DEDUPLICATED DATA PROCESSING CONGESTION CONTROL

• US 20130080409A1
• Filed: 11/09/2012
• Published: 03/28/2013
• Est. Priority Date: 05/04/2010
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: AKIRAV, Shay H., BACHAR, Yariv, LEVY, Asaf, MEIRI, Ehud
• Granted Patent: US 8,891,473 B2
• Time in Patent Office: Days
• US Class Current: 707/692
• CPC Class Codes: G06F 16/90 Details of database functions independent of the retrieved data types
• CPC Class Codes: H04L 47/122 by diverting traffic away from congested entities
• CPC Class Codes: H04L 67/1095 Replication or mirroring of data, e.g. scheduling or transport for data synchronisation between network nodes
• CPC Class Codes: H04L 67/1097 for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]

Abstract

Various embodiments for deduplicated data processing congestion control in a computing environment are provided. In one such embodiment, a congestion target setpoint is calculated using one of a proportional constant, an integral constant, and a derivative constant, wherein the congestion target setpoint is a virtual dimension setpoint. A single congestion metric is determined from a sampling of a plurality of combined deduplicated data processing congestion statistics in a number of active deduplicated data processes. A congestion limit is calculated from a comparison of the single congestion metric to the congestion target setpoint, the congestion limit being a manipulated variable. The congestion limit is compared to the number of active deduplicated data processes. If the number of active deduplicated data processes are less than the congestion limit, a new deduplicated data process of the number of active deduplicated data processes is spawned.

Claims

1. A method for deduplicated data processing congestion control using at least one processor device in a computing environment, the method comprising:

calculating a congestion target setpoint using one of a proportional constant, an integral constant, and a derivative constant, wherein the congestion target setpoint is a virtual dimension setpoint;

determining a single congestion metric from a sampling of a plurality of combined deduplicated data processing congestion statistics in a number of active deduplicated data processes;

calculating a congestion limit from a comparison of the single congestion metric to the congestion target setpoint, the congestion limit being a manipulated variable; and

comparing the congestion limit to the number of active deduplicated data processes, wherein if the number of active deduplicated data processes are less than the congestion limit, a new deduplicated data process of the number of active deduplicated data processes is spawned.

2. The method of claim 1, wherein the proportional constant represents a congestion error of a present deduplicated data processing activity, the integral constant indicates historical congestion errors of deduplicated data process activities, and the derivative constant forecasts a deduplicated data congestion error pattern trend.

3. The method of claim 1, further including, categorizing each of the number of active deduplicated data processes by a plurality of deduplicated data process types, and comparing the congestion limit to the number of active deduplicated data processes for each of the plurality of deduplicated data process types.

4. The method of claim 3, further including measuring the average response time of each of the number of active deduplicated data processes over a window of time, the window of time being a measurement of at least one of a normalized latency time period and a combination of a sum, an average, and a maximum time period.

5. The method of claim 4, further including normalizing each average response time and combining each average response time with times from other deduplicated data process types.

6. The method of claim 3, further comprising assigning a scalable congestion range to each of the deduplicated data process types from a low range of no congestion to a high range of heavy congestion.

7. A system for deduplicated data processing congestion control in a computing environment, comprising:

at least one processor device operable in the computing environment, wherein the at least one processor device is adapted for;

calculating a congestion target setpoint using one of a proportional constant, an integral constant, and a derivative constant, wherein the congestion target setpoint is a virtual dimension setpoint;

determining a single congestion metric from a sampling of a plurality of combined deduplicated data processing congestion statistics in a number of active deduplicated data processes;

calculating a congestion limit from a comparison of the single congestion metric to the congestion target setpoint, the congestion limit being a manipulated variable; and

comparing the congestion limit to the number of active deduplicated data processes, wherein if the number of active deduplicated data processes are less than the congestion limit, a new deduplicated data process of the number of active deduplicated data processes is spawned.

8. The system of claim 7, wherein the proportional constant represents a congestion error of a present deduplicated data processing activity, the integral constant indicates historical congestion errors of deduplicated data process activities, and the derivative constant forecasts a deduplicated data congestion error pattern trend.

9. The system of claim 7, wherein the at least one processor device is adapted for categorizing each of the number of active deduplicated data processes by a plurality of deduplicated data process types, and comparing the congestion limit to the number of active deduplicated data processes for each of the plurality of deduplicated data process types.

10. The system of claim 9, wherein the at least one processor device is adapted for measuring the average response time of each of the number of active deduplicated data processes over a window of time, the window of time being a measurement of at least one of a normalized latency time period and a combination of a sum, an average, and a maximum time period.

11. The system of claim 10, wherein the at least one processor device is adapted for normalizing each average response time and combining each average response time with times from other deduplicated data process types.

12. The system of claim 9, wherein the at least one processor device is adapted for assigning a scalable congestion range to each of the deduplicated data process types from a low range of no congestion to a high range of heavy congestion.

13. A computer program product for deduplicated data processing congestion control using at least one processor device in a computing environment, the computer program product comprising a computer-readable storage medium having computer-readable program code portions stored therein, the computer-readable program code portions comprising:

computer code for at least one processor device operable in the computing environment, wherein the at least one processor device is adapted for;

computer code for calculating a congestion target setpoint using one of a proportional constant, an integral constant, and a derivative constant, wherein the congestion target setpoint is a virtual dimension setpoint;

computer code for determining a single congestion metric from a sampling of a plurality of combined deduplicated data processing congestion statistics in a number of active deduplicated data processes;

computer code for calculating a congestion limit from a comparison of the single congestion metric to the congestion target setpoint, the congestion limit being a manipulated variable; and

computer code for comparing the congestion limit to the number of active deduplicated data processes, wherein if the number of active deduplicated data processes are less than the congestion limit, a new deduplicated data process of the number of active deduplicated data processes is spawned.

14. The program product of claim 13, wherein the proportional constant represents a congestion error of a present deduplicated data processing activity, the integral constant indicates historical congestion errors of deduplicated data process activities, and the derivative constant forecasts a deduplicated data congestion error pattern trend.

15. The program product of claim 13, further comprising computer code for categorizing each of the number of active deduplicated data processes by a plurality of deduplicated data process types, and comparing the congestion limit to the number of active deduplicated data processes for each of the plurality of deduplicated data process types.

16. The program product of claim 15, further comprising computer code for measuring the average response time of each of the number of active deduplicated data processes over a window of time, the window of time being a measurement of at least one of a normalized latency time period and a combination of a sum, an average, and a maximum time period.

17. The program product of claim 16, further comprising computer code for normalizing each average response time and combining each average response time with times from other deduplicated data process types.

18. The program product of claim 15, further comprising computer code for assigning a scalable congestion range to each of the deduplicated data process types from a low range of no congestion to a high range of heavy congestion.