Multi-dimensional associative search engine
AssigneeHyWire Ltd.
InventorMoshe Stark
Life of this patent
Estimating the term…
Loading patent…
AssigneeHyWire Ltd.
InventorMoshe Stark
Estimating the term…
A device for, and method of, storing data in a memory, and for extracting the data therefrom based on a multi-dimensional input (MDI) key, the device including a first and at least a second associative search engine (ASE), each including: (i) a memory having: (A) a first array of cells containing a field of entries, each of the cells being accessible via an input key including a string corresponding to at least a portion of the MDI key, and (B) a second array of cells having a plurality of associated data entries, each of the associated data entries being associated with a particular one of the entries in the first array, and (ii) control logic for operatively connecting the first and the second ASE, the control logic for processing at least a portion of the entries in the first array from each ASE, in response to the input key, so as to determine a match between the input key and an entry of said entries in said field; the control logic for producing a result pertaining to an associated data entry based on the determination, and wherein the control logic utilizes the result from the first ASE in the processing of the second ASE, so as to narrow searching within the second ASE.
A device for storing data in a memory, and for extracting the data therefrom based on a multi-dimensional input (MDI) key, the device comprising:
A method of storing data in a memory, and for extracting the data therefrom based on a multi-dimensional input (MDI) key, the method comprising the steps of:
A method of storing data in a memory, and for extracting the data therefrom based on a multi-dimensional input (MDI) key, the method comprising the steps of:
Dependencies are inferred from the claim text.
Drawings are temporarily unavailable. Open the PDF at USPTO
From the US filing history. Foreign family members aren’t listed.
Backward citations are indexed for grants only.
Details of database functions independent of the retrieved data types: clustering; Classification
Retrieval from the web: indexing; Web crawling techniques
ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or…: for…
Query processing, i.e. searching: pattern matching access
Term end not estimated
The present invention relates to associative search engines (ASEs) and, in particular, to associative search engines for performing operations on multi-dimensional entries having associated data or actions, in response to a multi-dimensional input key.
Packet classification, a well-known problem in computer science, can be defined as follows: given a data packet with certain predefined multiple components, also called packet fields, and an action taken on field-specific packet, this packet can be sorted on the basis of a different action to be taken. Classification is the action of sorting, among all the data packets, the packets that result in the same action.
The set of rules that determines the class to which the packet belongs, is stored in an information base called Classification Information Base (CIB) or a Classifier. The relevant components or fields of a specific data packet classification are assembled into a construct called a Classification Key. The Classification Key forms a CIB query, which results in a specific action to be taken on the packet. Since the Classification Key incorporates multiple fields, and since the CIB results are multiple-field dependent, the result is defined as a Multi-Field Classification.
The invention is herein described, by way of example only, with reference to the accompanying drawings. With specific reference now to the drawings in detail, it is stressed that the particulars shown are by way of example and for purposes of illustrative discussion of the preferred embodiments of the present invention only, and are presented in the cause of providing what is believed to be the most useful and readily understood description of the principles and conceptual aspects of the invention. In this regard, no attempt is made to show structural details of the invention in more detail than is necessary for a fundamental understanding of the invention, the description taken with the drawings making apparent to those skilled in the art how the several forms of the invention may be embodied in practice.
In the drawings:
FIG. 1 is a graphical illustration of the combination of two overlapping ranges and the resulting equivalent set of non-overlapping ranges;
The present invention is a method of, and device performing operations on multi-dimensional entries having associated data or actions, in response to a multi-dimensional input key, using associative search engines. A novel concept, Multi-Dimensional Ranges (MDRs), is introduced, and the utility of such MDRs in compactly representing multi-field information bases is shown. The present invention is particularly suitable for packet classification in the communications field.
The concept presented here provides the capability to decompose a d-Dimensional Range Information Base into d one-dimensional ranges, and each d-Dimensional Key into d one-dimensional keys, and thus transform the difficult search procedure in a d-Dimensional Information Base into d one-dimensional range searches, which can be readily performed by d one-dimensional Range Search Engines (RSEs). This concept can be used for a d-Dimensional Information Base with single (or “binary”) integers, which can be viewed as a special case of a d-Dimensional Range Information Base.
The search algorithm presented involves the use of a Tag as a result (or associated data) of each one-dimensional range search, and concatenate this Tag with the next one-dimensional key searched in the next dimension of the Range Information Base in the following step. This sequence of d concatenated steps results in an Associated Data A (or Action A), which is associated with the d-dimensional range. The search algorithm steps can be pipelined to maximize the d-dimensional search throughput, regardless of the latency of the search procedure.