Pre-Grant Publication Number: 20070174746
Please help the USPTO examine the application by evaluating the relevance of the publicly submitted prior art to the patent application.
Peer-to-Patent forwards the Top 10 most relevant prior art submissions and their annotations to the United States Patent and Trademark Office.
Review this prior art and click on the thumbs up (or down) to indicate whether this submission should be forwarded to the USPTO.
If you login then you can add an annotation by typing in the box at the bottom of the screen to comment on the relevance of the prior art to the claims of the patent application.
Review this prior art and click on the thumbs up (or down) to indicate whether this submission should be forwarded to the USPTO.
If you login then you can add an annotation by typing in the box at the bottom of the screen to comment on the relevance of the prior art to the claims of the patent application.
Prior Art Detail
Summary / Description
| Summary / Description | A codesign methodology incorporates timing speculation into a low-power micropocessor pipeline and shaves energy levels far below the point permitted by worst-case computation paths |
Basic Information
| Type of Prior Art | Print Publication |
| Publication Title * | Computer |
| Author | Austin, Todd et al. |
| ISBN | |
| Page Range | 57 - 65 |
| Medium | Journal article |
| Publication Date * | March 2004 |
| URL | |
Notes / To Do
| Notes | |
Excerpt
Excerpt Self-tuning systems:
In its current form, Razor sets voltage globally—
chip-wide, but we could refine it to allow distributed voltage control. Under a distributed control system, each processor pipeline stage could operate at a separate, potentially different voltage determined by monitoring pipeline stage error rates. Releasing the constraint of a single operating voltage enables significant optimizations
of voltage assignments across the processor stages, leading to further power savings.
|
Relevance
Claims
1
A method for software execution, comprising:
varying core voltages of plural processors operating in lockstep to determine an operating range for each of the plural processors; and
adjusting the core voltages of the plural processors within the operating range to tune the plural processors.
Relevance
Figures 2 and 3 show a plurality of processors operating in lockstep with error rates determined manually as a function of voltage. Page 63 teaches that the voltages of multiple parallel processors could be individually tuned based on a monitored error rate.
Figures 2 and 3 show a plurality of processors operating in lockstep with error rates determined manually as a function of voltage. Page 63 teaches that the voltages of multiple parallel processors could be individually tuned based on a monitored error rate.
Claim Chart
All
2
The method of Claim 1 further comprising:
determining a first voltage value where the plural processors exhibit anomalies while executing instructions; and
determining a second voltage value where the plural processors exhibit failures while executing instructions.
Relevance
Figure 2 shows voltages ranged from low error rates to 100% error rates.
Figure 2 shows voltages ranged from low error rates to 100% error rates.
Claim Chart
All
8
A computer readable medium having instructions for causing a computer to execute a method, comprising:
lowering a core voltage of one of two processors executing instructions in synchronization to determine a first voltage value;
raising the core voltage of the one processor to determine a second voltage value; and
using the first and second voltage values to tune the core voltage of the one processor to increase performance of the two processors.
Relevance
See claim 1
See claim 1
Claim Chart
All
14
A computer system, comprising:
memory for storing an algorithm; and
a processor for executing the algorithm to:
raise and lower core voltages of plural processors operating in lockstep to determine an operating range having no anomalies for each of the plural processors; and
change the core voltages of the plural processors within the operating range to tune the plural processors.
Relevance
See claim 1
See claim 1
Claim Chart
All
