Data staging for results of analytics

• US 9,251,215 B2
• Filed: 01/14/2011
• Issued: 02/02/2016
• Est. Priority Date: 01/14/2011
• Status: Active Grant

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Inventors: Chen, Qiming, Hsu, Meichun
• Primary Examiner(s): Gorney, Boris
• Assistant Examiner(s): TRAN, BAO G
• Application Number:
  US13/007,467
• Publication Number:
  US 20120185439A1
• Time in Patent Office: 1,845 Days
• Field of Search: 707/661, 707/665, 707/667, 707/673, 707/756, 707/795, 370/351
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24568 Data stream processing; Continuous queries

Abstract

Data staging for results of analytics according to an example method includes maintaining current results from the analytics in a first data structure, the first data structure having a label identifying first data structure as a target for queries. The method also includes maintaining prior results from the analytics in at least one other data structure. The method also includes changing the label of the first data structure after a predetermined time. The method also includes assigning the label to one of the other data structures, wherein the label identifies the one of the other data structures as the target for queries.

Claims

1. A method of data staging for results of analytics, the method stored as machine readable instructions on a computer-readable media executable by a processor, the method comprising:

maintaining current results from the analytics in a first data structure, the first data structure labeled with a first label identifying the first data structure as a target for queries;

maintaining prior results from the analytics in at least one other data structure;

changing the first label of the first data structure to a second label after a predetermined time by incrementing based on time, wherein the labels represent a time boundary;

assigning the first label to one of the other data structures, wherein the first label then identifies the one of the other data structures as the target for queries;

archiving the other data structures according to a round-robin approach; and

switching into-relations for the target queries while queries are running.

2. The method of claim 1, wherein the analytics is continuous stream analytics (CSA).

3. The method of claim 1, further comprising archiving results in the one of the other data structures before assigning the label.

4. The method of claim 1, further comprising providing a stable interface for clients regardless of which of the data structures is the target for queries.

5. The method of claim 4, further comprising maintaining an index with each data structure and corresponding labels for the first data structure.

6. A system with data staging for results of analytics, comprising:

a first data structure for maintaining the results of the analytics, the first data structure labeled with a target metatag during a current time;

a plurality of other data structures operatively associated with the first data structure;

a host device operatively associated with the first data structure and the plurality of other data structures, the host device changing the target metatag from the first data structure to label one of the other data structures with the target metatag after a predetermined time by incrementing based on time, wherein the target metatags represent a time boundary, the host device switching into-relations for the target queries while queries are running and archiving the other data structures according to a round-robin approach.

7. The system of claim 6, wherein the client is a mobile device.

8. The system of claim 6, wherein the host device changes metatags for each of the other data structures after the predetermined time.

9. The system of claim 6, wherein an oldest data structure is archived before the predetermined time.

10. The system of claim 9, wherein the oldest data structure is assigned the target metatag after archiving.

11. The system of claim 6, wherein the metatags for each of the data structures is changed by incrementing based on time.

12. A system comprising:

a cloud service with data staging, the cloud service providing results of analytics to at least one client;

a first data structure labeled with a first target label, the first data structure operatively associated with the cloud service for maintaining current results from the analytics for a predetermined time;

a host device operatively associated with the first data structure and a plurality of other data structures, the host device changing the target label of the first data structure to another label and labeling one of the other data structures with the target label after the predetermined time by incrementing based on time, wherein the target label represents a time boundary, the host device switching into-relations for the target queries while queries are running and archiving the other data structures according to a round-robin approach.

13. The system of claim 12, wherein the target label provides a stable interface for clients accessing the current results from a plurality of data structures at different times.