18 C; segment 12 B includes cells 18 D, 18 E and 18 F; and so forth. Each of those cells include a central processing unit and a memory element, interconnected along an intracellular processor bus (not shown). In accord with the preferred practice of the invention, the memory element contained in each cells stores all control and data signals used by its associated central processing unit.

Certain cells of the processing system 10 are connected to secondary storage devices. In the illustrated system, for example, cell 18 C is coupled with disk drive 19 A, cell 18 D is coupled with disk drive 19 B, and cell 18 O is coupled with disk drive 19 C. The disk drives 19 A- 19 C are of conventional design and can be selected from any of several commercially available devices. It will be appreciated that secondary storage devices other than disk drives, e.g., tape drives, can also be used to store information.

FIG. 2 illustrates in greater detail processing cells and their interconnection within the processing system of FIG. 1 . In the drawing, plural central processing units 40 A, 40 B and 40 C are coupled, respectively, to associated memory elements 42 A, 42 B and 42 C. Communications between the processing and memory units of each pair are carried along buses 44 A, 44 B and 44 C, as shown. Network 46 , representing the aforementioned level segments and routing cells, transfers information packets (passed to the network 46 over buses 48 A, 48 B and 48 C) between the illustrated processing cells 42 A- 42 C.

In the illustrated embodiment, the central processing units 40 A, 40 B and 40 C each include an access request element, labeled 50 A, 50 B and 50 C, respectively. These access request elements generate requests for access to data stored in the memory elements 42 A, 42 B and 42 C. Among access requests signals generated by elements 50 A, 50 B and 50 C is the ownership-request, representing a request for exclusive, modification access to a datum stored in the memory elements. In a preferred embodiment, access request elements 50 A, 50 B and 50 C comprise a subset of an instruction set implemented on CPU's 40 A, 40 B and 40 C. This instruction subset is described below.

The central processing units 40 A, 40 B, 40 C operate under control of an operating system 51 , portions 51 A, 51 B and 51 C of which are resident on respective ones of the central processing units. The operating system 51 provides an interface between applications programs executing on the central processing units and the system 10 facilities, and includes a virtual memory management system for managing data accesses and allocations.

A preferred operating system for controlling central processing units 40 A, 40 B and 40 C is a UNIX-like operating system and, more preferably, OSF/1, modified in accord with the teachings herein.

The memory elements 40 A, 40 B and 40 C include cache control units 52 A, 52 B and 52 C, respectively. Each of these cache control units interfaces a data storage area 54 A, 54 B and 54 C via a corresponding directory element 56 A, 56 B and 56 C, as shown. Stores 54 A, 54 B and 54 C are utilized by the illustrated system to provide physical storage space for data and instruction signals needed by their respective central processing units.

A further appreciation of the structure and operation of the illustrated digital data processing system 10 may be attained by reference to the following co-pending, commonly assigned applications, the teachings of which are incorporated herein by reference:

Application No.	Title	Filing Date
07/136,930	MULTIPROCESSOR DIGITAL	12/22/87
(now U.S. Pat. No.	DATA PROCESSING SYSTEM
5,055,999)
07/696,291	MULTIPROCESSOR SYSTEM	04/26/91
(now U.S. Pat. No.	WITH SHIFT REGISTER BUS
5,119,481)
07/370,341	SHARED MEMORY	06/22/89
(now U.S. Pat. No.	MULTIPROCESSOR SYSTEM
5,297,265)	AND METHOD OF OPERATION
	THEREOF
08/100,100	IMPROVED MEMORY SYSTEM	7/30/93
(now abandoned)	FOR A MULTIPROCESSOR
07/370,287	IMPROVED MULTIPROCESSOR	06/22/89
(now U.S. Pat. No.	SYSTEM
5,251,308)
07/521,798	DYNAMIC PACKET ROUTING	05/10/90
(now U.S. Pat. No.	NETWORK
5,182,201)
07/763,507	PARALLEL PROCESSING	09/20/91
(now abandoned)	APPARATUS AND METHOD FOR
	UTILIZING TILING
07/499,182	HIGH-SPEED PACKET	03/26/90
(now U.S. Pat. No.	SWITCHING APPARATUS AND
5,335,325)	METHOD
07/526,396	PACKET ROUTING SWITCH	05/18/90
(now, U.S. Pat. No.
5,226,039)
07/531,506	DYNAMIC HIERARCHICAL	05/31/90
(now U.S. Pat. No.	ASSOCIATIVE MEMORY
5,344,483)
07/763,368	DIGITAL DATA PROCESSOR	09/20/91
(now abandoned)	WITH IMPROVED PAGING
07/763,505	DIGITAL DATA PROCESSOR	09/20/91
(now U.S. Pat. No.	WITH IMPROVED
5,313,647)	CHECKPOINTING AND FORKING
07/763,132	IMPROVED DIGITAL DATA	09/20/91
(now abandoned)	PROCESSOR WITH
	DISTRIBUTED MEMORY
	SYSTEM
07/763,677	FAULT CONTAINMENT SYSTEM	09/23/91
(now abandoned)	FOR MULTIPROCESSOR WITH
	SHARED MEMORY

Query Decomposition

FIG. 3A depicts a standard arrangement of processes and software modules utilized in digital data processor 10 without query decomposition and data access according to the invention.

FIG. 3B depicts a preferred arrangement of processes and software modules utilized in digital data processor 10 for query decomposition and data access according to the invention. An initiating process 70 generates a query for accessing data stored in relational database 72 having data partitions 72 A, 72 B, 72 C. The query is generated in a conventional format otherwise intended for a conventional DBMS 76 . In a preferred embodiment, that conventional format is SQL and that conventional DBMS is the ORACLE 7 ™ Database Management System (hereinafter, “ORACLE” or “ORACLE Version 7”) of Oracle Corporation. Those skilled in the art will appreciate that other DBMS's and query formats may be substituted for the preferred ones without deviating from the spirit of the invention. However, those skilled in the art will also appreciate that a DBMS (such as ORACLE Version 7) used in connection with the preferred embodiments of invention disclosed below must be capable of efficiently running queries that specify “intersecting predicates” against relevant database partitions, i.e., they must avoid searching partitions other than those specified in those predicates.

Rather than being routed directly to DBMS 76 , the query is intercepted by the parallel user program interface (“PUPI” or “parallel interface”). Element 74 A (responsible for decomposing the query) routes queries not susceptible to decomposition to DBMS 76 , but for a decomposable query it generates a set of subqueries, each of which is based on the initial query but which is directed to data in one or more respective of the partitions 72 A, 72 B, 72 C of database 72 . Then element 74 A initiates and invokes threads 78 A, 78 B, 78 C, which initiate execution of the subqueries. The subqueries corresponding to threads 78 A, 78 B, 78 C are routed to the user program interface (“UPI” or “standard interface”) of DBMS 76 (in lieu of the intercepted query), as shown in the drawing. Multiple subqueries are preferably applied to the UPI of DBMS 76 in parallel with one another, thus capitalizing on the database partitions and on the multiprocessing nature of the preferred digital data processing system 10 . Each thread routes its subquery to a separate server process in DBMS 76 .

The DBMS 76 responds in the conventional manner to each subquery by generating appropriate requests (e.g., a disk read) for access to the database 73 and, particularly, for access to respective partitions of that database (unless the data requested is already in memory). Data retrieved from the database 72 in response to each subquery is processed in the normal manner by DBMS 76 and is routed to processes 76 A, 76 D and 76 G. Those responses, in turn, are routed to parallel interface assembly section 74 B which assembles a response like that which would have been generated by the DBMS 76 had the intercepted response been applied directly to it. The assembled response produced by assembly section 74 B is generally returned to the initiating process 70 more quickly than that which would have been generated by the DBMS 76 had the intercepted query been applied directly to it. This is a consequence of decomposition of the intercepted query and its parallel application to the UPI of DBMS 76 . It is also a consequence of the architecture of the underlying multiprocessor, which permits multiple server processes to run simultaneously. Though it will be appreciated that, even when running on a uniprocessor, the concurrent execution of multiple subqueries could speed access where there is overlapping I/O and CPU processing.

As noted above, the decomposer 74 A generates subqueries based on the conventional-format query intercepted from the initiating process. For simple, single-table queries, the decomposer 74 A generates corresponding subqueries by duplicating the query and appending a predicate for matching records in the corresponding table partition. Thus, for example, a query in the form

SELECT name, department_number

FROM employee

WHERE department_number=10

would result in the first subquery of the form:

SELECT name, department_number

FROM employee

WHERE department_number=10 AND

employee.rowid>=0.0.1 AND

employee.rowid<0.0.2

where rowid has three parts, the last of which indicates the partition number. Other subqueries would be of similar form, with changes to the partition numbers referenced in the rowid predicates.

For queries joining two or more tables, the decomposer 74 A generates corresponding subqueries by duplicating the query and appending a predicate for matching records in the corresponding table partition of the driving table, which is selected by the decomposer 74 A based on the access strategy chosen by the query optimizer portion 76 B of the DBMS 76 . Those skilled in the art will appreciate that information from the optimizer 76 B, including possible tables to be chosen as the driving table, can be obtained from data files generated by the DBMS 76 in connection with the query, and accessed by use of the “EXPLAIN” command.

FIG. 4 shows the operation of assembler 74 B on results generated by the UPI of DBMS 76 and threads 78 A, 78 B, 78 C in response to the subquery signals. More particularly, the drawing shows that for intercepted queries that call for aggregate data functions, element 74 C performs a like or related data function of the results of the subqueries. Thus, for example, if the intercepted query seeks a minimum data value from the database table—and, likewise, the subqueries seek the same minimum value from their respective partitions—then element 74 C generates a final result signal representing the minimum among those reported to the assembler 74 B by the DBMS 76 and threads 78 A, 78 B, 78 C.

Likewise, if the intercepted query seeks an average value from the database table—and, likewise, the subqueries seek a sum and a count from the respective partitions—then element 74 C generates an average table value through a weighted average of the reported subquery results. Moreover, if the intercepted query seeks a standard deviation or variance from the database tables, the decomposer 74 A generates subqueries requesting related functions of the data, e.g., the sum, count and sum of the squares of the data.

Such aggregate processing is preferably applied to, for example, intercepted queries requesting (i) a minimum or maximum of an item in the records (ii) an average of selected items, (iii) a standard deviation and variance of selected items, and (iv) a sum and a count of selected items.

As further shown in FIG. 4 , for intercepted queries that call for non-aggregate data functions, element 74 D generates a final result signal by interleaving the results of the subqueries. For example, if the intercepted query seeks a sorted list of data values from the database table—and, likewise, the subqueries seek sorted lists from their respective partitions—then element 74 D generates a final result signal by interleaving (in the specified sort order) the items presented in the results reported to the assembler 74 B by the DBMS 76 and threads 78 A, 78 B, 78 C. Other non-aggregate queries involving, for example, (i) a distinct value of an entire result row, (ii) a nested selection of items, and/or (iii) a correlated selection of items are processed accordingly.

For queries that combine aggregate and non-aggregate functions, a combination of elements 74 C and 74 D are invoked.

For queries involving grouping operations, the decomposer 74 A generates corresponding subqueries by duplicating the query, along with the grouping clause in its predicate list. For each group, data retrieved by the DBMS in response to those subqueries is placed in a temporary table. For that group, the assembly section 74 B generates and passes to the DBMS a “group by” combining query to be applied to the temporary table. The results of those queries are returned to the initiating process 70 in lieu of the response that would have been generated by the DBMS 76 had the intercepted query been applied directly to it.

For queries involving grouping operations and including a “having” clause, the decomposer 74 A and assembly section 74 B operate in the manner describe above, except, that the “having” clause is not included in the subqueries. That clause is, however, incorporated into the combining queries that are executed on the temporary table.

FIG. 5 depicts a preferred mechanism, referred to as “scatter clustering” or “small bucket hashing,” for storing and retrieving data from database 72 . The mechanism combines cluster-storage and index-access techniques to disperse and retrieve data records from storage media 80 A, 80 B, 80 C (e.g., disk drives) upon which database 72 is contained. Data records are stored using the DBMS's 76 cluster-storing capabilities, based on a conventional hash function of its key value (as generated by element 76 B), and using a smaller-than-normal bucket size chosen to insure that at least one overflow hash bucket will be created for each root bucket. More preferably, the bucket size is chosen to insure that hash buckets are spread over storage devices to maximize the potential for parallel access. Each stored record is simultaneously indexed for direct access in accord with the same key value(s) used by the hash function.

In operation, the DBMS 76 responds to requests to store data records by invoking the hashing element 76 B to store those data records in accord with a hash on their key values. The DBMS 76 also populates index 76 C by invoking DBMS's 76 corresponding indexing functionality. When accessing data records, the decomposer 74 A generates subqueries specifying that requested data records are to be accessed via the index element 76 c, not the hashing element 76 b.

It will be appreciated that, to maximize the performance of the system depicted in FIG. 3B , the database 72 is organized to achieve the best mix of I/O parallelism and hit ratio. Generally, the greater the former (I/O parallelism), the more threads 78 A, 78 B, 78 C can be used, in parallel, to initiate data retrievals. The greater the latter (hit ratio), the greater the number of relevant records each thread 78 A, 78 B, 78 C gets with each retrieval.

Traditional indexed access schemes lend themselves to high degree of I/O parallelism, but low hit ratio. Parallelism is good because new records are allocated randomly in the physical disk structure. The hit ratio is low, however, because each disk access is likely to get little more of interest than the specific record sought (i.e., the data in neighbors of any given record are unlikely to have any relationship to the data in the given record).

Traditional hashing schemes are generally of low I/O parallelism, but have a high hit ratio. Parallelism is low because most of the data with a given key value is stuffed into just a few buckets: the root and a few necessary overflows. The hit ratio is high, however, because each disk access will get several records of related data (i.e., the. neighbors of any given record are likely to be related to the data in the given record).

By combining the DBMS's 76 indexing and hashing mechanisms in the manner described above, the aforementioned scatter clustering technique achieves a good mix of I/O parallelism and hit ratio. It does this by storing the data records using the DBMS's 76 hash-based storage techniques with abnormally small bucket size, thereby distributing small bucket-size clusters of related information around the disk, and by retrieving the data using the DBMS's indexing mechanism.

Those skilled in the art will, of course, appreciate that the invention contemplates operating on database tables with any plurality of partitions. And, that the invention contemplates using any plurality of subqueries (and corresponding threads) to execute retrievals against those partitions. Moreover, it will be appreciated that the invention does not require that the number of partitions and subqueries be identical. Preferably, the number of subqueries (and threads) is an integral divisor, greater than one, of the number of partitions. Thus, for example, three subqueries can be beneficially run against six partitions.

The sections which follow discuss the design considerations of the illustrated, preferred embodiment of the invention, to wit, a system hereinafter referred to as the “Query Decomposer” or “QD ” for parallelizing decision support queries for use on a multiprocessor system of the type shown in FIG. 1 (and commercially available from the assignee hereof, Kendall Square Research Corporation) in connection with version 7 of the ORACLE™ database management system (which is commercially available from Oracle Corporation and can be adapted for operation with a number of computer systems, including the Kendall Square Research Corporation multiprocessors). Each of the sections which follow is identified by a “Database Note Number” (or DBN #). Those identifications are used to cross-reference the sections, typically, in lieu of their titles. The inventors are alternatively referred to as “we,” “I,” “KSR,” and other like terms.

Notwithstanding the grammatical tense of the sections which follow, those skilled in the art will attain the requisite understanding of the invention and the disclosed system upon reading the sections which follow in connection with the other portions of this patent application. In this regard, it will also be appreciated that when the text of the section refers to material “below” or “above,” such reference is typically with respect to material contained within that section itself

Those skilled in the art will attain from study of the sections that follow, not only an appreciation of the workings of an exemplary, preferred illustrated embodiment, but also of its application to other computer systems and DBMS's.

In this regard, a still better appreciation of a preferred embodiment of the invention may be attained by reference to the software appendices filed herewith.

The sections which immediately follow overview the operation and use of a preferred query decomposition system according to the invention.

SUMMARY & CLAIMS

The foregoing describes a digital data processing apparatus and method meeting the aforementioned objects. Particularly, it describes an improved digital data processing system that intercepts selected queries prior to processing by a database management system, that decomposes those queries to generate multiple subqueries for application, in parallel, to the DBMS, in lieu of the intercepted query, and that assembles responses by the DBMS to generate a final response. The foregoing also describes methods and apparatus for storage and retrieval of records from a database utilizing the DBMS's cluster storage and index retrieval facilities, in combination with a smaller-than-usual hash bucket size, to improve parallel access to the database.

Those skilled in the art will appreciate that the embodiments described above are exemplary only, and that other apparatuses and methods—including modifications, additions and deletions—fall within the scope and spirit of the invention. Thus, for example, it will be appreciated that the techniques described above may be utilized on different computing systems and in connection with database management systems different than those described above. It will also be appreciated that differing data structures than those described in the detailed description may be used. And, by way of further example, that equivalent, but varied, procedures may be used to decompose queries and reassemble results without changing the spirit of the invention.

Claims

1. A database query system configured for use with a database management system, said database management system including a standard interface configured to receive database queries, the query system comprising:a parallel interface configured to receive a first database query; and a query decomposer configured to: detect a decomposition directive corresponding to the received database query; generate a plurality of subqueries from the received first database query, wherein said subqueries correspond to said directive; and convey said plurality of subqueries in parallel to the standard interface of the database management system; wherein said directive is embedded within a comment.

2. The system of claim 1, further comprising a result assembler configured to:receive results from said database management system generated in response to said subqueries; and generate an assembled result representative of a response to said first database query.

3. The system of claim 1, wherein said first database query is directed to said standard interface, and wherein receiving the first database query by the parallel interface comprises intercepting the first database query.

4. The system of claim 1, wherein said decomposer is configured to generate said subqueries in response to detecting said first database query is decomposable.

5. The system of claim 4, wherein said decomposer is further configured to route a second database query received by said parallel interface directly to said standard interface without decomposing said second database query, in response to determining said second database query is non-decomposable.

6. The system of claim 1, wherein the database management system is coupled to a database comprising multiple partitions, each of said partitions being independently accessible.

7. The system of claim 1, wherein generating said plurality of subqueries comprises generating a separate thread corresponding to each of said subqueries.

8. The system of claim 7, wherein each thread of said threads corresponding to said subqueries is executed in parallel on separate processors.

9. A computer system comprising:a processor; a storage element; and a database system configured to: receive a first database query; detect a decomposition directive corresponding to the received database query; generate a plurality of subqueries from the received first database query, wherein said subqueries correspond to said directive; and convey said plurality of subqueries in parallel to a database management system; wherein said directive is embedded within a comment.

10. The system of claim 9, further comprising a result assembler configured to:receive results from said database management system generated in response to said subqueries; and generate an assembled result representative of a response to said first database query.

11. The system of claim 9, wherein said first database query is directed to said standard interface, and wherein receiving the first database query by the parallel interface comprises intercepting the first database query.

12. The system of claim 9, wherein said decomposer is configured to generate said subqueries in response to detecting said first database query is decomposable.

13. The system of claim 12, wherein said decomposer is further configured to route a second database query received by said parallel interface directly to said standard interface without decomposing said second database query, in response to determining said second database query is non-decomposable.

14. The system of claim 9, wherein the database management system is coupled to a database comprising multiple partitions, each of said partitions being independently accessible.

15. The system of claim 9, wherein generating said plurality of subqueries comprises generating a separate thread corresponding to each of said subqueries.

16. The system of claim 15, wherein each thread of said threads corresponding to said subqueries is executed in parallel on separate processors.

17. A computer system comprising:a processor; a storage element; and a database system configured to: receive a first database query; detect a decomposition directive corresponding to the received database query; generate a plurality of subqueries from the received first database query, wherein said subqueries correspond to said directive; and convey said plurality of subqueries in parallel to a database management system.

18. The system of claim 17, wherein said database system comprises a query decomposer coupled to said database management system, wherein said query decomposer is configured to receive said first database query and generate said subqueries.

19. The system of claim 18, wherein said database system further comprises a result assembler configured to:receive results from said database management system generated in response to said subqueries; and generate an assembled result representative of a response to said first database query.

20. A computer readable medium comprising program instructions configured to perform database operations, said program instructions executable to:receive a first database query; detect a decomposition directive corresponding to the received database query; generate a plurality of subqueries from the received first database query, wherein said subqueries correspond to said directive; and convey said plurality of subqueries in parallel to a standard interface of a database management system; wherein said directive is embedded within a comment.

21. The system of claim 18, wherein said decomposer is configured to generate said subqueries in response to detecting said first database query is decomposable.

22. The system of claim 21, wherein said decomposer is further configured to route a second database query directly to said database management system without decomposing said second database query, in response to determining said second database query is non-decomposable.

23. The system of claim 17, wherein the database management system is coupled to a database comprising multiple partitions, each of said partitions being independently accessible.

24. The system of claim 23, wherein said database comprises data stored on said storage element.

25. The system of claim 23, wherein said database comprises data stored on a plurality of storage elements, each said partition corresponds to a separate storage element of said plurality of storage elements.

26. The system of claim 17, wherein generating said plurality of subqueries comprises generating a separate thread corresponding to each of said subqueries.