IP Library › Granted Patent US 8,312,305
Granted Patent B2
US 8,312,305 · App. 12/700,513 · Granted Nov 13, 2012

Power management for systems on a chip

Assignee: International Business Machines Corporation
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 8,312,305
App. No.
12/700,513
Granted
Nov 13, 2012
Kind
B2
Abstract

A system for controlling a multitasking microprocessor system includes an interconnect, a plurality of processing units connected to the interconnect forming a single-source, single-sink flow network, wherein the plurality of processing units pass data between one another from the single-source to the single-sink, and a monitor connected to the interconnect for monitoring a portion of a resource consumed by each of the plurality of processing units and for controlling the plurality of processing units according to a predetermined budget for the resource to control a data overflow condition, wherein the monitor controls performance and power modes of the plurality of processing units.

Claims (24)

1. A system for controlling a multitasking microprocessor system, comprising:

an interconnect;

a plurality of processing units connected to the interconnect forming a single-source, single-sink flow network, wherein the plurality of processing units pass data between one another from the single-source to the single-sink; and

a monitor connected to the interconnect for monitoring a portion of a resource consumed by each of the plurality of processing units and for controlling the plurality of processing units according to a predetermined budget for the resource to control a data overflow condition, wherein the monitor controls performance and power modes of the plurality of processing units,

wherein the monitor performs an iterative max-flow method.

2. The system of claim 1 , further comprising:

a cache connected to the interconnect for receiving overflow data; and

a memory connected to the cache, wherein the monitor monitors reads and writes to a page of the memory.

3. The system of claim 1 , wherein the monitor performs the iterative max-flow method and improves a performance target given the predetermined budget.

4. The system of claim 3 , wherein the predetermined budget includes one of a power, bandwidth and temperature budget.

5. The system of claim 1 , wherein at least two of the processing units are arranged in series.

6. The system of claim 1 , wherein at least two of the processing units are arranged in parallel.

7. A system for controlling a multitasking microprocessor system, comprising:

an interconnect;

a plurality of processing units connected to the interconnect forming a single-source, single-sink flow network, wherein the plurality of processing units pass data between one another from the single-source to the single-sink; and

a monitor connected to the interconnect for monitoring a portion of a resource consumed by each of the plurality of processing units and for controlling the plurality of processing units according to a predetermined budget for the resource to control a data overflow condition, wherein the monitor controls performance and power modes of the plurality of processing units,

wherein the monitor performs an iterative min-flow method.

8. The system of claim 7 , further comprising:

a cache connected to the interconnect for receiving overflow data; and

a memory connected to the cache, wherein the monitor monitors reads and writes to a page of the memory.

9. The system of claim 7 , wherein the monitor performs the iterative min-flow method and improves a performance target given the predetermined budget.

10. The system of claim 9 , wherein the predetermined budget includes one of a power, bandwidth and temperature budget.

11. The system of claim 7 , wherein at least two of the processing units are arranged in series.

12. The system of claim 7 , wherein at least two of the processing units are arranged in parallel.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 4, 2010
From: CHER, CHEN-YONG; KARKHANIS, TEJAS S.; RAMANI, SRINIVASAN
To: INTERNATIONAL BUSINESS MACHINES CORPORATION
Reel/Frame 023900/0712 →
Continuity (1)
Related Publication 20110191603A1 · Aug 4, 2011