Maximally and arbitrarily fast implementation of linear and feedback linear computations

Potkonjak, Miodrag; Rabaey, Jan M.

doi:10.1109/43.822618

Cited by 14 publications

(17 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We repeat this process until all inputs are included in the tree. Note that the maximally fast tree in [9] is a solution for a special case of this problem when all the inputs arrive at time 0. We can verify that the new algorithm constructs the tree with the same height as the maximally fast tree for the special case.…”

Section: Algorithm For Throughputmentioning

confidence: 99%

“…These computations may have multiplications and divisions between primary inputs, between primary inputs and algorithmic delays in feedback cycles, or between algorithmic delays, not in feedback cycles. The maximally fast approach for linear computation can be applied to this class of computations to result in arbitrary speed-up of throughput [2,9].…”

Section: Preliminariesmentioning

confidence: 99%

“…The first set of techniques is based on transformations [6,8,9]. For control-flow intensive computations, Lakshminarayana and Jha [4] proposed a set of techniques for throughput and power optimization.…”

Section: Related Workmentioning

confidence: 99%

“…A system can also exhibit linearity over either min or max in place of the addition, and addition in place of the multiplication with constants. Given arbitrary linear computations, the maximally fast technique [9], which combines several algebraic and redundancy manipulation transformations, reduces the critical path length to C = dlog2 N + 1 e + 1 , where N is the number of states. Combined with time-loop unfolding, the technique arbitrarily speeds up the linear computations such that the effective critical path length is reduced to C=k + 1 , where k is the unfolding factor.…”

Section: Preliminariesmentioning

confidence: 99%

“…Transformations alter the structure of a computation in such a way that the user specified input/output relationship is maintained. Numerous approaches have been proposed for throughput optimization using transformations, but most was restricted to limited classes of computations such as linear computations and even more restricted instances of linear computations [8,9]. They are inapplicable or ineffective when applied to large complex non-linear DSP and communication computations.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Throughput optimization of general non-linear computations

Hong¹,

Potkonjak²,

Guerra³

1999 IEEE/ACM International Conference on Computer-Aided Design. Digest of Technical Papers (Cat. No.99CH37051)

Self Cite

View full text Add to dashboard Cite

This paper addresses an optimal technique for throughput optimization of general non-linear data flow computations using a set of transformations. Throughput is widely recognized as the most important design metric of the modern DSP and communication applications. Numerous approaches have been proposed for throughput optimization, but most was restricted to limited classes of computations. They have limited effectiveness when applied to large complex non-linear DSP and communication computations. The new technique is used as an optimization engine in a divide-andconquer global approach for throughput optimization. We demonstrate the effectiveness of the new technique on numerous real-life non-linear designs.

show abstract

Section: Algorithm For Throughputmentioning

confidence: 99%

Section: Preliminariesmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Preliminariesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Throughput optimization of general non-linear computations

Hong¹,

Potkonjak²,

Guerra³

1999 IEEE/ACM International Conference on Computer-Aided Design. Digest of Technical Papers (Cat. No.99CH37051)

Self Cite

View full text Add to dashboard Cite

show abstract

Symbolic debugging of embedded hardware and software

Koushanfar

Kirovski

Hong

et al. 2001

IEEE Trans. Comput.-Aided Des. Integr. Circuits Syst.

Self Cite

View full text Add to dashboard Cite

Symbolic debuggers are system-development tools that can accelerate the validation speed of behavioral specifications by allowing a user to interact with an executing code at the source level. In response to a user query, the debugger must retrieve and display the value of a source variable in a manner consistent with user expectations with respect to the source statement where execution has halted. However, when a behavioral specification has been optimized using transformations, values of variables may either be inaccessible in the runtime state or inconsistent with user expectations. We address the problem that pertains to the retrieval of source values for the globally optimized behavioral specifications. We present a new approach for symbolic debugging. The implementation of the new debugging approach poses several optimization tasks. We formulate the optimization tasks and develop heuristics to solve them. We demonstrate the effectiveness of the proposed approach on a set of designs.

show abstract

Loop transformation methodology for fixed-rate video, image and telecom processing applications

Catthoor

Geurts

Man

Proceedings of IEEE International Conference on Application Specific Array Processors (ASSAP'94)

View full text Add to dashboard Cite

Many real-time signal pmessing applications are dominated by iterative loop constructs which exhibit a large amount of parallelism. In general, a realisation matched to the required mte of these applications exploits only a relatively small part of the parallelism available in the algorithm. This paper addresses the important problem of selecting the appropriate algorithmic-level decisions, in particular loop manipulations and the like, to arrive at an area-optimized specification for use in register-transfer level synthesis tools. One of the crucial cost factors in this optimisation is memory storage related. A n effective model and methodology are proposed to derive an optimized architecture with fully matched throughput, while avoiding a full traversal of the large search space. The effectiveness of our apprwch is substantiated with seveml realistic test cases. 1: IntroductionMany real-time signal processing (RSP) applications in video, image, speech and telecom processing are dominated by iterative loop constructs which exhibit a large amount of parallelism. Still, the loop organisation is not regular for most realistic algorithms in this target domain [l, 251. In general, a ked-rate realisation of these applications exploits only a relatively small part of the available parallelism. Hence, the main problem is not only to derive upper bounds on the amount of parallelism and on the achievable throughput [2, 21, 23, 16, 221, but especially to select a loop organisation matched to the required throughput. This important problem necessitates the traversal of a very large search space and has not been well researched yet. Only very recently, some research has addressed part of this problem. It is oriented mostly to good estimators to guide the search for feasible solutions under certain restrictions, combined with a hierarchical design space exploration This paper addresses the problem of selecting the appropriate loop manipulations on the initial behavioural description, to arrive at an area-optimized specification with matched throughput. The latter can then be used as input for register-transfer level synthesis tools, i.e. detailed allocation, assignment and scheduling. The cost function to steer these loop transformations should not only include the data-path area but it has to incorporates also the crucial foreground and background memory. Indeed, these storage related factors can be quite dominant over the other costs in multi-dimensional signal processing as in video, image and speech processing [I, 251. Up to now, they have been mostly neglected in the conventional high-level synthesis approaches. Also in the array synthesis community (see most P21. *Professor at the Katholieke Universiteit Leuven'This research has been partly sponsored by the E.C. projects ESPRIT-2260 (SPRITE) and HCM-ERBCHRXCT930382 (Vision algorithms and optical computer architectures).

show abstract

Maximally and arbitrarily fast implementation of linear and feedback linear computations

Cited by 14 publications

References 35 publications

Throughput optimization of general non-linear computations

Throughput optimization of general non-linear computations

Symbolic debugging of embedded hardware and software

Loop transformation methodology for fixed-rate video, image and telecom processing applications

Contact Info

Product

Resources

About