Patent Grant
7567943
This invention is related to systems and methods of constructing a mapping between database schemas, which mapping is composed from existing mappings.
The advent of global communications networks such as the Internet has facilitated access to enormous amounts of data. The world is populated with information sources where in many cases the data is represented differently from source to source. The free flow of information prevalent today in wired and wireless regimes demands that the source and destination be compatible insofar as storing and interpreting the data for use. A major problem facing companies and individuals today is that the data existing in one model/schema needs to be accessed via a different model/schema. However, such processes are being hampered by a largely disparate and ever-changing set of models/schemas. Such an example can be found in data warehousing where data is received from many different sources for storage and quick access from other sources. Converting data from one model to another model is not only time-consuming and resource intensive, but can be fraught with conversion problems.
Schema mappings specify the relationships between heterogeneous database sources and are used to support data transformation, data exchange, schema evolution, constraint checking, and other tasks. Mapping composition is an operation that takes as input two mappings mapAB and mapBC between schemas A, B, C and produces a single mapping mapAC that specifies the same set of constraints between A and C that is given by the combined mappings mapAB and mapBC.
To illustrate the use of mapping composition, consider the following schema evolution scenario. Assume that S1, S2, S3, S4 are versions of a schema used in successive releases of a product. The mapping map12 is used to migrate data from schema S1's format to schema S2's format. Similarly, mapping map23 is used to migrate data from schema S2's format to schema S3's format, and finally, map34 is used to migrate data from schema S3's format to schema S4's format. A conventional way of migrating the data from version S1 to version S4 is by executing map12, map23, map34 one by one, which is time-consuming and costly. To migrate the data from S1 to S4 in a single step, mapping composition is required. The mapping map14 can be obtained by first composing map12 and map23, and then composing the resulting mapping with map34.
Now suppose V1 is a view defined on S1 and map1V1 is a function. To migrate view V1 from S1 to S2, composition is again used. The inverse of mapping map12 is composed with map1V1 to obtain a mapping map2V1 from S2 to V1.
Algorithms for mapping composition are well-known for the case where each mapping is a function (i.e., maps one database state to exactly one database state, and both mappings have the same directionality). However, there is a substantial unmet need for an algorithm which is suitable for a broader class of mappings where one or both mappings are not functions.
The following presents a simplified summary of the invention in order to provide a basic understanding of some aspects of the invention. This summary is not an extensive overview of the invention. It is not intended to identify key/critical elements of the invention or to delineate the scope of the invention. Its sole purpose is to present some concepts of the invention in a simplified form as a prelude to the more detailed description that is presented later.
The invention is a general algorithm that is suitable for composing a broader class of mappings, where one or both mappings are not functions, such as constraints between two schemas and the inverse of functions. This enables construction of a mapping between database schemas from two existing mappings by composing them. The algorithm for composition includes a procedure that tests whether the composition algorithm will terminate.
The invention disclosed and claimed herein, in one aspect thereof, comprises a system that facilitates composition of schema mappings. A composition component performs composition on existing schema mappings of disparate data sources, which schema mappings can be expressed by at least one of full, embedded, and second-order dependencies, wherein the second-order dependencies need not be in source-to-target form.
In another aspect of the subject invention, a methodology is provided for composing two schema mappings by Skolemizing the schema mappings expressed by the embedded dependencies to obtain schema mappings expressed by the second-order dependencies; computing a finite axiomatization of all constraints over input and output signatures which can be deduced from the constraints that give the schema mappings; and de-Skolemizing a finite axiomatization to obtain a schema mapping expressed by an embedded dependency.
In still another aspect thereof, a methodology is provided where the act of de-Skolemizing the finite axiomatization further comprises unnesting Skolem functions using equality to generate a conclusion; checking that all variables in the conclusion appear in a variable dependency set of the Skolem functions; generating all possible clauses that can be obtained by unifying Skolem functions that appear in the conclusion; and de-Skolemizing each clause separately by substituting each Skolem term by a new existential variable.
To the accomplishment of the foregoing and related ends, certain illustrative aspects of the invention are described herein in connection with the following description and the annexed drawings. These aspects are indicative, however, of but a few of the various ways in which the principles of the invention can be employed and the subject invention is intended to include all such aspects and their equivalents. Other advantages and novel features of the invention will become apparent from the following detailed description of the invention when considered in conjunction with the drawings.
The invention is now described with reference to the drawings, wherein like reference numerals are used to refer to like elements throughout. In the following description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the subject invention. It may be evident, however, that the invention can be practiced without these specific details. In other instances, well-known structures and devices are shown in block diagram form in order to facilitate describing the invention.
As used in this application, the terms “component” and “system” are intended to refer to a computer-related entity, either hardware, a combination of hardware and software, software, or software in execution. For example, a component can be, but is not limited to being, a process running on a processor, a processor, an object, an executable, a thread of execution, a program, and/or a computer. By way of illustration, both an application running on a server and the server can be a component. One or more components can reside within a process and/or thread of execution, and a component can be localized on one computer and/or distributed between two or more computers.
Composition of Mappings
The subject invention is a general algorithm that is suitable for composing a broader class of mappings, where one or both mappings are not functions, such as constraints between two schemas (such as map12, map23, and map34) and the inverse of functions (such as the inverse of map12).
As illustrated by the dashed lines, the following scenarios are addressed: when the first schema mapping 104 is a function, the second schema mapping 106 is a non-function; when the first schema mapping 104 is a non-function, the second schema mapping 106 is a function; and, when the first schema mapping 104 is a non-function, the second schema mapping 106 is a non-function. The schema mappings (104 and 106) are composed using a composition algorithm component 108 to generate the resulting mapping 102 (denoted SCHEMA MAPPING13). The composition component 108 employs a deductive procedure to obtain a finite axiomatization of all constraints that need to appear in the composition mapping. This is described in detail infra.
In any of the three non-function mappings, the mappings can be represented by logical constraints, or clauses.
A compose component 208 (similar to component 108) facilitates composition of the two schema mappings (204 and 206) into a resulting schema mapping 210 (denoted MAPPING13). The considered constraint languages include full dependencies (CQ0= mappings), embedded dependencies (CQ= mappings), or second-order dependencies (SOCQ= mappings, also called second-order tuple-generating dependencies). Thus, each of the first and second mappings (204 and 206) can include constraints which are full, embedded and/or second-order dependencies. Given two finite CQ0=, CQ=, or SOCQ= mappings map12 and map23 that are not necessarily functional, the challenge is to find a CQ0=, CQ=, or SOCQ= mapping that specifies the resulting mapping 210 of the composition of map12 and map23.
Referring now to
The composition is computed in three steps: the CQ= mappings are Skolemized to get the second-order dependencies (SOCQ=) mappings. Skolemization is a way to remove existential quantifiers from a formula. Next, a finite axiomatization of all constraints over the input and output signatures is found that can be deduced from the constraints that give the mappings. Finally, the finite axiomatization is de-Skolemized to get a CQ= mapping. Accordingly there is provided a methodology, such that, at 300, the embedded dependencies are Skolemized to obtain the second-order dependencies. At 302, a finite axiomatization of all constraints over input and output signatures is found. The axiomatizations can be deduced from constraints that give the mappings. At 304, the finite axiomatization is de-Skolemized to obtain the embedded dependencies.
In the case of CQ0= mappings and SOCQ= mappings, only act 302 of
Algorithm compose (Σ12, Σ23):
When compose terminates, it is proven that the algorithm gives the correct answer. However, compose may not terminate. At 402, conditions are imposed to check termination of the algorithm. The following conditions can be included to test the sufficiency for algorithm compose to terminate. These conditions can be checked in polynomial time, as indicated at 404, by running k−1 iterations of compose(Σ12, Σ23), where k is the number of relation symbols in S2 and then checking whether a constraint of the form φ(z),R(y)→R(x) appears in Σ, such that,
In order to obtain the embedded dependency, as indicated by act 304 (Step 3) of
Alternatively, or in addition to the algorithm of
All algorithms and variations considered infra can be applied to multiple mappings in this fashion. Specifically, the algorithm presented in
Algorithm compose (Σ12, Σ23, . . . , Σ(n−1)n):
Analysis of Composition and Non-Functional Mappings
The term “mapping” has become established in the database literature to refer to a binary relation on instances, such as database states or XML documents. A general framework for managing mappings, called model management, studies general operators on mappings and on sets of instances, called models. Among these operators, there are such basic operators as domain, range, composition, inverse, etc.
If a mapping is a function, it is said to be functional. Non-functional mappings appear in many contexts, in particular in data integration and data exchange. There are many natural sources of non-functional mappings. For example, the inverse of a functional mapping which is not injective is a nonfunctional mapping.
Significant interest in the art is found in the development of mappings and models given by constraints, and in particular, mappings given by embedded dependencies and by full dependencies. Such mappings are called CQ=-mappings and CQ0=-mappings, respectively, because they are given by inclusions of conjunctive queries with equality (CQ=) and inclusions of conjunctive queries with equality, but without existential quantification (CQ0=). Similarly, models given by embedded dependencies and by full dependencies are called CQ=-models and CQ0=-models, respectively.
Several fundamental questions are explored for basic operators in the relational setting. Let be a language for models and mappings. Given an operator
and
-models and
-mappings for input: (1) Is the output always an
-model or an
-mapping? (That is,
is closed under
); (2) If not, is there a decision procedure to determine when the output is an
-model or an
-mapping?; and (3) In case the output is an
-model or an
-mapping, how hard is it to find an expression for the output?
The present description addresses at least the case when is CQ0=, CQ=, or SOCQ=. For many operators (e.g., identity, intersection, and Cartesian product) it can be seen that the answer to (1) is ‘yes’ and the answer to (3) is ‘very easy’ (see Proposition 3 infra). For some other operators (e.g., range, domain, and composition) it can be shown that the answers to (1) and (2) are ‘no’. Furthermore, large subclasses of CQ0=, CQ=, and SOCQ= are introduced for which the answer to (1) is ‘yes’ and the answer to (3) is ‘easy.’
It can be seen that domain, range, and composition are closely related and can be reduced to each other (see Proposition 4 infra). Therefore, it suffices to study composition and inverse. The latter is easy if symmetric restrictions are placed on the languages that define the mappings. On the other hand, composition turns out to be very hard.
Composition of mappings has been studied in the art. In one study, the class of source-to-target second-order tuple-generating dependencies (ST SO tgds) is introduced and shown that the class of mappings given by ST SO tgds is closed under composition. Moreover, several examples are found which illustrate cases where smaller classes of mappings are not closed under composition. In particular, it can be shown that CQ=-mappings are not closed under composition, even when further restricted to source-to-target dependencies.
The subject invention extends work in the prior art in several directions. Herein, mappings given by SO tgds are called SOCQ=-mappings. Notice that Skolemizing CQ=-mappings gives SOCQ=-mappings. Thus, composition of three major classes of mappings is addressed, each properly containing the previous ones:
One case of interest is the case of embedded dependencies (CQ=-mappings). One way to compute composition of such schema mappings, as referenced in
The subject invention extends the restrictions of languages considered to obtain closure under inverse, which fails in the case of source-to-target constraints, because the inverse of a source-to-target constraint is a target-to-source constraint.
Subclasses of CQ0=, CQ= and SOCQ= are provided which are closed under composition and inverse, and which ideally, include constraints that arise from CQ0= and CQ= queries, which are known to be closed under composition.
It is desired to have widely applicable, polynomial-time-checkable, sufficient conditions for the composition of two CQ0=-mappings (CQ=, SOCQ=) to be a CQ0=-mapping (CQ=, SOCQ=). Moreover, it is preferred to know whether there are computable necessary and sufficient conditions or whether the problem is undecidable.
Summarizing some aspects provided herein, it is shown that CQ0=-mappings are not closed under composition. The problem of determining whether the composition of two CQ0= mappings is a CQ0=-mapping is shown to be undecidable. These results carry over to the case of CQ=-mappings and SOCQ=-mappings. Additionally, several equivalent (non-effective) conditions for composition are provided.
Polynomial-time-checkable sufficient conditions for composition are introduced, which include the case of constraints given by queries. Moreover, subsets of CQ0=, CQ=, and SOCQ= are identified which are closed under composition and inverse, and which include the case of constraints given by queries. Finally, a polynomial-time algorithm is presented for computing the composition of CQ0=, CQ=, and SOCQ=-mappings which satisfy the conditions above.
Preliminary Information
A signature σ={R1, R2, . . . , RN} is a set of relation symbols. A relational atom is a formula of the kind R(t1, . . . , tm), where ti is a variable or constant. An equality atom is a formula of the kind t1=t2, where ti is a term, such as a variable or a constant. CQ is defined as the set of conjunctive queries, i.e., formulas of the kind ∃x1, . . . ∃xm (φ(x1, . . . , xm, y1, . . . , yn)), where φ is a conjunction of relational or equality atoms and each of the free variables y1, . . . , yn occurs in some relational atom. CQ0 is the set of queries with conjunction, but no existential quantifiers (i.e., when m=0; also called select-join queries). Given a query class ,
= is the corresponding query class that also includes equality.
Skolem(Q) is defined to be the formula obtained from Q by replacing existential quantifiers with Skolem functions; it is an existential second-order formula that asserts the existence of the Skolem functions and which does not have first-order existential quantifiers. For example, if
Q=∃y(Exy, Eyz)→Exz
then
Skolem(Q)=∃ƒ(E(x, ƒ(xz)), E(ƒ(xz), z)→Exz).
Set Sk:={Skolem(Q): Q ∈
}. Given a set of constraints ψ over the signature σ1∪σ2, ψ|σ
Models and mappings. A model is a set of instances. A model can be expressed in a concrete language, such as SQL DDL, XML Schema, etc. For example, a relational schema denotes a set of database states; an XML schema denotes a set of XML documents; a workflow definition denotes a set of workflow instances; a programming interface denotes a set of implementations that conform to the interface. A mapping is a relation on instances. Unless stated otherwise, it is assumed that all mappings are binary, i.e., they hold between two models. A mapping can be expressed in a concrete language, such as SQL DML, XSLT, etc.
Constraints for specifying mappings. A constraint is a Boolean query. Sets of constraints are denoted with capital Greek letters, and individual constraints with lowercase Greek letters. The constraints to be considered are in
IC():={∀
}
IC(SOCQ=):=∃
Given a query language , associate to every expression of the form
(σ1, σ2, Σ12)
where Σ12 is a finite subset of IC() over the signature σ1∪σ2 the mapping
{<A, B>: A ∈ mod(σ1), B ∈ mod(σ2), (A, B) Σ12}.
σ1 is the input (or source) signature, and σ2 is the output (or target) signature. Without loss of generality, it is assumed that σ1, σ2 are disjoint or identical.
It is said that m is given by expression E if the mapping that corresponds to E is m. Furthermore, m is an -mapping if m is given by an expression (σ1, σ2, Σ12) with Σ12 a finite subset of IC(
).
Composition. Given two mappings m12 and m23, the composition m12·m23 is the unique mapping
{<A,C>: ∃B(<A, B>∈m12 ^<B,C>∈m23)}.
Deductions and Chase
Following are additional preliminaries provided to explain the algorithm and the correctness proofs in the following sections. The following fixes a specific deductive system, defines the standard notions of ├DC (•), and chase (•), and gives Propositions 1 and 2.
Constraints in IC(SOCQ=) are written as rules of the form
Q1(
leaving the second-order quantifiers over Skolem functions and the outermost first-order universal quantifiers ∀
Definition 1. A deduction from rules Σ is a sequence of rules, each obtained in one of three ways:
A deductive procedure is the one that derives new rules using deductions. A deduction has length n if it consists of n lines. A rule r obtained by expand/rename from rule r′ may have additional atoms in the premise, may have variables replaced consistently by arbitrary terms, and may have replacements in the conclusion consistent with equations in the premise. For example,
R(xy), y=ƒ(xy)→S(xy)
is a valid result of applying expand/rename to R(uv)→S(u ƒ(uv)).
Two formulas can be unified when there exists a variable substitution that renders both formulas syntactically identical. Resolution is a complete logical inference procedure that derives a new rule from two given rules, when an atom of the premise of one rule can be unified with an atom in the conclusion of another rule. For example,
R(xy), S(yz)→S(xz)
is the result of applying resolution to rules
R(xy)→S(xy)
S(xy), S(yz)→S(xz).
The example illustrates a special case of resolution known as modus ponens. A rule r obtained by applying modus ponens to rules r′, r″ consists of the conclusion in r″ and the premise atoms in r′, together with those in r′ which do not appear in the conclusion of r′.
Deductions are annotated by numbering the rules in ascending order and by adding justifications to each line indicating how that line was obtained. It is enough to justify a resolution rule with just two numbers and an expand/rename rule with a single number and a variable assignment. Axiom rules can be indicated through a lack of any other annotation. A variable assignment is a list of items of the form x:=y where x is a variable and y is a term.
Given
Δ={R(1, 1)} and
Σ={R(xy)→S(xy), S(zz)→T(zz)},
the following is a valid deduction from Σ∪Δ:
Here, rules 1, 2, and 5 are axioms, 3 and 6 are expand/rename, and 4 and 7 are resolution.
Within a deduction, a resolution step is called a σ2-resolution if the resolution is performed by unifying an atom from the signature σ2. The rule ξ obtained by σ2- resolution is called σ2-resolvent. A rule ξ is a variant of ξ′ if it can be deduced from ξ′ without using resolution.
Following is an introduction of prerequisites for proving the correctness of the composition algorithm and the undecidability results.
If there is a deduction from a set of constraints Σ where the last line contains a constraint ξ, it can be said that ξ is deduced from Σ, which is written Σ├ξ. The -deductive closure of Σ is
DC(, Σ):={ξ∈
:Σ├ξ}.
Write DC(Σ) when is clear from context. Write D
Σ if all constraints in Σ are true in D. Write Σ
Σ′ if for all instances D, D
Σ implies D
Σ′. It can be checked that if Σ
Σ′, then also Σ
Σ′; i.e., the deductive system is sound. A deductive system is called complete if it guarantees that if a set of sentences is unsatisfiable, then it will derive a contradiction. A deduction witnesses that the derived rules are implied by the axiom rules.
Proposition 1. If Δ is a set of facts, then the following are equivalent:
Proof. If (2) holds then there exist deductions γ and γ′ witnessing Σ├ξ and Δ, ξ├φ. Then γ that witnesses Σ∪Δ├φis obtained by appending γ′ to γ. Now assume that (1) holds and γ witnesses Σ∪Δ├φ. Set γ to γ except for the following replacements, which are made rule by rule from the first rule in γ to the last:
Given the deduction γ from Example 1, γ is:
The following replacements have been made for each rule:
γ is a valid deduction witnessing Σ├ξ since axioms from Δ are no longer used, and since replacements 1 through 5 above ensure that rule R is correctly deduced from the previously replaced rules. Additionally, Δ, ξ├R(
Definition 2. Given an instance D, the result of chasing D with constraints Σ⊂IC(SOCQ=) and the set of Skolem functions F is abbreviated by chase (D, Σ, F) and is denoted by the database D″ obtained from
D′:={Ri(
where
Now to obtain D″ from D′, pick one constant c0 from every equivalence class and replace every constant in that equivalence class with c0. That is, D″:=D′/≡. All functions in F are required to have the same domain. If they are finite, then chase (D, Σ, F) is finite. This definition is a variation on the usual definition, where the functions F are constructed during the chase process.
Proposition 2. Chase (D, Σ, F)Σ.
Composition of CQ0= -Mappings
The following description of composition begins by studying composition of CQ0-mappings and CQ0=-mappings. Later, the techniques introduced to handle these cases are extended to handle SOCQ= and CQ=-mappings. Necessary and sufficient, non-computable conditions for the composition of two CQ0=-mappings to be a CQ0=-mapping are given (see Theorem 1 below). Then, CQ0 and CQ0= are shown not to be closed under composition (see Example 3 below) and, furthermore, that determining whether the composition of two CQ0=-mappings is a CQ0=-mapping is undecidable (see Theorem 2 infra). An algorithm for composition (CQ0=
The composition of two CQ0=-mappings given by (σ1, σ2, Σ12) and (σ2, σ3, Σ23) can be computed using the following algorithm:
Algorithm CQ0=
The next step is to verify the correctness of the algorithm and study the conditions under which the algorithm terminates. Notice that CQ0 and CQ0= are not closed under composition, as the following example shows.
Consider the CQ0-mappings m12 and m23 given by (σ1, σ2, Σ12) and (σ2, σ3, Σ23) where Σ12 is
R(xy)→S(xy)
S(xy), S(yz)→S(xz)
and Σ23 is
S(xy)→T(xy)
and where σ1, ={R}, σ2, ={S}, and σ3, ={T}. Together, these constraints say that R⊂S⊂T and that S is closed under transitive closure. This implies tc(R)⊂T, where tc(R) denotes the transitive closure of R. It can be seen that the only CQ0=-constraints that hold between R and T are constraints of the form
R(xv1), R(v1, v2), . . . ,R(vi−1, vi), R(vi, y)→T(xy)
but no finite set of them expresses tc(R)⊂T.
In fact, the composition m12·m23 is not even expressible in FO (first-order logic). Otherwise, given an FO sentence φ such that
<R, T>∈m12·m23 iff (R,T)φ
an FO formula ψ(xy) can be created, which is obtained by replacing every occurrence of T(uv) in φ with x≠u V y≠v. Then given a domain D with R⊂D2, Rψ[ab]
iff (R, D2−<a,b>)φ
iff tc(R)⊂D2−<a,b>
iff <a,b>∈tc(R)
and therefore, ∀(x,y)ψ(x,y) says that R is a connected graph, which cannot be expressed in FO.
When is the composition of two CQ0=-mappings a CQ0=-mapping? Theorem 1 herein gives necessary and sufficient conditions. An obstacle for the composition to be a CQ0=-mapping can be recursion, yet recursion is not always a problem. In Theorem 3, sufficient conditions are provided that can be checked efficiently, and in Theorem 4, it is shown that good-CQ0=, a subset of CQ0= defined below, is closed under composition. All these results also hold for CQ0.
Theorem 1. If the CQ0=-mappings m1, m2 are given by (σ1, σ2, Σ12) and (σ2, σ3, Σ23) with Σ123:=Σ12∪Σ23 and σ13=σ1∪σ3, then the following are equivalent:
Proof. The proof uses Lemmas 1 and 2 below. The proof first shows the equivalence of (1) and (2), and then shows the equivalence of (2) and (3).
Σ123
DC(CQ0=, Σ123) |σ
DC(CQ0=, Σ13)
Σ13.
This shows that (1) holds.
Σ123
Σ13
DC(CQ0=, Σ13).
Now assume (3) holds. Set Σ to the set of all constraints in DC(CQ0=, Σ123)|σ
Assume that there is a deduction γ witnessing Σ123├ξ. Since (3) holds, it can be assumed that γ has m′ σ2-resolutions. By Lemma 2, there is a deduction γ witnessing Σ123├ξ also with m′ σ2-resolutions, and with all of them occurring before any other resolutions. Since the last line of γ does not contain any symbols from σ2, it can be assumed that γ does not contain any lines containing symbols from σ2 after the last σ2-resolution.
Break γ into two parts: γ1 the initial segment of γ up to and including the last σ2-resolution; and γ2 the remainder of γ. By definition of Σ, every constraint ψ in γ1 must be in Σ, and therefore Σ13├ψ holds. Every constraint ψ in γ2 does not contain any symbols from σ2, and Σ123|σ
Conversely, assume (2) holds. Take m to be the total number of σ2-resolutions needed to deduce every ψ∈Σ13 from Σ123. Assume Σ123├ξ. Then there is a deduction γ witnessing Σ13├ξ. Clearly, γ has no σ2-resolutions. From γ, one can obtain γ witnessing Σ123├ξ by appending to γ a deduction of every constraint in Σ13 and by replacing every line where an axiom from Σ13 is used by a vacuous expand/rename of the line where the deduction of that axiom ends. Clearly, γ has exactly m σ2-resolutions as desired. This shows that (3) holds.
Lemma 1. Under the hypotheses of Theorem 1, the following are equivalent:
Proof. Assume (A, B, C)Σ123 for some B. Then (A, B, C)
DC(CQ0=, Σ123) (by soundness) and therefore (A, C)
DC(CQ0=, Σ123)|σ
Conversely, assume (A, C)DC(CQ0=, Σ123) |σ
(A′, B′, C):=chase((A, Ø, C),Σ123).
If the chase terminates and A=A′ and C=C′, then (A, B, C)Σ123 by Proposition 2, which implies (A, B)
Σ12 and (B, C)
Σ23, as desired.
It is clear that the chase terminates, since no new constants are introduced. Now assume, to get a contradiction, that A≠A′ and C≠C′. Set ΔAC to the set of facts given by A and C. Then,
Σ123∪ΔAC├R(
where R is a relation in A or C not containing
Σ123∪ΔAC├c0=c1
where c0, c1 are distinct constants in A or C.
Consider the former case; the latter is similar. If Σ123∪ΔAC├R( ΔAC, it follows that (A, C)
ξ and ξ∈DC(CQ0=, Σ123)|σ
Lemma 2. Under the hypotheses of Theorem 1, if there is a deduction γ witnessing Σ123├ξ with at most m σ2-resolutions, then there is γ witnessing Σ123├ξ with at most m σ2-resolutions and where furthermore, all σ2-resolutions occur before all other resolutions.
Proof. Assume k<m and γk,l witness Σ123├ξ with
The proof proceeds by induction on k and l. Given γk,l with l>0, the proof shows how to obtain γk,l−1. To prove the case where l=0, simply set γk+1 ,l′:=γk,0 where l′ fulfills the conditions above. Once γm,l′ is obtained for some l′, set γ:=γm,l′ and the lemma is proved.
Consider the line s containing δ, the (k+1)th σ2-resolution in γk,l of, for example, lines i and j containing, respectively, α and β. Consider also the line r containing λ, the l-th non-σ2-resolution in γk,l of, for example, lines r1 and r2 containing, respectively, λ1 and λ2.
Now consider several cases. If i, j<r<s, then obtain γk,l−1 by moving line s to just before r. If lines i, j are not derived from line r, then obtain γk,l−1 by first rearranging γk,l to γk,l so that i, j<r<s, then proceeding as above.
Otherwise, either α or β have been obtained through expand/rename from line r. To simplify the presentation, assume that both have been obtained through a single expand/rename from line r (the other cases are similar). There now exists r<i, j<s. By rearranging γk,l if needed, assume i=r+1, j=r+2 and s=r+3. Since α and β can be obtained from r by expand/rename, α1, α2 and β1, β2 can be obtained, respectively, from r1, r2 by expand/rename so that α is the resolution of α1, α2 and β is the resolution of β1, β2. Replace the four contiguous lines r, i, j, s with the following seven lines.
An important point is that line r+4 now contains a σ2-resolution, since α2 and β1 must resolve through a relation symbol of σ2, because α and β do. Notice that δ is on line r+6, since the result of resolution on α1, α2, β1 and β2 is the same as the result of resolution on α and β (this is because resolution is “associative”).
Corollary 1. Under the hypotheses of Theorem 1, whenever CQ0=
Notice that after the m-th iteration of the main loop, Σ will contain a variant of every constraint that can be deduced using at most m σ2-resolution steps. The constraints of Example 3 fail to satisfy (3) of Theorem 1, and therefore, CQ0=
Consider the CQ0-mappings m12 and m23 given by (σ1, σ2, Σ12) and (σ2, σ3, Σ23) where Σ12 is
R(xy)→S(xy)
S(xy), S(yz)→R(xz),
and Σ23 is
S(xy)→T(xy),
and where σ1={R}, σ2={S}, and σ3={T}. Together, these constraints say that R⊂S⊂T, that R and S are closed under transitive closure (because the constraints
S(xy), S(yz)→S(xz)
R(xy), R(yz)→R(xz)
can be deduced from the constraints above), and that R(xz) holds whenever S(xy), S(yz) holds. The constraints
R(xy), R(yz)→R(xz)
R(xy)→T(xy)
express exactly the composition m1·m2, and are exactly those found by CQ0=
Theorem 2. Checking whether the composition of two CQ0-mappings is a CQ0-mapping is undecidable (in fact, coRE-hard). The same holds with CQ0= instead of CQ0.
Proof. The proof reduces Post's correspondence problem (PCP)—known to be undecidable—to the problem of deciding whether m12·m23 is a CQ0-mapping where m12 and m23 are CQ0 mappings. Given a PCP problem, define m23 so that there is a solution to the PCP problem iff m12·m23 is not a CQ0-mapping (m12 does not depend on the PCP problem).
Write x0011y for the set of atoms
A(x), Z(xa), Z(ab), O(bc), O(cy), B(y)
over σ1, which corresponds to a path from A to B through Z, O (stands for “zero” and O stands for “one”). Similarly, write x′x0011y′y for the set of atoms
Â(xx′), {circumflex over (Z)}(xa), {circumflex over (Z)}(ab), Ô(bc), {circumflex over (Z)}(cy), {circumflex over (Z)}(x′a′), {circumflex over (Z)}(a′b′), Ô(b′c′), Ô(c′y′), {circumflex over (B)}(yy′)
over σ2. Σ12 contains the constraint
A(x), Z(xy), A (x′), Z(x′y′)→Â(xx′), {circumflex over (Z)}(xy), {circumflex over (Z)}(x′y′)
and all corresponding constraints for all combinations AO, ZZ, ZO, OZ, OO, ZB, OB. These constraints imply that xSy, x′Sy′→x′xSy′y can be deduced for any string S.
Encode each PCP “domino” by a constraint in Σ23 as follows. For example, encode the domino 0011/011 as the constraint
Â(xx′), {circumflex over (Z)}(xa), {circumflex over (Z)}(ab), Ô(bc), Ô(cy), {circumflex over (Z)}(x′a′), Ô(a′b′), Ô(b′y′)→Â(xx′), Â(yy′)
Finally, add to Σ23 the constraint
Â(xx′), Â(yy′), {circumflex over (B)}(yy′)→{circumflex over (T)}(xy).
Now assume that the PCP problem has a solution for the string S over {0, 1}*. Then deduce infinitely many constraints of the form
xSk y, x′Sky′→T(xy)
where Sk is the string S repeated k times as follows. Notice that these constraints are over σ1∪σ3 and that none of them can be obtained from any other.
For k=1 (the other cases are similar), first obtain
xSy, x′SyS→x′xSy′y
by using the constraints in Σ12, and then obtain
xSy, x′SyS→A(xx′), A(yy′), B(yy′)
by resolution, in order, using the rules in Σ23 which correspond to the sequence of blocks necessary to obtain S. Finally, obtain
xSy, x′SyS→T(x,y)
using the last constraint in Σ23.
Conversely, if the PCP problem has no solutions, no deduction with premise in σ1∪σ3 and conclusion
Â(xx′), Â(yy′), {circumflex over (B)}(yy′)
is possible. Since this is the only premise matching a conclusion in σ1∪σ3, no constraints in σ1∪σ3 are deduced. In this case Σ13=Ø gives the composition m12·m23.
The coRE-hardness from Theorem 2 implies that algorithm CQ0=
Consider the CQ0-mappings m12 and m23 given by (σ1, σ2, Σ12) and (σ2, σ3, Σ23) where Σ12 is
R(xy)→S(xy)
R(xy), R(yz)→R(xz), R′(y)
S(xy), S(yz)→S(xz), S′(y)
and Σ23 is
S(xy)→T(xy)
and where σ1, ={R, R′}, σ2={S, S′}, and σ3={T}. The constraints
R(xy), R(yz)→R(xz), R(y)
R(xy)→T(xy)
express exactly the composition m1·m2, but algorithm CQ0=
The following conditions are sufficient for algorithm CQ0=
Theorem 3. Under the hypotheses of Theorem 1, if no constraint of the form φ(
Proof. Assume the hypotheses hold. Consider any deduction γ witnessing Σ123 ├ξ which uses only σ2-unifications where ξ contains a σ2 atom in the conclusion. Assume without loss of generality that all rename operations are performed first, followed by all resolution operations. Assume also that every rule in γ contains a single atom in its conclusion and that every rule is used in at most one resolution step.
Such a deduction can be represented as a tree T where every node is one atom. Every non-leaf node in T is the conclusion of some rule r. The children of r are the atoms in the premise of rule r. The premise of ξ consists of all the leaves of T and its conclusion is the root of T.
It is easy to check that any subtree T′ of T which contains, for every node, either all its children in T or none of them, can be converted into a deduction γ witnessing Σ123├ξ′ where the premise of ξ′ consists of all the leaves of T′ and its conclusion is the root of T′.
Since the hypothesis holds, no such subtree may contain S(
Definition 3. A CQ0=-mapping is a good-CQ0=-mapping if it is given by (σ1, σ2, Σ12) such that no constraint of the form φ(
Theorem 4. Good-CQ0= and good-CQ0 are closed under composition and inverse.
Comment 1. CQ0=-queries are included in good-CQ0=-mappings. Notice that total and surjective CQ0=-mappings are not closed under composition.
Consider the CQ0-mappings m12 and m23 given by (σ1, σ2, Σ12) and (σ2, σ3, Σ23) where Σ12 is
R(xy)→S(xy)
R(xy), S(yz)→S(xz)
and Σ23 is
S(xy)→T(xy)
and where σ1={R}, σ2={S}, and σ3={T}. Here m12 and m23 are total and surjective and their composition says that tc(R) ⊂T, which can be seen in Example 5, is not expressible even in FO.
Composition of SOCQ=-Mappings
The focus of this section is the composition of SOCQ=-mappings. In order to handle existential quantifiers in CQ=-mappings, first convert CQ=-mappings constraints into Skolem form, which yields SOCQ=-mappings. Theorem 5 below shows that SOCQ=-mappings are closed under composition, because it can assert the existence of the relations in the intermediate signature or σ2 using existentially-quantified function symbols.
The following related question is also very important:
Conservative composition is important because ultimately there is a need to eliminate all function symbols to obtain CQ=-mappings, and it is not known how to eliminate function symbols introduced in equations such as those used in the proof of Theorem 5 below. Theorems 1 and 2 from the previous section show that SOCQ and SOCQ= are not closed under conservative composition, and that determining whether the conservative composition of two SOCQ=-mappings is a SOCQ=-mapping is undecidable. As in the case of CQ0=-mappings, necessary and sufficient non-computable conditions are given for two SOCQ=-mappings to have a conservative composition (see Theorem 6 below), and sufficient conditions for conservative composition are given that can be checked efficiently. An algorithm for conservative composition (SOCQ=
Before proceeding to the results, the semantics of IC(SOCQ=) constraints are briefly described. A question is, what is the universe from which the functions can take values, i.e., what is their allowed range? Intuitively, the problem is with the universe of the existentially quantified intermediate database. All databases are required to be finite (i.e., all relations are finite), but to have an implicit countably infinite universe. The functions are allowed to take any values from this implicit universe, as long as their range is finite. These assumptions ensure that the deductive system introduced in Section 2 is sound for IC(SOCQ=), which is needed for Theorem 6 below.
Theorem 5. SOCQ=-mappings are closed under composition. Proof. If SOCQ=-mappings m12 and m23 are given by (σ1, σ2, Σ12) and (σ2, σ3, Σ23), Σ13 is obtained such that the composition m12·m23 is given by (σ1, σ2, Σ23) as follows. Set Σ13:=Σ′12∪Σ′23 where Σ′12 and Σ′23 are obtained from Σ12 and Σ23 by replacing every occurrence of an atom R(
Then if <A,C>∈m12·m23, then there is a B such that (A,B)m12 and (B,C)
m23. Set gR(
Σ13. Conversely, if (A,C)
Σ13, then set R:={
m12 and (B,C)
m23. fin(
Theorem 6. If the SOCQ=-mappings m12, m23 are given by (σ1, σ2, Σ12) and (σ2, σ3, Σ23) with Σ123:=Σ12∪Σ23 and σ13=σ1∪σ3, then the following are equivalent;
Lemma 3. Under the hypotheses of Theorem 6, the following are equivalent:
Composition of CQ=-Mappings
This section considers how to convert SOCQ=-mappings back to CQ=-mappings. Consider the case of CQ=-mappings and CQ=-mappings. In order to compute the composition of two CQ=-mappings m12, m23 given by (σ1, σ2, Σ12) and (σ2, σ3, Σ23) progress is in three steps, as follows.
The first step, S
The following example, from prior art, shows that de-Skolemization is not always possible.
Consider the CQ=-mappings m12 and m23 given by (σ1, σ2, Σ12) and (σ2, σ3, Σ23) where
The algorithm D. That is, if Σ├*φ, then Σ
φ. The converse may not hold. There are known cases in which there exists such ├* which is also complete (i.e., the converse does hold). For example, this is the case when Σ has stratified witnesses. For sake of brevity, options for the implementation of ├* are not described. D
Procedure D
At 800, an unnesting process is initiated. Set Λ1:={φ′:φ∈Σ} where φ′ is equivalent to and obtained from φ by “unnesting” terms and eliminating terms from relational atoms and from the conclusion so that in φ:
Relational atoms in which a term variable occurs are called restricting atoms. If A is a restricting relational atom or a restricting equation in which v occurs, where v is a term variable for ƒ, A is called an ƒ-restricting atom. A constraint is called restricted if it has restricting atoms in the premise.
At 802, a check for cycles is performed. For every φ∈Λ1, construct the graph Gφ where the edges are variables in φ and where there is an edge (v, u) iff there is an equation of the form v=ƒ(. . . u . . . ). If Gφ has a cycle, abort. Otherwise, set Λ2:=Λ1.
At 804, a check for argument compatibility is performed. For every φ∈Λ2, check that φ does not contain two atoms with the same function symbol. If it does, abort. Otherwise, set Λ3:=Λ2.
At 806, variables are aligned. Rename the variables in Λ3 to obtain Λ4 satisfying:
At 808, restricting atoms are eliminated. Pick some ordering of the function symbols in Λ4:ƒ1, . . . ƒk. Set Δ0=Λ4. For n=1, . . . , k−1, set Δn+1:={φ′:φ∈Δn} where φ′ is obtained from φ as follows. Set ψ to be φ with the ƒn+1-restricting atoms removed from the premise. If Δn├*ψ, set φ′:=ψ; otherwise set φ′:=φ. In any case, Δn+1≡Δn. Set Λ5:=Δk.
At 810, constraints with restricting atoms are eliminated. Set Λ6 to be the set of constraints φ∈Λ5 which can not be eliminated according to the following test: φ can be eliminated if
At 812, a check for restricted constraints is performed. Set Λ to the set of unrestricted constraints in Λ6. If there is any φ∈Λ6 such that Λ*φ, abort. Otherwise, set Λ7:=Λ.
At 814, a check for dependencies is performed. For every φ∈Λ7 and every variable v in φ, define Dφ,v as follows. If v is not a term variable, set Dφ,v={v}. If v is a term variable and v=ƒ(ū) is its defining equation in φ, then set Dφ,v:=∪u∈{ū}Dφ,u. Intuitively, Dφ,v is the set of variables on which v depends. Set Vφ:=the set of variables which appear in the conclusion of φ. For every term variable v in Vφ, check that Dφ,v=∪u∈VφDφ,u. If this fails, abort. Otherwise, set Λ8:=Λ7.
At 816, dependencies are combined. Set Λ9:={ψΦ:Ø≠Φ⊂Λ8} where ψΦ is defined as follows. If there is a function ƒ which appears in every φ∈Φ, then the premise of ψΦ consists of the atoms in all the premises in Φ and the conclusion of ψΦ consists of the atoms in all the conclusions of Φ (remove duplicate atoms). Otherwise, ψΦ is some constraint in Φ. Notice that Λ9 ⊂Λ8 since ψ{φ}=φ.
At 818, redundant constraints are removed. Pick some set Λ10 Λ9 such that Λ10├φ for every φ∈Λg, and such that this does not hold for any proper subset of Λ10.
Additionally, at 818, functions with ∃ variables are replaced. Set Λ11:={φ′:φ∈Λ10} where the premise of φ′ is the premise of φ with all equations removed and where the conclusion of φ′ is the conclusion of φ, with all variables appearing on the left of equations in φ existentially quantified.
At 820, unnecessary ∃ variables are eliminated. Set Λ12:={φ′:φ∈Λ11} and return Λ12 where φ′ is like φ, but where existentially quantified variables which do not appear in the conclusion atom have been removed (with their corresponding existential quantifier).
Consider three runs of the algorithm D
For completeness, note that each Σ13i is obtained by first de-Skolemizing CQ=-mappings given by Σ12i and Σ23i, which are shown below, and then invoking SOCQ= C
Dependencies α1, α2, α3, β1, β2 are specified as:
In all three runs of D
In the run D
In the run D
In the run D
ψ:=R2(x), y=ƒ(x)→T1(x).
Clearly, Δ0├*ψ, since Δ0⊃{γ1, γ3}├*ψ. Δ1 has no restricting constraints, so process 810 has no effect and process 812 passes. Process 814 succeeds with Λ8=Λ7={ψ, γ2, γ3}, since every dependency in Δ1 has at most one term variable y in its conclusion. For example, doing the calculation for γ3 yields Vγ3={x, y}, Dγ3,x=Dγ3,y=∪u∈V γ3Dγ3,u={x}. In process 816, combining the dependencies for Φ=Λ8 yields
γ4:=R1(y), R2(x), y=ƒ(x)→T1(x), T2(y), R1(y)
(Combinations resulting from proper subsets of Λ8 are not shown for brevity.) In process 818, the redundant constraints which include ψ, γ2, γ3 are removed because they share the premise with γ4 and their conclusion is subsumed by that of γ4; the result is Λ10={γ4}. Finally, at 818, replacing function ƒ by an existential variable in γ4 yields
Λ12={R2(x)→∃y(T1(x)^T2(y)^R1(y))}
Thus, D
Theorem 7. If D
Σ′⊂IC(CQ=) and Σ′≡Σ.
The following follows from the description of the algorithm.
Proposition 3.
From the first part of Proposition 3, it follows that sufficient conditions for success of CQ= C
Since D
Theorem 8. For any subexponential function ƒ, there are source-to-target CQ-mappings m12 and m23 given by (σ1, σ2, Σ12) and (σ2, σ3, Σ23) such that m13:=m12·m23 is a CQ-mapping which is not given by (σ1, Σ3, Σ13) for any Σ13 whose size is less than ƒ(s) where s is the size of Σ12∪Σ23.
Proof. (Outline) Pick k large enough so that ƒ(2kl)<2k where l is the size of the longest constraint below (which 9 is about 18+log k characters long). Such k must exist, since ƒ is subexponential. Set [k]:={1, . . . , k}. Consider the CQ=-mappings m12 and m23 given by (σ1, σ2, Σ12) and (σ2, σ3, Σ23) where
Domain, Range, and Other Operators
Composition is one of the operators that take models and/or mappings as inputs and give models and/or mappings as output. Composition can be performed using a combination of other operators, which in turn can be implemented using similar techniques as outlined above. These other operators include:
These operators are defined as follows.
dom(m):={A:∃B<A, B>∈m}.
rng(m):={B:∃A<A, B>∈m}.∩
:={A:A ∈
, A∈B}.
id():={<A, A>:A∈
}.
×
:={<A, B>:A∈
, B ∈
}.
m1∩m2:={<A, B>:<A, B>∈m1<A, B>∈m2}.
m−1:={<B, A>:<A, B>∈m}.
In a relational case, with signatures, a signature is associated to every model and, input and output signatures to every mapping, and the obvious signature-compatibility conditions on these operators are required.
Proposition 4. Every ⊂CQ0 is closed under identity, cross product and intersection.
Proof If m12 and m34 are given by (σ1, σ2, Σ12) and (σ3, σ4, Σ34), and and
are given by (σ1, Σ1) and (σ2, Σ2), then
To express identity, refer to the third auxiliary signature σ′2 (which is normally ignored) which contains, for every relation symbol R in or σ2, a relation symbol R of the same arity. In this case, σ1=σ2, so σ2=σ′1.
Proposition 5. Each one of the operators, composition, range, and domain can be reduced to any one of the others.
This proves the composition can be expressed by either domain or range.
Conversely, suppose
Then dom(m12) and rng(m12) are given respectively by
(σ1, Σ1) and (σ2, Σ2).
This proves that domain and range can be expressed by composition.
All the languages considered satisfy the premises of Proposition 4. Therefore, Proposition 5 indicates that attention can be focused on closure under composition and inverse. Notice that if an -mapping m is given by (σ1, σ2, Σ12), then its inverse is given by (σ2, σ1, Σ12). However, the restrictions on
may be such that the second expression no longer gives an
-mapping. For example, this happens with “source-to-target” expressions. In contrast, CQ0=, CQ=, and SOCQ= mappings are closed under inverse, so it is sufficient to study composition.
Proposition 5 implies that any algorithmic procedure that can be used to implement domain and range can be deployed for implementing composition. This can be achieved either directly, by invoking the algorithms presented supra, or indirectly, by first executing the operators domain and range, and then using their results to obtain the composition.
Computing Environments
Referring now to
Generally, program modules include routines, programs, components, data structures, etc., that perform particular tasks or implement particular abstract data types. Moreover, those skilled in the art will appreciate that the inventive methods can be practiced with other computer system configurations, including single-processor or multiprocessor computer systems, minicomputers, mainframe computers, as well as personal computers, hand-held computing devices, microprocessor-based or programmable consumer electronics, and the like, each of which can be operatively coupled to one or more associated devices.
The illustrated aspects of the invention may also be practiced in distributed computing environments where certain tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules can be located in both local and remote memory storage devices.
A computer typically includes a variety of computer-readable media. Computer-readable media can be any available media that can be accessed by the computer and includes both volatile and nonvolatile media, removable and non-removable media. By way of example, and not limitation, computer readable media can comprise computer storage media and communication media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital video disk (DVD) or other optical disk storage, or any other medium which can be used to store the desired information and which can be accessed by the computer.
Communication media typically embodies computer-readable instructions, data structures, program modules, and includes any information delivery media. By way of example, and not limitation, communication media includes wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, RF, infrared and other wireless media. Combinations of the any of the above should also be included within the scope of computer-readable media.
With reference again to
The system bus 908 can be any of several types of bus structure that may further interconnect to a memory bus (with or without a memory controller), a peripheral bus, and a local bus using any of a variety of commercially available bus architectures. The system memory 906 includes read only memory (ROM) 910 and random access memory (RAM) 912. A basic input/output system (BIOS) is stored in a non-volatile memory 910 such as ROM, EPROM, EEPROM, which BIOS contains the basic routines that help to transfer information between elements within the computer 902, such as during start-up. The RAM 912 can also include a high-speed RAM such as static RAM for caching data.
The computer 902 further includes an internal hard disk drive (HDD) 914 (e.g., EIDE, SATA), which internal hard disk drive 914 may also be configured for external use in a suitable chassis (not shown), a magnetic floppy disk drive (FDD) 916, (e.g., to read from or write to a removable diskette 918) and an optical disk drive 920, (e.g., reading a CD-ROM disk 922 or, to read from or write to other high capacity optical media such as the DVD). The hard disk drive 914, magnetic disk drive 916 and optical disk drive 920 can be connected to the system bus 908 by a hard disk drive interface 924, a magnetic disk drive interface 926 and an optical drive interface 928, respectively. The interface 924 for external drive implementations includes at least one or both of Universal Serial Bus (USB) and IEEE 1394 interface technologies.
The drives and their associated computer-readable media provide nonvolatile storage of data, data structures, computer-executable instructions, and so forth. For the computer 902, the drives and media accommodate the storage of any data in a suitable digital format. Although the description of computer-readable media above refers to a HDD, a removable magnetic diskette, and a removable optical media such as a CD or DVD, it should be appreciated by those skilled in the art that other types of media which are readable by a computer, such as zip drives, magnetic cassettes, flash memory cards, cartridges, and the like, may also be used in the exemplary operating environment, and further, that any such media may contain computer-executable instructions for performing the methods of the invention.
A number of program modules can be stored in the drives and RAM 912, including an operating system 930, one or more application programs 932, other program modules 934 and program data 936. All or portions of the operating system, applications, modules, and/or data can also be cached in the RAM 912. It is appreciated that the invention can be implemented with various commercially available operating systems or combinations of operating systems.
A user can enter commands and information into the computer 902 through one or more wired/wireless input devices, e.g., a keyboard 938 and a pointing device, such as a mouse 940. Other input devices (not shown) may include a microphone, an IR remote control, a joystick, a game pad, a stylus pen, touch screen, or the like. These and other input devices are often connected to the processing unit 904 through an input device interface 942 that is coupled to the system bus 908, but can be connected by other interfaces, such as a parallel port, an IEEE 1394 serial port, a game port, a USB port, an IR interface, etc.
A monitor 944 or other type of display device is also connected to the system bus 908 via an interface, such as a video adapter 946. In addition to the monitor 944, a computer typically includes other peripheral output devices (not shown), such as speakers, printers, etc.
The computer 902 may operate in a networked environment using logical connections via wired and/or wireless communications to one or more remote computers, such as a remote computer(s) 948. The remote computer(s) 948 can be a workstation, a server computer, a router, a personal computer, portable computer, microprocessor-based entertainment appliance, a peer device or other common network node, and typically includes many or all of the elements described relative to the computer 902, although, for purposes of brevity, only a memory storage device 950 is illustrated. The logical connections depicted include wired/wireless connectivity to a local area network (LAN) 952 and/or larger networks, e.g., a wide area network (WAN) 954. Such LAN and WAN networking environments are commonplace in offices, and companies, and facilitate enterprise-wide computer networks, such as intranets, all of which may connect to a global communication network, e.g., the Internet.
When used in a LAN networking environment, the computer 902 is connected to the local network 952 through a wired and/or wireless communication network interface or adapter 956. The adaptor 956 may facilitate wired or wireless communication to the LAN 952, which may also include a wireless access point disposed thereon for communicating with the wireless adaptor 956.
When used in a WAN networking environment, the computer 902 can include a modem 958, or is connected to a communications server on the WAN 954, or has other means for establishing communications over the WAN 954, such as by way of the Internet. The modem 958, which can be internal or external and a wired or wireless device, is connected to the system bus 908 via the serial port interface 942. In a networked environment, program modules depicted relative to the computer 902, or portions thereof, can be stored in the remote memory/storage device 950. It will be appreciated that the network connections shown are exemplary and other means of establishing a communications link between the computers can be used.
The computer 902 is operable to communicate with any wireless devices or entities operatively disposed in wireless communication, e.g., a printer, scanner, desktop and/or portable computer, portable data assistant, communications satellite, any piece of equipment or location associated with a wirelessly detectable tag (e.g., a kiosk, news stand, restroom), and telephone. This includes at least Wi-Fi and Bluetooth™ wireless technologies. Thus, the communication can be a predefined structure as with a conventional network or simply an ad hoc communication between at least two devices.
Wi-Fi, or Wireless Fidelity, allows connection to the Internet from a couch at home, a bed in a hotel room, or a conference room at work, without wires. Wi-Fi is a wireless technology similar to that used in a cell phone that enables such devices, e.g., computers, to send and receive data indoors and out; anywhere within the range of a base station. Wi-Fi networks use radio technologies called IEEE 802.11(a, b, g, etc.) to provide secure, reliable, fast wireless connectivity. A Wi-Fi network can be used to connect computers to each other, to the Internet, and to wired networks (which use IEEE 802.3 or Ethernet). Wi-Fi networks operate in the unlicensed 2.4 and 5 GHz radio bands, at an 11 Mbps (802.11a) or 54 Mbps (802.11b) data rate, for example, or with products that contain both bands (dual band), so the networks can provide real-world performance similar to the basic 10BaseT wired Ethernet networks used in many offices.
Referring now to
The system 1000 also includes one or more server(s) 1004. The server(s) 1004 can also be hardware and/or software (e.g., threads, processes, computing devices). The servers 1004 can house threads to perform transformations by employing the invention, for example. One possible communication between a client 1002 and a server 1004 can be in the form of a data packet adapted to be transmitted between two or more computer processes. The data packet may include a cookie and/or associated contextual information, for example. The system 1000 includes a communication framework 1006 (e.g., a global communication network such as the Internet) that can be employed to facilitate communications between the client(s) 1002 and the server(s) 1004.
Communications can be facilitated via a wired (including optical fiber) and/or wireless technology. The client(s) 1002 are operatively connected to one or more client data store(s) 1008 that can be employed to store information local to the client(s) 1002 (e.g., cookie(s) and/or associated contextual information). Similarly, the server(s) 1004 are operatively connected to one or more server data store(s) 1010 that can be employed to store information local to the servers 1004.
What has been described above includes examples of the invention. It is, of course, not possible to describe every conceivable combination of components or methodologies for purposes of describing the subject invention, but one of ordinary skill in the art may recognize that many further combinations and permutations of the invention are possible. Accordingly, the invention is intended to embrace all such alterations, modifications and variations that fall within the spirit and scope of the appended claims. Furthermore, to the extent that the term “includes” is used in either the detailed description or the claims, such term is intended to be inclusive in a manner similar to the term “comprising” as “comprising” is interpreted when employed as a transitional word in a claim.
| Number | Name | Date | Kind |
|---|---|---|---|
| 6785673 | Fernandez et al. | Aug 2004 | B1 |
| Number | Date | Country | |
|---|---|---|---|
| 20060136463 A1 | Jun 2006 | US |
