Method and Apparatus for Efficient Aggregate Computation over Data Streams
Improved techniques are disclosed for processing data stream queries wherein a data stream is obtained, a set of aggregate queries to be executed on the data stream is obtained, and a query plan for executing the set of aggregate queries on the data stream is generated. In a first method, the generated query plan includes generating at least one intermediate aggregate query, wherein the intermediate aggregate query combines a subset of aggregate queries from the set of aggregate queries so as to preaggregate data from the data stream prior to execution of the subset of aggregate queries such that the generated query plan is optimized for computational expense based on a given cost model. In a second method, the generated query plan includes identifying similar filters in two or more aggregate queries of the set of aggregate queries and combining the similar filters into a single filter such that the single filter is usable to prefilter data input to the two or more aggregate queries.
The present invention relates generally to data processing systems and, more particularly, to improved techniques for processing data stream queries in such data processing systems.
BACKGROUND OF THE INVENTIONExamples of data streaming applications include applications that process data such as network traffic records, stock quotes, Web clicks, sensor data, and call records. One type of network traffic record is known as a NetFlow record, which is a record generated in accordance with NetFlow protocol available from Cisco Systems, Inc. (San Jose, Calif.).
Such data streams can generate hundreds of gigabytes of information each day. Processing of such vast amounts of data can obviously place a heavy load on the data processing system that performs such processing. The situation is further exacerbated since analyzing huge volumes of data can require a large number of aggregate queries to be processed. As is known, an aggregate query is a query that performs an aggregate computation (e.g., summation, average, max, min, etc.) on a given data set (e.g., a data stream). These queries may be generated by system administrators seeking to obtain information about the system.
Thus, for realworld deployment, scalability is a key requirement for these types of collection systems. Naïve query answering systems that process the queries separately for each incoming record can not keep up with the high stream rates.
Accordingly, what is required for scalability is an improved technique for processing data stream queries.
SUMMARY OF THE INVENTIONPrinciples of the invention provide an improved technique for processing data stream queries.
For example, in one aspect of the invention, a method includes the following steps. A data stream is obtained. A set of aggregate queries to be executed on the data stream is obtained. A query plan for executing the set of aggregate queries on the data stream is generated. The generated query plan includes generating at least one intermediate aggregate query, wherein the intermediate aggregate query combines a subset of aggregate queries from the set of aggregate queries so as to preaggregate data from the data stream prior to execution of the subset of aggregate queries such that the generated query plan is optimized for computational expense based on a given cost model. By preaggregating the data, the intermediate aggregate query preferably reduces the number of computations that would otherwise be required to generate results of the subset of aggregate queries.
The generated query plan for executing the set of aggregate queries for the data stream may be substantially entirely executed using a main memory of a machine hosting the generated query plan.
The generated query plan may include a tree structure. The query plan generating step may further include determining an optimal query plan with a lowest computation cost by determining a minimumcost aggregate tree. The minimumcost aggregate tree may be determined using a heuristic which performs one or more locallyoptimal modifications to the aggregate tree such that a maximum cost reduction is realized. The minimumcost aggregate tree may be determined using a heuristic which adds one or more random aggregate queries to the aggregate tree to form an expanded aggregate graph, and uses a directed steiner tree heuristic to find the minimumcost aggregate subtree of the expanded aggregate graph.
The generated query plan may further include generating other intermediate aggregate queries, wherein a first one of the other intermediate aggregate queries combines second and third ones of the other intermediate aggregate queries.
The data stream may include records received from a data network, wherein each of the data records includes attributes that describe flow statistics in the data network.
In another aspect of the invention, a method includes the following steps. A data stream is obtained. A set of aggregate queries to be executed on the data stream is obtained. A query plan for executing the set of aggregate queries on the data stream is generated. The generated query plan includes identifying similar filters in two or more aggregate queries of the set of aggregate queries and combining the similar filters into a single filter such that the single filter is usable to prefilter data input to the two or more aggregate queries.
The generated query plan may further include generating other filters, wherein a first one of the other generated filters prefilters data prior to the data entering a second one of the other generated filters, and the second one of the other generated filters prefilters data prior to the data entering one or more of the set of aggregate queries.
In yet another aspect of the invention, apparatus includes a memory, and a processor coupled to the memory and operative to: obtain a data stream; obtain a set of aggregate queries to be executed on the data stream; and generate a query plan for executing the set of aggregate queries on the data stream, wherein the generated query plan comprises at least one of: (i) generating at least one intermediate aggregate query, wherein the intermediate aggregate query combines a subset of aggregate queries from the set of aggregate queries so as to preaggregate data from the data stream prior to execution of the subset of aggregate queries such that the generated query plan is optimized for computational expense based on a given cost model; and (ii) identifying similar filters in two or more aggregate queries of the set of aggregate queries and combining the similar filters into a single filter such that the single filter is usable to prefilter data input to the two or more aggregate queries.
These and other objects, features and advantages of the present invention will become apparent from the following detailed description of illustrative embodiments thereof, which is to be read in connection with the accompanying drawings.
Principles of the invention implement the concept of a query execution plan. Given a set of aggregate queries (also referred to herein more simply as “aggregates”), each of which may or may not involve filters, principles of the invention provide techniques for generating a query execution plan. A query execution plan is basically a structure that describes in which order the queries are to be executed.
As will be explained in detail below, the query execution plan may contain certain one or more intermediate aggregates. These intermediate aggregates are finegrained aggregates, which are then used to generate coarsegrained aggregates. Advantageously, the intermediate aggregates will generally be much smaller than the input data stream itself and so computing multiple query results from an intermediate aggregate will cost much less than answering these queries directly from the data stream.
With respect to filters, principles of the invention provide techniques for coalescing similar filter conditions into a single filter, which is then used as a prefilter to reduce the amount of data input to the queries.
Furthermore, it is demonstrated below that query plans incorporating the above two computation sharing optimizations have a tree structure. Principles of the invention also provide a detailed cost model for aggregate query computation that takes into account hash computation and filter evaluation costs. Thus, the problem of finding the optimal query plan with the lowest computation cost is reduced to that of finding the minimumcost aggregate tree.
It is proven that the problem of finding a minimumcost aggregate tree is NPhard. In accordance with principles of the invention, two heuristics are provided, one greedy and one randomized, to find lowcost aggregate trees. In the greedy heuristic, small locally optimal modifications that deliver the maximum cost reduction in each local step are made to the aggregate tree. The randomized heuristic takes a more global approach. In each iteration, the randomized heuristic adds randomized intermediate aggregates to the tree and then uses a directed steiner tree heursitic (R. Wong, “A Dual Ascent Approach for Steiner Tree Problems on a Directed Graph,” In Mathematical Programming, 1984) to find the minimum cost steiner tree out of the expanded graph.
These and other principles of the invention will be illustrated below in conjunction with NetFlow records associated with an exemplary NetFlow collector (NFC) system (available from Cisco Systems, Inc. (San Jose Calif.)) as the exemplary type of data stream and the exemplary data processing system. It should be understood, however, that the invention is not limited to use with any particular type of data stream or data processing system. The disclosed techniques are suitable for use with a wide variety of other data processing systems which process various types of data streams, and in numerous alternative applications.
Cisco's NetFlow Collector (NFC) (“Cisco CNS NetFlow Collection Engine Installation and Configuration Guide, 3.0,” http://www.cisco.com/univercd/cc/td/doc/product/rtrmgmt/nfc/nfc_{—}3_{—}0/nfc_ug/index.htm) is representative of an emerging class of applications that require multiple OLAP (Online Analytical Processing) style aggregate queries to be processed over a continuous stream of data. NFC collects IP (Internet Protocol) flow records exported by network devices and allows users to run queries for estimating traffic demands between IP endpoints, computing the top hosts in terms of IP traffic, profiling applications, and detecting network attacks and intrusions. For this reason, it is extensively used by network administrators to manage realworld IP networks. However, besides IP networks, such multiplequery streaming applications can be found in other domains as well, for example, financial tickers, retail transactions, Web log records, sensor node readings, and call detail records in telecommunications.
Principles of the invention were at least in part motivated to improve the scalability of NFClike applications so that they can process hundreds of queries. In the following, we describe NFC in further detail.
In an IP network, a flow is essentially a continuous unidirectional sequence of packets from a source device to a destination device. NetFlow, first implemented in Cisco's routers, is the most widely used IP flow measurement solution today. A network device (e.g., router, switch) can be configured to export a single NetFlow data record for every IP flow that passes through it. Each NetFlow record has a number of attributes that describe the various flow statistics. Individual attributes can be classified into one of two categories:
Groupby attributes: These include source/destination IP addresses for the flow, source/destination ports, ToS byte, protocol, input and output interfaces, etc.
Measure attributes: These include the number of packets or bytes in the flow, begin/end timestamp, flow duration, etc.
NFC collects the NetFlow records exported by devices in the network, and processes userspecified aggregate queries on the collected NetFlow data. Each aggregate query consists of: (1) a subset of groupby attributes—records with matching values for attributes in the subset are aggregated together; (2) an aggregate operator (e.g., SUM, COUNT) on a measure attribute—the measure attribute values for aggregated records are combined using the specified aggregate operator; (3) a boolean filter condition on attributes; and (4) a time period over which aggregation is to be performed—after each successive time period, result tuples for the aggregate query (computed over NetFlow records that arrived during the time period) are output.
Below, we give an example NetFlow query that is a variant of the Cisco NFC predefined HostMatrix aggregation scheme (Cisco NFC has 22 predefined aggregation schemes):
groupby: {srcaddr, dstaddr}
aggregateop: SUM(bytes)
filter: (srcaddr ε 135.254.*.*̂ dstaddr ε 135.254.*.*)
period: 15 min
The above query returns the total traffic in bytes between every pair of IP addresses in subnet 135.254.*.*aggregated over 15 minute intervals (note that * is a wildcard that matches any integer between 0 and 255).
A production service provider network contains hundreds of routers which can easily generate massive amounts of NetFlow data. In fact, it is known that even with a high degree of sampling and aggregation, an IP backbone network alone can generate 500 GB (gigabytes) of NetFlow data per day (about ten billion fiftybyte records). The situation is further exacerbated since analyzing the huge volumes of NetFlow data (for diverse network management tasks) requires a large number of aggregation queries to be processed. Thus, for realworld deployment, scalability is a key requirement for a NetFlow data management system like NFC. The system must be able to process, in real time, hundreds of queries over highspeed NetFlow data streams. Naive query answering strategies that process the queries separately for each incoming NetFlow record can not keep up with the high NetFlow stream rates. Thus, we have realized that what is required for scalability are techniques that improve processing efficiency by avoiding redundant work and sharing computation among the various queries.
In an illustrative embodiment, we disclose two computation sharing techniques for scalable online processing of hundreds of aggregate queries on rapidrate data streams. A key idea underlying our techniques, in this embodiment, is to first identify similarities among the groupby attributes and filter conditions of queries, and then use these commonalities as building blocks to generate the final query answers.
In accordance with principles of the invention, we assume that the streaming environment has sufficient memory to process the input aggregate queries. This realization is made possible, for example, due to the fact that: (i) RAM (random access memory) prices have dropped considerably in the past few years, allowing machines to be equipped with several GBs of RAM; and (ii) in our experiments with a number of reallife data sets and aggregate queries, we found that query results can be easily accommodated in main memory. For instance, in the NetFlow record traces stored at the Abilene observatory (“Abilene Observatory Data Collections,” http://abilene.internet2.edu/observatory/datacollections.html), the New York Internet2 backbone router exported a total of 1.7 million NetFlow records in a 20 minute period (from 11:20 to 11:40 on May 8, 2006). For this data, the aggregate results for the 22 default Cisco NFC queries contain approximately 6 million result tuples that take up only 75 MB (megabytes) of memory.
Therefore, based on current technology trends, we have realized that it is practical to process hundreds of stream queries in main memory. Advantageously, as will be illustrated below, this realization leads to query processing approaches that focus on optimizing CPU (central processing unit) cycles as opposed to main memory usage.
For the sake of convenience, the remainder of the detailed description is organized as follows. In Section 1, we describe a system architecture for implementing query processing techniques of the invention. We describe the system model and cost model for processing queries in Section 2. In Section 3, we present our two heuristics for generating treestructured query plans for aggregate queries without filters. We extend our heuristics to handle filters in Section 4.
1. Illustrative System ArchitectureIn general, query plan generation module 105 receives input aggregate queries, filters (if any), and the epoch period. These inputs are defined by the user (e.g., system administrator). While input 106 is referred to as XML (Extensible Markup Language) input in the figure, the query plan generation module of the invention is not limited to processing input of this type. From this input (referred to as 106 in the figure), module 105 generates query plan 107.
Then, with query plan 107 generated by module 105, module 104 inputs the NetFlow records from the various routers and switches (this is more generally considered as streaming data from one or more sources) in the subject network (referred to as 108 in the figure) and generates aggregated output 110. Given generation of the query plan in accordance with the techniques of the invention, aggregated output 110 is generated by module 104.
It is within the data aggregation module and the query plan generation module that techniques of the invention, to be described in detail below in the following sections, are preferably implemented.
The system shown in
In this section, we first describe the aggregation queries supported by our illustrative query processing system, which may be generalized in a straightforward manner to support a broad range of applications including NetFlow data management similar to Cisco's NFC. We then present a naive approach that processes each query independently on the input stream, and finally, we develop a cost model for estimating the CPU cycles consumed for producing query answers.
2.1 System ModelWe consider a single stream consisting of an infinite sequence of tuples, each with groupby attributes a_{1}, . . . , a_{m }(e.g., source/destination IP addresses, source/destination ports), and a measure attribute a_{0 }(e.g., byte count). We are interested in answering a set of aggregate queries Θ={Q_{1}, . . . , Q_{n}} defined over the stream of tuples. A typical aggregate query Q_{i }has three main components, listed below:
Aggregation. This includes: (1) the subset of groupby attributes on which aggregation is performed—a result tuple is output for each distinct combination of these groupby attribute values; and (2) the aggregation operator that is applied to the measure attribute values of aggregated tuples—this is one of the typical SQL (Structured Query Language) aggregates like MIN, MAX, AVERAGE, SUM, or COUNT.
Filter. This is essentially a boolean expression (containing boolean operators and ) over attribute range conditions. Only tuples whose attribute values satisfy the range conditions specified in the filter expression are considered for aggregation. For instance, the filter (srcaddr ε 135.254.*.*̂ dstaddr ε 135.254.*.*) in the above example NetFlow query only aggregates NetFlow records between IP addresses in subnet 135.254.*.*.
Period. This is the time interval (referred to in
In this embodiment, we will assume the following: (1) the measure attribute and aggregation operator are the same for all aggregates; and (2) all aggregate queries in e have the same time period T; thus, result tuples for all aggregates are output at the same time. Our proposed aggregate and filter sharing techniques can, however, be easily extended to handle scenarios when these assumptions do not hold. For example, a straightforward way would be to partition the input query set into subsets of queries, each with identical measure attributes, aggregate operators, and time periods, and then apply our query processing techniques to each subset. Principles of the invention can be extended to other scenarios.
Thus, going back to the assumptions for this embodiment, aggregate queries in Θ differ only in their grouping attributes and filters. Consequently, if A_{i }and F_{i }denote the groupby attributes and filter expression, respectively, for query Q_{i}, then we can completely characterize each query Q_{i }by the pair (A_{i},F_{i}). In the remainder of the detailed description, we will use A to denote the collection of grouping attributes A_{i }for the queries, and Φ for the set of filters F_{i}. We will also use N to denote the number of stream tuples that arrive in time period T. And finally, in view of the abundance of RAM on modern machines, we will assume that there is adequate main memory for processing queries.
2.2 Naive Query Evaluation StrategyA naive strategy is to simply process each aggregation query independently for each incoming stream tuple. For each query Q_{i}, we maintain a separate hash table on the groupby attributes A_{i}. The steps involved in processing query Q_{i }for a tuple are: (1) check if the tuple satisfies the filter condition F_{i}—if not, then simply stop processing the tuple; and (2) hash on the groupby attributes to locate the hash bucket for the tuple, and then update the aggregate statistic for the groupby attribute values. Note that, in the second step, the first time a tuple with a specific combination of grouping attribute values is encountered, a new entry for that group is created (and initialized) in the bucket. If an entry for the group already exists in the bucket, then only the aggregate statistic for the group is updated.
Every time period T, the result tuples for all the aggregates are output by scanning the nonempty buckets in the hash table for each aggregate query, and writing to an output file the groupby attribute values and the aggregate value in every bucket entry. Once all the result tuples are written, all the hash tables are reinitialized by setting their buckets to be empty.
2.3 Query Evaluation Cost ModelNext, let us examine the CPU cost for answering a query Q_{i }using the above naive strategy. First, we introduce some notation. Let σ_{F}_{i }denote the selectivity of the filter condition F_{i}; thus, a fraction a F of stream tuples satisfy F_{i}. Further, let sz(A_{i},F_{i}) be the size of the result after tuples filtered through F_{i }are aggregated on attributes in A_{i}. Both σ_{F}_{i }and sz(A_{i},F_{i}) can be estimated by maintaining random samples of past stream tuples and applying known samplingbased techniques, for example, as disclosed in Moses Charikar et al., “Towards Estimation Error Guarantees for Distinct Values,” In PODS, 2000. Consider a random sample of size r of our stream data set with N tuples. Let f_{1 }and f_{2 }denote the number of values that occur exactly 1 time and 2 or more times, respectively, in the sample. Then the GEE estimator for the number of distinct values is
In this embodiment, we use the same random stream sample to estimate the size of all intermediate aggregates considered in our heuristics. Note that in the presence of filters, we require the values that contribute to the counts f_{1 }and f_{2 }to satisfy the filter.
We will use C_{H}(A_{i}) to denote the cost of hashing a tuple on its groupby attributes A_{i}. Similarly, C_{F}(F_{i}) will denote the cost of checking the filter condition F_{i }for the tuple. We use the UNIX ELF hash function (e.g., Andrew Binstock, “Hashing rehashed,” Dr. Dobbs, April 1996) in our hash table implementation; the function first computes a hash value by performing bit manipulation operations on successive bytes of the input value to be hashed. It then applies a mod function to compute the hash bucket from the hash value. Our filter evaluation operation considers a conjunction of attribute range conditions, and checks the range condition (by performing two comparisons) for each attribute in the filter. We measured the running times (in nanoseconds or ns) for hashing and filtering on a PC with a 3 GHz Intel Pentium 4 processor running Redhat Enterprise Linux 3.0. Both hashing and filtering costs increase linearly with the number of attributes. Hashing incurs about 50 ns for each additional attribute in A_{i}, while filtering requires about 5 ns per attribute range condition in F_{i}. Thus, it follows that hashing is about 10 times more expensive than filtering, for the same number of attributes. In our hash computation experiments, we found the overhead of the final mod function step to be negligible at only about 15 ns. Additionally, when inserting tuples into a hash table, we found that hashing is the dominant cost, and other actions like finding the appropriate bucket entry and updating it consume only a small fraction of the CPU cycles.
Now, the computation cost for query Q_{i }on each stream tuple includes the cost of applying the filter F_{i }to the tuple, and then inserting the tuple into the hash table on attributes A_{i }if it satisfies F_{i}. Thus, since there are N stream tuples in time period T, we get that the CPU cost for processing Q_{i }over time interval T is N·C_{F}(F_{i})+N·σ_{C}_{H}(A_{i}). At the end of time T, the sz(A_{i},F_{i}) result tuples for Q_{i }are output. In general, sz(A_{i}, F_{i}) will be small compared to N, and so we expect output costs to be negligible compared to the computation costs. Also, every query processing scheme will incur identical output costs. So in the remainder of the detailed description, we ignore the cost of writing the result tuples to an output file, and focus primarily on the result computation cost which comprises the CPU cycles for hashing and filtering the incoming stream tuples.
Processing each query in Θ independently (as is done by the naive strategy) may lead to redundant computation. In the following sections, we show that by sharing aggregate computation among the queries in Θ in accordance with principles of the invention, it is possible to achieve a significant reduction in computation overhead and boost overall system throughput.
3. Processing Aggregate Queries Without FiltersWe begin by considering queries without filters. Thus, each query Q_{i}εΘ is simply the groupby attributes A_{i }on which tuples are aggregated, and query processing costs are completely dominated by the hash function computation costs.
For the multiplequery scenario, the naive approach of maintaining separate hash tables for each aggregation query has the drawback that for each streaming tuple, the hash function value is computed n times, once for each input aggregate A_{i}. In this section, we show how we can reduce the hash function computation overhead by sharing hash tables across aggregates.
3.1 Execution Model and Problem FormulationTo reduce the number of hash operations, our technique instantiates a few intermediate aggregates B_{1}, . . . , B_{q }each of whose size is much smaller than N, and then uses them to compute the various A_{i}s. The reason for the small B_{j }sizes is that there will typically be many duplicate tuples in the stream when we restrict ourselves to only the grouping attributes in B_{j}—these will all be aggregated into a single result tuple. Now, it is easy to see that each intermediate aggregate B_{j }can be used to compute any aggregate A_{i}εA that it covers (that is, A_{i}⊂B_{j}). This is because all the groupby attribute values for A_{i }are present in the result tuples for B_{j}. Thus, by making a single pass over the result tuples for B_{j }and inserting them into the hash table for A_{i}, aggregate A_{i }can be computed. In this manner, the result tuples for these intermediate aggregates B_{j }can be used as input (instead of stream tuples) to compute the aggregates in A covered by them. Since the intermediate aggregates B_{j }are much smaller than the tuple stream, it follows that the number of hash computations is significantly reduced.
In general, our technique instantiates an intermediate aggregate if it is beneficial to the overall query processing plan. For an intermediate aggregate to be beneficial, it preferably has the following property. Assume that: N=input size; S=output size; X=sum of the number of groupby attributes in the queries composing the intermediate aggregate; and Y=number of groupby attributes in the intermediate aggregate. If S<(N*(X−Y)/X), then the intermediate aggregate is beneficial. For example, assume there are 1,000,000 records in the stream, and there are two children composing the intermediate aggregate with groupby attributes: A,B and B,C. N=1,000,000 and X=4. The intermediate aggregate would have groupby attributes: A,B,C. Thus, Y=3. Therefore, for the intermediate to be beneficial, S must be less than N*(X−Y)/X, i.e., 1,000,000*(4−3)/4=250,000. Therefore, if the output size is less than 250,000, then S is beneficial. So in this example, S must be ¼th the size of N. However, in practice, it is common to see this ratio exaggerated such that S is many orders or magnitude smaller than N. If the input stream is 1,000,000 records, it is possible for the output size of an intermediate aggregate to be 100 records or less, depending on the data set and the query in question.
More formally, suppose sz(B_{j}) denotes the size of aggregate B_{j}, that is, sz(B_{j}) is the number of distinct value combinations observed for groupby attributes B_{j }in the tuple stream over period T. Then the cost of computing aggregate A_{i }directly from the stream is N·C_{H}(A_{i}). On the other hand, the cost of further aggregating the result tuples for an intermediate B_{j }to compute an aggregate A_{i }that it covers is sz(B_{j})·C_{H}(A_{i}). Thus, by ensuring that sz(B_{j})=N, we can realize substantial cost savings. There is, of course, the additional cost of computing each B_{j }from the input stream, which is N·C_{H}(B_{j}). However, if we select the B_{j}s carefully, then this cost can be amortized across the multiple aggregates A_{i }that are covered by (and thus computed from) each B_{j}.
Next we address the question of what is the best set of intermediate aggregates B_{j }to instantiate? Our discussion above points to B_{j}S that are small and cover many input aggregates A_{i }as good candidates for instantiation. We illustrate the tradeoffs between the different alternatives in the following example.
EXAMPLE 1Consider a stream with attributes a,b,c and d. Also let the aggregates A_{i}εA be defined as follows: A_{i}={a,b}, A_{2}={a,c}, and A_{3}={c,d}. Below, we look at 3 strategies for computing the aggregates A_{i }(we assume that the hashing cost C_{H }(A_{i}) is proportional to the number of attributes in A_{i}).
Strategy 1. This is the naive strategy in which each aggregate A_{i }is computed directly from the stream (see
Strategy 2. This is the other extreme in which we instantiate a single intermediate aggregate that covers all the aggregates A_{i}. (see
Strategy 3. A possible middle ground between the above two extremes is to maintain a single intermediate aggregate B_{2}={a,b,c} and the aggregate A_{3}={c,d} directly on the input stream (see
Now, suppose that N>>sz(B_{2}). Further, suppose that sz(B_{1})≈N. This is entirely possible because B_{1 }contains result tuples for every possible combination of attribute values, and the number of such value combinations could be high. In such a scenario, both strategies 1 and 2 have high computation costs because of the large N and sz(B_{1}) values. In contrast, since sz(B_{2}) is small relative to N and sz(B_{1}), it is easy to verify that Strategy 3 results in the lowest cost among the 3 strategies. In fact, if for B_{3}={a,c,d}, it is the case that sz(B_{3})>sz(B_{2}), then Strategy 3 can be shown to be the best possible strategy for answering the aggregate queries.
Note that it is not necessary to compute every intermediate aggregate B_{j }directly from the stream. Rather, it may be possible to reduce hash computation costs by computing an intermediate B_{j }from another intermediate aggregate, and then using B_{j }to compute multiple aggregates A_{i}. For instance, in Example 1, if N>>sz(B_{1}) and sz(B_{1})>>sz(B_{2}), then the following strategy (depicted in
Also, observe that each of the query plans considered above (and shown in
Aggregate Trees. An aggregate tree is a directed tree with: (1) a special root node corresponding to the input stream; and (2) other nodes corresponding to aggregates. The aggregate for vertex v_{i }is denoted by A(v_{i}). At the root node, since the input stream is not aggregated, we use the special symbol T for A(root). T covers every other aggregate A(v_{i}) but not vice versa, that is, A(v_{i})⊂T for all A(v_{i})—this is because any aggregate can be generated from the input stream. Further, since the root includes all the stream tuples, sz(T)=N.
A directed edge v_{1},v_{2} from vertex v_{1 }to vertex v_{2 }can be present in the tree only if the aggregate for v_{1 }covers the aggregate for v_{2 }(that is, A(v_{2})⊂A(v_{1})). Note that there are no incoming edges into the root node. However, there are no restrictions on outgoing edges from the root, that is, there can be edges from the root to any other node in the tree. Further, all nodes in the aggregate tree are reachable from the root. Each edge v_{1},v_{2}in the tree has an associated cost given by sz(A(v_{1}))·C_{H}(A(v_{2})). Note that the cost of any edge v_{1},v_{2} originating at the root is N·C_{H}(A(v_{2})). The cost of a tree is simply the sum of the costs of all its edges.
Intuitively, an aggregate tree corresponds to a query plan capable of generating answers for every aggregate contained in the tree. The directed edge v_{1},v_{2} implies that node v_{2}'s aggregate is generated from that of node v_{1}'s. This is possible because A(v_{2})⊂A(v_{1}) for a nonroot v_{1}, and any aggregate can be generated from the input stream associated with the root node. The plan for a tree generates aggregates in two phases:
Realtime streaming phase. Only the child aggregates of the root node are maintained as tuples are streaming in. Each streaming tuple is inserted into the hash tables of each of the root's children.
Periodic results output phase. At time intervals of period T, the root's children are used to generate the remaining aggregates in the tree. Starting with each child, aggregates are generated by performing a depth first traversal of the tree. Every time a directed edge v_{1},v_{2} is traversed, the aggregate for v_{2 }A(v_{2}) is produced from the result tuples for A(v_{1}).
Observe that the cost of the edge v_{1},v_{2} is the hash computation cost of producing the aggregate A(v_{2}) from aggregate A(v_{1})—this is the cost of scanning the sz(A(v_{1})) result tuples for aggregate A(v_{1}) (or N stream tuples if v_{1 }is root) and inserting them into the hash table for aggregate A(v_{2}). Thus, the cost of an aggregate tree reflects the total computation cost of producing all the aggregates in the tree.
Thus, our problem of finding a good query plan (with low hash computation costs) to process the aggregate queries in A reduces to the following:
Given an aggregate set A, compute the minimumcost aggregate tree T that contains all the aggregates in A.
Our aggregate tree concept allows us to effectively capture, within a single unified framework, the computation costs incurred during the realtime streaming and periodic results output phases. In contrast, existing schemes such as that disclosed by Rui Zhang et al. (“Multiple Aggregations over Data Streams,” In SIGMOD, 2005) focus exclusively on optimizing the realtime streaming phase cost, which is the dominant cost when the available space is low and collision rates are high. However, this can lead to poor query plans for environments that are not necessarily memoryconstrained—this is because in such environments, the periodic results output phase cost becomes significant due to low collision rates, and this is not considered by Rui Zhang et al. Note that as shown above in Example 1, the minimumcost aggregation tree for A may contain intermediate aggregates not in A.
We have proven that the following decision version of our aggregate tree computation problem is NPhard: Given an aggregate set A and a constant τ, is there an aggregate tree T with cost at most r that also contains all the aggregates in A?
3.2 Heuristics for Computing Aggregate TreesIn this section, we present two heuristics for computing an appropriate aggregate tree. The first is a greedy heuristic that applies a series of local modifications to the tree, at each step, selecting the modification that leads to the biggest cost reduction. The second is a randomized heuristic that adopts a more global approach; it relies on the observation that the aggregate tree computation problem has strong similarities to computing a directed Steiner tree over the global aggregate space. So, directed Steiner approximation algorithms such as the one proposed in M. Charikar et al., “Approximation Algorithms for Directed Steiner Problems,” In SODA, 1998 or heuristics like the one in R. Wong, “A Dual Ascent Approach for Steiner Tree Problems on a Directed Graph,” In Mathematical Programming, 1984 can be used to compute an appropriate aggregate tree.
3.2.1 Greedy HeuristicAlgorithm 1 shown in
Now, lets look at the rationale behind our two local modifications. For a pair of aggregates A,B whose union C is much smaller than their current parent P, our first modification enables cost savings of sz(P)−2·sz(C)≈sz(P) to be realized by adding the new aggregate C to the tree. This is because generating C from P requires sz(P) hash computations, and then generating A,B from C incurs an additional 2·sz(C) hash operations, while generating A, B directly from P requires 2·sz(P) operations. The second modification considers the opposite situation when the size of an aggregate A is close to the size of its parent P in the tree—in this case, the extra cost of generating A from P does not offset the cost reduction when A's children are generated from A instead of P. Thus, it is more beneficial in this case to delete A from the tree and compute A's children directly from P.
Note that, in the worstcase, we may need to consider a quadratic (in n, the number of input aggregates) number of local modifications in a single iteration. Since the cost benefit of each local modification can be computed in constant time, each iteration has a worst case time complexity that is quadratic in the size of the input.
3.2.2 Randomized HeuristicAs is evident, the greedy heuristic considers local modifications like merging a pair of siblings. In contrast, the randomized heuristic that we propose in this section takes a more global perspective—in each merge step, it coalesces multiple randomly chosen aggregates from A to generate new intermediate aggregates.
Before discussing our randomized heuristic, we make an important observation that relates our aggregate tree computation problem to the problem of computing a directed steiner tree. Consider the graph containing a node for every possible aggregate (that is, every possible subset of groupby attributes), and also T for the input stream. In the aggregate graph, there is a directed edge from aggregate A to aggregate B if A covers B, and the cost of the edge is sz(A)·C_{H}(B). Now, it is easy to see that computing the optimal aggregate tree T is nothing but computing a directed steiner tree (in the graph) that connects the root T to the set of aggregates A.
Although computing a directed steiner tree is an NPhard problem, there exist approximation algorithms (e.g., M. Charikar et al., “Approximation Algorithms for Directed Steiner Problems,” In SODA, 1998) and heuristics (e.g., R. Wong, “A Dual Ascent Approach for Steiner Tree Problems on a Directed Graph,” In Mathematical Programming, 1984) in the literature for computing such a tree. Thus, we could theoretically use a directed steiner approximation algorithm to find a good aggregate tree in the full aggregate graph. However, the problem with this is that the full graph contains 2′ nodes (a node for every subset of groupby attributes). This is exponential in the number of attributes, and so any approach that is based on creating the full graph will only work for a small number of attributes.
As illustrated in
Advantageously, since the running time of each iteration of Algorithm 2 is dominated by steiner tree computation, our randomized heuristic scales well with the number of queries.
4. Processing Aggregate Queries With FiltersWe now turn our attention to aggregate queries with filters. So, each query Q_{i }now consists of a set A_{i }of grouping attributes and a filter F_{i}. In the following subsections, we will show how the aggregate tree concept and our heuristics for computing good trees can be extended to handle these richer query types.
4.1 Execution Model and Problem FormulationIn the presence of filters, principles of the invention can reduce computational overhead by sharing filter evaluation among the various queries. For instance, we can coalesce a group of similar query filters, and then with a single application of the coalesced filter, discard a significant fraction of stream tuples that are not relevant to the queries. Further, depending on the selectivity of filters, the location and order in which filters and hashing operations are executed in the aggregate tree can make a substantial difference to the overall computation costs. We illustrate these ideas in the following example.
EXAMPLE 2Consider a stream with attributes a, b, c, and d each with domain {0, . . . ,1000}. For purposes of illustration, we assume that attribute values are uniformly distributed and independent. Let there be three queries: (1) Q_{1 }with groupby attributes {a,b} and filter 0≦a≦95; (2) Q_{2 }with groupby attributes {a,c} and filter 50≦a≦100; and (3) Q_{3 }with groupby attributes {a,d} and filter 200≦a≦300. Now there are multiple query evaluation strategies possible here, which we consider below.
Strategy 1. The naive strategy is to process each query separately (see
Strategy 2. Now a more efficient strategy can be devised based on the observation that the filters F_{1 }and F_{2 }have a fair amount of overlap and so can be merged to create a new filter, G_{1}=0≦a≦100. Note that G_{1 }is equivalent to F_{1}F_{2}. The idea then would be to evaluate the filter G_{1 }for every stream tuple, and only if the tuple satisfies G_{1 }would we check the filters F_{1 }and F_{2 }for the queries Q_{1 }and Q_{2}, respectively. Of course, if the tuple does not satisfy G_{1}, then it cannot possibly satisfy F_{1 }or F_{2}, and thus, the tuple can be safely discarded. Thus, with Strategy 2 (depicted in
Strategy 3. Next observe that filter F_{1 }has significant overlap with filter G_{1}. Consequently, when F_{1 }is applied immediately after G_{1 }on stream tuples (as in
Now suppose that the aggregated result size sz(A_{1}, G_{1})=σ_{G}_{1}·N. Then, Strategy 3 (depicted in
Observe that the same argument does not hold for F_{2 }which filters (σ_{G}_{1}−σ_{F}_{2})·N tuples thus saving 0.05·N·C_{H}(A_{2}) in hashing costs. Since checking F_{2 }on the filtered stream from G_{1 }costs only 0.1·N·C_{F}(F_{2}), the cost savings from hashing fewer tuples far outweigh the additional cost of evaluating F_{2}—thus, in Strategy 3, we apply F_{2 }before tuples are inserted into the hash table for Q_{2}.
Strategy 4. Now if sz(B_{1})=N for aggregate B_{1}={a,b,c}, then in addition to applying the filter G_{1 }on the tuple stream, Strategy 4 (shown in
For simplicity of exposition, we will initially only consider filters that are conjunctions () of attribute range conditions. Thus, each filter is a multidimensional box whose boundaries along a dimension coincide with the range specified for the attribute corresponding to the dimension. Only tuples belonging to the box (with attribute values in the ranges specified in the filter) are considered for aggregation. The union F=_{1}∪F_{2 }of two filters F_{1 }and F_{2 }is a box that contains the boxes for F_{1 }and F_{2}. Essentially, in the union F, the range for each attribute a contains its ranges in F_{1 }and F_{2}. For example, if F_{1}=(0≦a≦5̂0≦b≦5) and F_{2}=(5≦a≦10̂5≦b≦10), then their union F=(0≦a≦10̂0≦b≦10). In Section 4.3, we will discuss how our techniques can be extended to handle filters containing disjunctions () as well.
We will also assume that for each query Q_{i}, the filter attributes in F_{i }are a subset of the groupby attributes A_{i}. We expect that this will be the case for a majority of the queries. For the few queries Q_{i }that do not satisfy this assumption, we can either: (1) process Q_{i }separately; or (2) process a variant Q′_{i }of Q_{i }jointly with other queries in Θ if we find that this leads to lower query processing costs. Here, Q′_{i}=(A′_{i}, F′_{i}) is derived from Q_{i}, and has the same filter as Q_{i }(that is, F′_{i}=F_{i}), but its groupby attributes set A′_{i }contains attributes in both A_{i }and F_{i}. Since A_{i}⊂A′_{i}, the answer for Q_{i }can be derived from the result for Q′_{i }by performing a final additional aggregation step. Note that the cost for the additional aggregation step needs to be added to the processing cost for Q′_{i}.
Aggregate Trees. In the presence of filters, each node of the aggregate tree is a (filter, grouping attributes) pair. Note that there is an implicit ordering of filter and aggregation operations in each node depending on the input tuples to the node. We discuss details below. The root node is special with a (filter, attributes) pair equal to (T, T), and corresponds to the input stream. Here, T is a special symbol that contains all other filters and grouping attributes, but not vice versa, Further, all tuples satisfy the filter condition T. Intuitively, nodes with groupby attributes equal to T perform no aggregation, and nodes with filters equal to T do no filter checks. In the aggregate tree, there can be an edge from a vertex v_{1 }to a vertex v_{2 }only if v_{1 }covers v_{2}, that is, the filter and groupby attributes of v_{1 }contain the filter and groupby attributes, respectively, of v_{2}. Note that since T contains every other filter and groupby attributes, the root can have edges to every other node in the tree.
Execution Plan for Aggregate Trees. Now, an aggregate tree essentially specifies an execution plan for answering the input aggregate queries. Let V denote the set of tree nodes where incoming stream tuples are first aggregated. More formally, V contains all tree nodes v such that: (1) the groupby attributes of v is not T (that is, v performs aggregation); and (2) the path from the root to v only has nodes with groupby attributes equal to T (that is, none of v's ancestors perform aggregation).
As before, the execution plan has two phases:
Realtime streaming phase: We maintain a hash table for each intermediate node v in V on the grouping attributes for v. Each incoming stream tuple is inserted into the hash table for v if and only if it satisfies all the filters in the path from the root to v.
Periodic results output phase: After time period T, the result tuples in the hash table for each intermediate node v in V are used to compute the result tuples for nodes in the aggregate subtree rooted at v. Essentially, the result tuples for v are used to compute the result tuples for v's children, and their result tuples, in turn, are used to compute the result tuples for their children, and so on. Let v_{1 }be V_{2}'s parent in the subtree (v_{1 }and v_{2 }differ in their filters or their grouping attributes). Also, let (G_{i},B_{i}) denote the (filter, groupby attributes) pair at node v_{i}. Then, when computing v_{2}'s result tuples from v_{1}'s tuples, we need to consider the following three cases.
Case 1: v_{2}'s filter is identical to v_{1}'s filter. Note that this covers the case that v_{2}'s filter is T. In this case, all the result tuples for v_{1 }are aggregated on v_{2}'s groupby attributes by inserting them into a hash table on v_{2}'s attributes (without any filtering). The aggregated tuples in the hash table are the result tuples for v_{2}, and the cost of computing these tuples is sz(B_{1}, G_{1})·C_{H}(B_{2}).
Case 2: v_{2}'s groupby attributes are identical to v_{1}'s attributes. Note that this covers the case that v_{2}'s grouping attributes are T. In this case, only v_{2}'s filter condition is applied to all the result tuples for v_{1 }(without any aggregation), and those that satisfy the filter constitute the result tuples for v_{2}. The cost of computing these tuples is sz(B_{1},G_{1})·C_{F}(G_{2}).
Case 3: v_{1 }and v_{2 }have different filters and groupby attributes. In this case, we have two options: (1) first apply V_{2}'s filter to v_{1}'s result tuples, and then aggregate the ones that satisfy the filter on v_{2}'s groupby attributes; or (2) first aggregate v_{1}'s result tuples on v_{2}'s groupby attributes, and then filter out the aggregate tuples that do not satisfy v_{2}'s filter. Depending on which of the two options has a lower cost, we will order the filtering and aggregation operations in v_{2 }differently. The costs of the two options are as follows:
Option (1)cost=sz(B_{1}, G_{1})·C_{F}(G_{2})+sz(B_{1}, G_{2})·C_{H}(B_{2})
Option (2)cost=sz(B_{1}, G_{1})·C_{h}(B_{2})+sz(B_{2}, G_{1})·C_{F}(G_{2})
Thus, the cost of computing v_{2}'s result tuples is the minimum of the costs of options (1) and (2) above. Intuitively, if sz(B_{1}, G_{2})=sz(B_{1},G_{1}), then Option (1) is preferable. If this is not the case and if sz(B_{2},G_{1})=sz(B_{1},G_{1}), then Option (2) may prove to be better.
Problem Definition. We assign a cost to each tree edge (v_{1}, v_{2}) equal to the CPU cost of materializing the result tuples for v_{2 }using the tuples of v_{1 }(as described in the 3 cases above). Thus, the aggregate tree cost (which is the sum of the edge costs) reflects the total CPU cost of processing all the input aggregate queries. Our objective then is to find the minimumcost aggregate tree containing all the input aggregate queries in Θ.
4.2 Heuristics for Computing Aggregate TreesIt can be proven that the more general problem of computing the optimal aggregate tree for queries containing filters is NPhard. In the following subsections, we extend the greedy and randomized heuristics presented above in sections 3.2.1 and 3.2.2, respectively, to compute a satisfactory lowcost aggregate tree.
4.2.1 Greedy HeuristicIn each iteration, our modified greedy heuristic applies four types of local modifications to the tree, and selects the one that results in the largest cost reduction. Of the four modifications listed below, the first two are variants of previously proposed modifications for queries without filters (see Algorithm 1 in
1. For every pair of sibling nodes v_{1}, v_{2 }(with parent p), create a new node v with p as parent, and make v_{1},v_{2 }children of v. Set node v's filter and groupby attributes equal to the union of the filters and groupby attributes, respectively, of v_{1 }and v_{2}.
2. For every node v∉Θ (with parent p), delete v from the tree, and make p the parent of v's children.
3. For every node v∉Θ, modify v's groupby attributes to be equal to its parent's groupby attributes.
4. For every node v∉Θ, modify v's filter to be equal to its parent's filter.
Similar to Algorithm 2 (in
1. Randomly select a subset of input query nodes from Θ.
2. Let v denote the union of (filters and groupby attributes of) the nodes selected above. Add v to R.
3. For every other node u in S that covers v, we add the following two additional nodes x and y to R:

 Node x with v's filter, but u's groupby attributes.
Node y with v's groupby attributes, but u's filter.
4.3 Handling Complex FiltersOur proposed techniques can be extended to handle complex filters containing disjunctions (in addition to conjunctions). We will assume that each filter F is in disjunctive normal form, that is, each filter has the form D_{1} . . . D_{1 }where each D_{1 }is a conjunction of attribute range conditions. Thus, our filter F now is a union of multiple boxes instead of a single box. Consequently, we can model the cost C_{F}(F) of evaluating filter F as Σ_{i}C_{F}(D_{i}), and for estimating the size of aggregates with filters, we can use the samplingbased estimator described in the previous subsection.
Now, in our heuristics, we compute the filter F for a new node in the aggregate tree as the union F_{1}∪ . . . ∪F_{q }of multiple filters. When each F_{i }is a single box, their union is simply the box that contains all the filter boxes. However, when each F_{i }is a set of boxes {D_{1}^{i}, . . . , D_{l}_{i}^{i}}, the union computation for F=F_{1}∪ . . . ∪F_{q }is somewhat more involved. We begin by initializing the union F to be the set of all the boxes D_{j}^{i}, that is, F={D_{j}^{i}:1≦i≦q, 1≦j≦l_{i}}. Now, if F is used to prefilter tuples into the filters F_{i}, then the filtering cost per tuple is C_{F}(F)+σ_{F}·Σ_{i}C_{F}(F_{i})—here the first term is the cost of checking whether the tuple satisfies F and the second term is the cost of checking filters F_{i }if the tuple satisfies F. Clearly, the ideal value for the union F is one that minimizes the filtering cost C_{F}(F)+σ_{F}·Σ_{i}C_{F}(F_{i}) So we repeat the following step until no further improvement in filtering cost is possible: Let D_{1},D_{2 }be the pair of filter boxes in F whose merging results in an F with the smallest filtering cost; merge D_{1}, D_{2 }(by taking their union) into a single box.
As described above in detail, principles of the invention provide two techniques for sharing computation among multiple aggregate queries over a data stream: (1) instantiating certain intermediate aggregates; and (2) coalescing similar filters and using the coalesced filter to prefilter stream tuples. We proposed two heuristics, one greedy and another randomized, for finding lowcost query plans incorporating the above optimizations. In our experiments with reallife NetFlow data sets, we found that our randomized heuristic generated the best query plans with maximum sharing—this is because it adopts a more global approach, continuously interleaving optimization steps with random perturbations to the query plan. In fact, query plans output by our randomized heuristic boosted system throughput by over a factor of three compared to a naive approach that processes queries separately.
Although illustrative embodiments of the present invention have been described herein with reference to the accompanying drawings, it is to be understood that the invention is not limited to those precise embodiments, and that various other changes and modifications may be made by one skilled in the art without departing from the scope or spirit of the invention.
Claims
1. A method, comprising:
 obtaining a data stream;
 obtaining a set of aggregate queries to be executed on the data stream; and
 generating a query plan for executing the set of aggregate queries on the data stream, wherein the generated query plan comprises generating at least one intermediate aggregate query, wherein the intermediate aggregate query combines a subset of aggregate queries from the set of aggregate queries so as to preaggregate data from the data stream prior to execution of the subset of aggregate queries such that the generated query plan is optimized for computational expense based on a given cost model.
2. The method of claim 1, wherein the intermediate aggregate query reduces the number of computations that would otherwise be required to generate results of the subset of aggregate queries.
3. The method of claim 1, wherein the generated query plan for executing the set of aggregate queries for the data stream is substantially entirely executed using a main memory of a machine hosting the generated query plan.
4. The method of claim 1, wherein the generated query plan comprises a tree structure.
5. The method of claim 4, wherein the query plan generating step further comprises determining an optimal query plan with a lowest computation cost by determining a minimumcost aggregate tree.
6. The method of claim 5, wherein the minimumcost aggregate tree is determined using a heuristic which performs one or more locallyoptimal modifications to the aggregate tree such that a maximum cost reduction is realized.
7. The method of claim 5, wherein the minimumcost aggregate tree is determined using a heuristic which adds one or more random aggregate queries to the aggregate tree to form an expanded aggregate graph, and uses a directed steiner tree heuristic to find the minimumcost aggregate subtree of the expanded aggregate graph.
8. The method of claim 1, wherein the generated query plan further comprises generating other intermediate aggregate queries, wherein a first one of the other intermediate aggregate queries combines second and third ones of the other intermediate aggregate queries.
9. The method of claim 1, wherein the data stream comprises records received from a data network, wherein each of the data records comprises attributes that describe flow statistics in the data network.
10. An article of manufacture comprising a processorreadable storage medium storing one or more software programs which when executed by a processor perform the steps of the method of claim 1.
11. A method, comprising:
 obtaining a data stream;
 obtaining a set of aggregate queries to be executed on the data stream; and
 generating a query plan for executing the set of aggregate queries on the data stream, wherein the generated query plan comprises identifying similar filters in two or more aggregate queries of the set of aggregate queries and combining the similar filters into a single filter such that the single filter is usable to prefilter data input to the two or more aggregate queries.
12. The method of claim 11, wherein the generated query plan further comprises generating other filters, wherein a first one of the other generated filters prefilters data prior to the data entering a second one of the other generated filters, and the second one of the other generated filters prefilters data prior to the data entering one or more of the set of aggregate queries.
13. An article of manufacture comprising a processorreadable storage medium storing one or more software programs which when executed by a processor perform the steps of the method of claim 11.
14. Apparatus, comprising:
 a memory; and
 a processor coupled to the memory and operative to: obtain a data stream; obtain a set of aggregate queries to be executed on the data stream; and generate a query plan for executing the set of aggregate queries on the data stream, wherein the generated query plan comprises at least one of: (i) generating at least one intermediate aggregate query, wherein the intermediate aggregate query combines a subset of aggregate queries from the set of aggregate queries so as to preaggregate data from the data stream prior to execution of the subset of aggregate queries such that the generated query plan is optimized for computational expense based on a given cost model; and (ii) identifying similar filters in two or more aggregate queries of the set of aggregate queries and combining the similar filters into a single filter such that the single filter is usable to prefilter data input to the two or more aggregate queries.
15. The apparatus of claim 14, wherein the intermediate aggregate query reduces the number of computations that would otherwise be required to generate results of the subset of aggregate queries.
16. The apparatus of claim 14, wherein the memory comprises main memory and the generated query plan for executing the set of aggregate queries for the data stream is substantially entirely executed using the main memory.
17. The apparatus of claim 14, wherein the generated query plan comprises a tree structure.
18. The apparatus of claim 17, wherein the query plan generating operation further comprises determining an optimal query plan with a lowest computation cost by determining a minimumcost aggregate tree.
19. The apparatus of claim 18, wherein the minimumcost aggregate tree is determined using a heuristic which performs one or more locallyoptimal modifications to the aggregate tree such that a maximum cost reduction is realized.
20. The apparatus of claim 18, wherein the minimumcost aggregate tree is determined using a heuristic which adds one or more random aggregate queries to the aggregate tree to form an expanded aggregate graph, and uses a directed steiner tree heuristic to find the minimumcost aggregate subtree of the expanded aggregate graph.
Type: Application
Filed: Jun 29, 2007
Publication Date: Jan 1, 2009
Patent Grant number: 8832073
Inventors: Kanthi C N (Bangalore), Naidu K V M (Bangalore), Rajeev Rastogi (Bangalore), Scott Satkin (Westfield, NJ)
Application Number: 11/770,926
International Classification: G06F 17/30 (20060101);