US6086628A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Power-related hardware-software co-synthesis of heterogeneous distributed embedded systems
Assignee: Nokia of America Corp
•
Filing: 1998-02-17
•
Granted: 2000-07-11
•
Expires: 2018-02-17
Abstract
Abstract Hardware-software co-synthesis is the process of partitioning an embedded system specification into hardware and software modules to meet performance, power, and cost goals. Embedded systems are generally specified in terms of a set of acyclic task graphs. According to one embodiment of the present invention, a co-synthesis algorithm, called COSYN, starts with periodic task graphs with real-time constraints and produces a low-cost heterogeneous distributed embedded system architecture meeting these constraints. The algorithm has a pre-processing phase during which task graphs, system/task constraints, and a resource library for the embedded system are parsed, wherein the resource library has different PEs requiring different power supply voltages. The algorithm also has a synthesis phase, following the pre-processing phase, during which groups of tasks in the task groups are allocated to the PEs in the resource library and edges in the task graphs are allocated to communicate links in the resource library, based on performance evaluation of one or more possible allocations for each of the groups and edges in light of the system/task constraints, wherein (1) two or more PEs in the embedded system use two or more different power supply voltages (2) power dissipation is considered, along with cost and real-time constraints, during at least one of task grouping, group/edge allocation, and/or performance evaluation.
Inventors
- 👤 Bharat P. Dave
- 👤 Niraj K. Jha
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-08 02:54
CPC Classifications
G; G06; G06F; G06F30/00; G06F30/30; G06F2119/00; G06F2119/06
Claims Analysis
Total Claims: 20
Claim 1
Independent Claim
148 words
A method for designing the architecture of an embedded system, comprising: (a) a pre-processing phase comprising the step of parsing one or more task graphs, one or more system/task constraints, and a resource library for the embedded system, wherein the resource library has different PEs requiring different power supply voltages; and (b) a synthesis phase, following the pre-processing phase, comprising the step of allocating one or more groups of one or more tasks in the task graphs to one or more processing elements (PEs) in the resource library and allocating one or more edges in the task graphs to one or more communication links in the resource library, based on performance evaluation of one or more possible allocation for each of the groups and edges in light of the system/task constraints, wherein two or more PEs in the embedded system use two or more different power supply voltages.
Claim 2
Independent Claim
148 words
A method for designing the architecture of an embedded system, comprising: (a) a pre-processing phase comprising the step of parsing one or more task graphs, one or more system/task constraints, and a resource library for the embedded system, wherein the resource library has different PEs requiring different power supply voltages; and (b) a synthesis phase, following the pre-processing phase, comprising the step of allocating one or more groups of one or more tasks in the task graphs to one or more processing elements (PEs) in the resource library and allocating one or more edges in the task graphs to one or more communication links in the resource library, based on performance evaluation of one or more possible allocation for each of the groups and edges in light of the system/task constraints, wherein two or more PEs in the embedded system use two or more different power supply voltages.
Claim 3
Dependent Claim
17 words
The method of claim 1, wherein signal voltage level compatibility is checked for each communication link/PE interface.
Claim 4
Dependent Claim
17 words
The method of claim 1, wherein the embedded system comprises one or more voltage level translation buffers.
Claim 5
Dependent Claim
50 words
The method of claim 3, wherein power dissipation in a voltage level translation buffer is calculated considering its average quiescent power dissipation, frequency of operation of the corresponding communication link, and an activity factor of the corresponding signal, wherein the activity factor defines an effective transition rate of the signal.
Claim 6
Dependent Claim
14 words
The method of claim 1, wherein power requirements are estimated for multiple voltage levels.
Claim 7
Dependent Claim
13 words
An embedded system having an architecture generated using the method of claim 1.
Claim 8
Independent Claim
142 words
A method for designing the architecture of an embedded system, comprising: (a) a pre-processing phase comprising the step of parsing one or more task graphs, one or more system/task constraints, and a resource library for the embedded system; and (b) a synthesis phase, following the pre-processing phase, comprising the step of allocating one or more groups of one or more tasks in the task graphs to one or more processing elements (PEs) in the resource library and allocating one or more edges in the task graphs to one or more communication links in the resource library, based on performance evaluation of one or more possible allocation for each of the groups and edges in light of the system/task constraints, wherein power dissipation is considered, along with cost and real-time constraints, during at least one of task grouping, group/edge allocation, and/or performance evaluation.
Claim 9
Independent Claim
142 words
A method for designing the architecture of an embedded system, comprising: (a) a pre-processing phase comprising the step of parsing one or more task graphs, one or more system/task constraints, and a resource library for the embedded system; and (b) a synthesis phase, following the pre-processing phase, comprising the step of allocating one or more groups of one or more tasks in the task graphs to one or more processing elements (PEs) in the resource library and allocating one or more edges in the task graphs to one or more communication links in the resource library, based on performance evaluation of one or more possible allocation for each of the groups and edges in light of the system/task constraints, wherein power dissipation is considered, along with cost and real-time constraints, during at least one of task grouping, group/edge allocation, and/or performance evaluation.
Claim 10
Dependent Claim
40 words
The method of claim 7, wherein, during task grouping, deadline-based priority levels are used to select a task for grouping, and the selected task is then grouped with another task along which the selected task has the highest energy level.
Claim 11
Dependent Claim
22 words
The method of claim 8, wherein, after each task is grouped, energy levels of one or more remaining ungrouped tasks are recalculated.
Claim 12
Dependent Claim
21 words
The method of claim 7, wherein, during group/edge allocation, a voltage translation buffer is added to each communication link, if needed.
Claim 13
Dependent Claim
17 words
The method of claim 7, wherein during group/edge allocations are ordered based on increasing average power dissipation.
Claim 14
Dependent Claim
33 words
The method of claim 11, wherein, if two or more allocations have equal average power dissipation, then the allocation with the least dollar cost is chosen first among the two of more allocations.
Claim 15
Dependent Claim
37 words
The method of claim 12, wherein, if two or more allocations have equal average power dissipation and equal dollar cost, then the allocation with least peak power dissipation is chosen first among the two or more allocations.
Claim 16
Dependent Claim
12 words
The method of claim 7, wherein performance evaluation comprises architecture energy/power estimation.
Claim 17
Dependent Claim
23 words
The method of claim 14, wherein, during power estimation, power dissipation, execution time, and communication time of each task and edge are considered.
Claim 18
Dependent Claim
18 words
The method of claim 15, wherein quiescent power dissipation of each PE and each communication link are considered.
Claim 19
Dependent Claim
22 words
The method of claim 15, wherein the power dissipation during context switching of a task allocated to a general-purpose processor is considered.
Claim 20
Dependent Claim
13 words
An embedded system having an architecture generated using the method of claim 7.
References Cited (Backward Citations)
US4885684A
Cited
US5452461A
Cited
US5202987A
Cited
US5210872A
Cited
US5504670A
Cited
US5828568A
Cited
US5590323A
Cited
US5781787A
Cited
US5768594A
Cited
US5870588A
Cited
US5742821A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2003-07-11 | 2004-01-10 | 2004-07-11 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2007-07-11 | 2008-01-10 | 2008-07-11 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2011-07-11 | 2012-01-10 | 2012-07-11 | $7700.00 | Window Closed / Lapsed |