Distributing and processing streams over one or more networks for on-the-fly schema evolution

• US 9,959,301 B2
• Filed: 07/25/2014
• Issued: 05/01/2018
• Est. Priority Date: 07/25/2014
• Status: Expired due to Fees

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Inventors: Krishnamurthy, Saileshwar
• Primary Examiner(s): Waldron, Scott A
• Assistant Examiner(s): MINA, FATIMA P
• Application Number:
  US14/341,676
• Publication Number:
  US 20160026663A1
• Time in Patent Office: 1,376 Days
• CPC Class Codes: G06F 16/2228 Indexing structures
• CPC Class Codes: G06F 16/24568 Data stream processing; Continuous queries
• CPC Class Codes: G06F 9/50 Allocation of resources, e.g. of the central processing unit [CPU]
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46

Abstract

In an embodiment, a method comprises receiving a first record comprising one or more first fields from a first source computer; in response to determining that the first record is a first type and a first node is associated with the first type, sending the first record to the first node to be processed; receiving a second record comprising the one or more first fields and one or more second fields from a second source computer; in response to determining that the second record is a second type and a second node is associated with the second type, sending the second record to the second node; in response to determining that the second type is a first subtype of the first type, sending the second record to the first node to be processed, without suspending the first node.

Claims

1. A computer system comprising:

a memory;

a processor coupled to the memory;

processor logic coupled to the processor and the memory configured to;

receive a first record comprising one or more first fields from a first source computer;

in response to determining that the first record is of a first type, send the first record to a first node computer that processes records of the first type but is suspended when processing records of a second type;

wherein records of the first type are organized according to a first schema and records of the second type are organized according to a second schema that is different from the first schema;

receive a second record comprising the one or more first fields and one or more second fields from a second source computer;

in response to determining that the second record is of the second type, send the second record to a second node computer that processes records of the second type;

after sending the first record to the first node computer and the second record to the second node computer, determine whether the second type is a first subtype of the first type, without suspending and reconfiguring the first node computer;

in response to determining that the second type is the first subtype of the first type, sending the second record to the first node computer to be processed, without suspending and reconfiguring the first node computer.

2. The computer system of claim 1, wherein the processor logic is configured to:

in response to determining that the first record is the first type and the first node computer is associated with the first type, send a first signature with the first record to the first node computer;

in response to determining that the second record is the second type and the second node computer is associated with the second type, send a second signature with the second record to the first node computer;

in response to determining that the second type is the first subtype of the first type, send the first signature with the second record to the first node computer.

3. The computer system of claim 1, wherein the processor logic is configured to:

organize the one or more first fields in the first record according to one or more first indices;

organize the one or more first fields in the second record according to the one or more first indices;

organize the one or more second fields in the second record according to one or more second indices, wherein each index in the one or more first indices references a memory address that is before the memory address referenced by each index in the one or more second indices.

4. The computer system of claim 1 comprising a third node computer comprising a third processor and a third memory configured to process one or more third type of records based on a third operation, wherein the processor logic is configured to:

receive, from a third source computer, a third record comprising the one or more first fields, the one or more second fields, and one or more third fields;

in response to determining that the third record is a third type and the third node computer is associated with the third type, send the third record to the third node computer to be processed;

in response to determining that the third type is a second subtype of the second type, send the third record to the second node computer to be processed without suspending and reconfiguring the second node computer;

in response to determining that the second type is the first subtype of the first type, send the third record to the first node computer to be processed without suspending and reconfiguring the second node computer.

5. The computer system of claim 1, wherein the processor logic is configured to:

determine whether the first record is the first type based on the one or more first fields included in the first record;

determine whether the second record is the second type based on the one or more first fields and the one or more second fields included in the second record;

determine whether the second type is the first subtype of the first type based on each field of the one or more first fields being included in the second record.

6. The computer system of claim 1, wherein the processor logic is configured to:

associate a first signature with the first type;

associate a second signature with the second type;

associate the second signature with the first signature as the first subtype;

determine the first record has the first signature based on one or more first field names of the one or more first fields, which are included in the first record;

determine whether the first record is the first type based on the first signature;

determine the second record has the second signature based on one or more second field names of the one or more first fields and the one or more second fields, which are included in the second record;

determine whether the second record is the second type based on the second signature;

determine whether the second type is the first subtype of the first type based on the second signature being associated with the first signature;

in response to determining that the second type is the first subtype of the first type, send the first signature with the second record to the first node computer.

7. The computer system of claim 1, wherein the processor logic is configured to:

associate a first signature with the first type;

associate a second signature with the second type;

associate the second signature with the first signature as the first subtype;

determine the first record has the first signature based on one or more first data structures in the one or more first fields, which are included in the first record;

determine whether the first record is the first type based on the first signature;

determine the second record has the second signature based on one or more second data structures in the one or more first fields and the one or more second fields, which are included in the second record;

determine whether the second record is the second type based on the second signature;

determine whether the second type is the first subtype of the first type based on the second signature being associated with the first signature;

in response to determining that the second type is the first subtype of the first type, send the first signature with the second record to the first node computer.

8. The computer system of claim 1, wherein the processor logic is configured to:

determine whether the first record is the first type based on whether the first record was received from a first source;

determine whether the second record is the second type based on whether the second record was received from a second source.

9. The computer system of claim 1, wherein:

the first node computer comprising a first processor and a first memory configured to process one or more first type of records based on a first operation;

the second node computer comprising a second processor and a second memory configured to process one or more second type of records based on a second operation;

the processor logic is configured to;

determine the second node computer has finished processing the second record;

send the second record to the first node computer after determining the second node computer has finished processing the second record.

10. The computer system of claim 9, wherein the processor logic is configured to:

send the second record to one or more sub-node computers associated with the second node computer in a particular order;

determine that each sub-node computer of the one or more sub-node computers has processed the second record in the particular order;

determine the second node computer is finished processing the second record comprises after determining that each sub-node computer of the one or more sub-node computers has processed the second record in the particular order.

11. A method comprising:

receiving a first record comprising one or more first fields from a first source computer;

in response to determining that the first record is of a first type, sending the first record to a first node computer, which is executed on a first computer, that processes records of the first type but is suspended when processing records of a second type;

wherein records of the first type are organized according to a first schema and records of the second type are organized according to a second schema that is different from the first schema;

receiving a second record comprising the one or more first fields and one or more second fields from a second source computer;

in response to determining that the second record is of the second type, sending the second record to a second node computer, which is executed on a second computer, that processes records of the second type;

after sending the first record to the first node computer and the second record to the second node computer, determining whether the second type is a first subtype of the first type, without suspending and reconfiguring the first node computer;

in response to determining that the second type is the first subtype of the first type, sending the second record to the first node computer to be processed, without suspending the first node computer;

wherein the method is performed by one or more computing devices.

12. The method of claim 11 comprising:

in response to determining that the first record is the first type and the first node computer is associated with the first type, sending a first signature with the first record to the first node computer;

in response to determining that the second record is the second type and the second node is associated with the second type, sending a second signature with the second record to the first node computer;

in response to determining that the second type is the first subtype of the first type, sending the first signature with the second record to the first node computer.

13. The method of claim 11 comprising:

organizing the one or more first fields in the first record according to one or more first indices;

organizing the one or more first fields in the second record according to the one or more first indices;

organizing the one or more second fields in the second record according to one or more second indices, wherein each index in the one or more first indices references a memory address that is before the memory address referenced by each index in the one or more second indices.

14. The method of claim 11 comprising:

receiving, from a third source computer, a third record comprising the one or more first fields, the one or more second fields, and one or more third fields;

in response to determining that the third record is a third type and a third node computer is associated with the third type, sending the third record to the third node computer, which is executed on a third computer, to be processed;

in response to determining that the third type is a second subtype of the second type, sending the third record to the second node computer to be processed without suspending the second node computer;

in response to determining that the second type is the first subtype of the first type, sending the third record to the first node computer to be processed without suspending the first node computer.

15. The method of claim 11 comprising:

determining whether the first record is the first type based on the one or more first fields included in the first record;

determining whether the second record is the second type based on the one or more first fields and the one or more second fields included in the second record;

determining whether the second type is the first subtype of the first type based on each field of the one or more first fields being included in the second record.

16. The method of claim 11 comprising:

associating a first signature with the first type;

associating a second signature with the second type;

associating the second signature with the first signature as the first subtype;

determining the first record has the first signature based on one or more first field names of the one or more first fields, which are included in the first record;

determining whether the first record is the first type based on the first signature;

determining the second record has the second signature based on one or more second field names of the one or more first fields and the one or more second fields, which are included in the second record;

determining whether the second record is the second type based on the second signature;

determining whether the second type is the first subtype of the first type based on the second signature being associated with the first signature;

in response to determining that the second type is the first subtype of the first type, sending the first signature with the second record to the first node computer.

17. The method of claim 11 comprising:

associating a first signature with the first type;

associating a second signature with the second type;

associating the second signature with the first signature as the first subtype;

determining the first record has the first signature based on one or more first data structures in the one or more first fields, which are included in the first record;

determining whether the first record is the first type based on the first signature;

determining the second record has the second signature based on one or more second data structures in the one or more first fields and the one or more second fields, which are included in the second record;

determining whether the second record is the second type based on the second signature;

determining whether the second type is the first subtype of the first type based on the second signature being associated with the first signature;

in response to determining that the second type is the first subtype of the first type, sending the first signature with the second record to the first node computer.

18. The method of claim 11 comprising:

determining whether the first record is the first type based on whether the first record was received from a first source;

determining whether the second record is the second type based on whether the second record was received from a second source.

19. The method of claim 11 comprising:

determining the second node computer has finished processing the second record;

wherein sending the second record to the first node computer is performed after determining the second node computer has finished processing the second record.

20. The method of claim 19 comprising:

sending the second record to one or more sub-node computers associated with the second node computer in a particular order;

wherein determining the second node computer is finished processing the second record comprises determining that each sub-node computer of the one or more sub-node computers has processed the second record in the particular order.