Apparatus and method for processing a data stream

• US 9,378,244 B2
• Filed: 06/29/2010
• Issued: 06/28/2016
• Est. Priority Date: 12/11/2009
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Kim, Jeong-Ja, Jin, Yeong-Ho, Noh, Ye-Jin
• Primary Examiner(s): GORTAYO, DANGELINO N
• Application Number:
  US12/826,529
• Publication Number:
  US 20110145221A1
• Time in Patent Office: 2,191 Days
• Field of Search: 707/1, 707/2, 707/3, 707/718, 707/705, 707/708, 707/713, 707/720, 707/769, 707/802, 707/719, 707/759
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24539 using cached or materialised query results
• CPC Class Codes: G06F 16/2454 Optimisation of common expressions
• CPC Class Codes: G06F 16/24542 Plan optimisation
• CPC Class Codes: G06F 16/2471 Distributed queries

Abstract

An apparatus and method for processing data stream are provided. A query execution plan for processing data stream is shared between the server and the terminal. The server builds a query execution plan and determines a part of the built query execution plan to be executed in the terminal. The terminal executes the part of the built query execution plan and transfers a result of the execution. The server executes the remaining part of the built query execution plan based on the execution result transferred by the terminal. The part of the query execution plan to be executed in the terminal is determined depending on the performance of the terminal.

Claims

1. An apparatus for processing a data stream, the apparatus comprising:

a determination unit configured to;

determine a first part and a second part of a query execution plan shared between a server and a terminal, the first part comprising a part to be executed by the terminal served by the server, and the second part comprising a part to be executed by the server,

estimate an operation cost, the operation cost corresponding to a computing capacity of the terminal enabling execution of a query operator of the query execution plan by the terminal,

set a respective execution threshold for execution of the query operator by the terminal, corresponding to a maximum computing capacity enabling execution of the query operator by the terminal and being based on the estimating of the operation cost at the terminal; and

determine the first part based on the estimated operation cost;

a transmitter configured to transmit the first part to the terminal; and

a combining unit configured to receive a query execution result of the first part from the terminal and to combine the received query execution result with the second part,

wherein execution of the second part is based on an actual operation cost of executing the query operator at the terminal compared to the estimated operation cost.

2. The apparatus of claim 1, further comprising a terminal information storage unit configured to store terminal-related information comprising at least one of:

identification information of the terminal, Central Processing Unit (CPU) performance information of the terminal, a memory size of the terminal, and an access speed of the terminal to a network.

3. The apparatus of claim 2, wherein the operation cost is estimated based on the terminal-related information.

4. The apparatus of claim 2, wherein the combining unit is further configured to combine the query execution result corresponding to the first part with the second part based on the terminal-related information.

5. The apparatus of claim 1, wherein the query execution plan comprises:

at least one query operator; and

an execution order of the query operator.

6. The apparatus of claim 1, wherein the first part of the query execution plan transmitted by the apparatus to the terminal comprises a first set of query operators to be performed by the terminal and the second part of the query execution plan comprises a second set of query operators to be performed by the server.

7. An apparatus for processing a data stream, the apparatus comprising:

a receiver configured to receive a first part of a query execution plan shared between a server and a terminal from the server, the first part being determined as a part to be executed by the terminal served by the server based on an estimated operation cost; and

a partial query execution unit configured to execute the first part of the query execution plan and transmit a result of the execution to the server,

wherein the query execution plan comprises a query operator, an indication of a location within an execution sequence for executing the query operator, and the estimated operation cost, which corresponds to a computing capacity of the terminal enabling execution of the query operator of the query execution plan by the terminal,

wherein the first part is executed within a respective execution threshold for the query operator, set by the server and the corresponding to maximum computing capacity enabling execution of the query operator by the terminal, and based on the estimation of the operation cost at the terminal; and

wherein execution of a part of the query execution plan shared by the server is based on an actual operation cost of executing the query operator at the terminal compared to the estimated operation cost.

8. The apparatus of claim 7, wherein the partial query execution unit is further configured to transmit identification information of the terminal, together with the execution result, to the server.

9. A system for processing a data stream based on a query execution plan for queries, the system comprising:

a data sensing terminal that is served by a data stream management server, the data sensing terminal being configured to execute a part of the query execution plan that is determined by and received from the data stream management server, and in which the execution of the query execution plan is shared between the server and the data sensing terminal; and

the data stream management server configured to;

receive a result of the part of query execution plan from the data sensing terminal,

execute another part of the query execution plan based on the received result, and

estimate an operation cost of the data sensing terminal corresponding to a computing capacity of the data sensing terminal enabling execution of query operators of the query execution plan, and

set a respective execution threshold for the part of the execution plan transmitted to the data sensing terminal corresponding to a maximum computing capacity enabling execution of the part of the query execution plan transmitted to the data sensing terminal, based on the estimating of the operation cost at the terminal,

wherein execution of a part of the query execution plan shared by the server is based on an actual operation cost of executing the part of the query execution plan at the data sensing terminal compared to the estimated operation cost.

10. The system of claim 9, wherein the data stream management server is configured to:

determine the part of the query execution plan to be executed in the data sensing terminal based on the estimated operation cost, the part of the query execution plan to be executed in the data sensing terminal.

11. A method of processing a data stream, the method comprising:

at a server, determining a first part and a second part of a query execution plan shared between the server and a terminal, the first part comprising a part that is to be executed in the terminal that is served by the server and the second part comprising a part to be executed by the server,

wherein the determining of the first part and the second part comprises estimating an operation cost of the terminal corresponding to a computing capacity of the terminal enabling execution of at least one query operator of the query execution plan, setting a respective execution threshold for execution of query operators by the terminal corresponding to a maximum computing capacity enabling execution of the at least one query operator by the terminal, based on the estimating of the operation cost at the terminal, and determining the first part based on the estimated operation cost;

at the server, transmitting the first part to the terminal;

at the terminal, executing the first part and transmitting a result of the execution to the server; and

at the server, combining the result of execution of the first part with the second part,

wherein execution of the second part is based on an actual operation cost of executing the at lest one query operator at the terminal compared to the estimated operation cost.

12. The method of claim 11, wherein the operation cost is estimated based on at least one of:

identification information of the terminal, Central Processing Unit (CPU) performance information of the terminal, a memory size of the terminal, and an access speed of the terminal to a network.

13. The method of claim 11, wherein, in the combining, the execution result of the first part is combined with the second part based on identification information of the terminal.