Time-constrained loop pipelining
Fermin Sanchez,Jordi Cortadella +1 more
- 01 Dec 1995
- pp 592-596
TL;DR: The results show that TCLP obtains optimal schedules in most cases.
read more
Abstract: This paper addresses the problem of Time-Constrained Loop Pipelining, i.e. given a fixed throughput, finding a schedule of a loop which minimizes resource requirements. This paper proposes a methodology, called TCLP, based on dividing the problem into two simpler and independent tasks: retiming and scheduling. TCLP explores different sets of resources, searching for a maximum resource utilization. This reduces area requirements. After a minimum set of resources has been found, the execution throughput is increased and the number of registers required by the loop schedule is reduced. TCLP attempts to generate a schedule which minimizes cost in time and area (resources and registers). The results show that TCLP obtains optimal schedules in most cases.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Flushing-Enabled Loop Pipelining for High-Level Synthesis
Steve Dai,Mingxing Tan,Kecheng Hao,Zhiru Zhang +3 more
- 01 Jun 2014
TL;DR: Novel techniques are proposed for synthesizing a conflict-aware flushing-enabled pipeline that is robust against potential resource collisions and conserving hardware resources and achieving near-optimal performance.
Automata-based symbolic scheduling
Steve Haynal,Forrest Brewer +1 more
- 01 Jan 2000
TL;DR: This dissertation presents a set of techniques for representing the high-level of a digital subsystem as a collection of nondeterministic finite automata, and demonstrates that techniques are scalable to practical problem sizes and complexities.
26
Realistic performance-constrained pipelining in high-level synthesis
Alex Kondratyev,Luciano Lavagno,Mike Meyer,Yosinori Watanabe +3 more
- 14 Mar 2011
TL;DR: An approach to pipelining in high-level synthesis that modifies the control/data flow graph before and after scheduling enables the direct re-use of a pre-existing, timing- and area-aware non-pipelined simultaneous scheduler and binder.
IMPACCT: Methodology and Tools for Power-Aware Embedded Systems
TL;DR: A new design tool framework called IMPACCT is proposed, which correctly combines the state-of-the-arttechniques at the system level, thereby saving even experienced designers from many pitfalls of system-level power management.
21
Automata-based symbolic scheduling for looping DFGs
Steve Haynal,Forrest Brewer +1 more
TL;DR: This paper presents an exact technique for scheduling looping data-flow graphs that implicitly supports functional pipelining and loop winding, and presents results for traditionally referenced benchmarks, several large synthetic benchmarks, and a practical industrial example.
20
References
A characterization of the minimum cycle mean in a digraph
TL;DR: A simple characterization of λ∗, as well as an algorithm for computing it efficiently, is given, which is called the minimum cycle mean.
863
Some scheduling techniques and an easily schedulable horizontal architecture for high performance scientific computing
B. R. Rau,C. D. Glaeser +1 more
- 01 Dec 1981
TL;DR: A novel architecture and a powerful scheduling algorithm that yields optimal and near-optimal schedules for iterative computations for horizontal architectures are developed.
455
Rotation Scheduling: A Loop Pipelining Algorithm
Liang-Fang Chao,Andrea S. LaPaugh,Edwin H.-M. Sha +2 more
- 01 Jul 1993
TL;DR: This work designs a novel and flexible technique, called rotation scheduling, for scheduling cyclic DFGs using loop pipelining, and provides a theoretical basis for the operations based on retiming.
200
Optimizing resource utilization using transformations
Miodrag Potkonjak,Jan M. Rabaey +1 more
- 11 Nov 1991
TL;DR: A novel learning while searching iterative improvement probabilistic algorithm has been developed and is used to resolve the associated NP-complete combinatorial optimization problem.
142