Memory estimation for high level synthesis

Verbauwhede, Ingrid; Scheers, C.; Rabaey, Jan M.

doi:10.1145/196244.196313

Cited by 51 publications

(36 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Table 1 summarizes the results of our experiments, carried out on a PC with a 1.85 GHz Athlon XP processor and 512 MB memory. The benchmarks used are: (1) a motion detection algorithm used in the transmission of real-time video signals on data networks [3]; (2) a real-time regularity detection algorithm used in robot vision; (3) Durbin's algorithm for solving Toeplitz systems with n unknowns [9]; (4) the kernel of a motion estimation algorithm for moving objects (MPEG-4); (5) a singular value decomposition (SVD) updating algorithm used in spatial division multiplex access (SDMA) modulation, in beamforming, and Kalman filtering; (6) the kernel of a voice coding application.…”

Section: Resultsmentioning

confidence: 99%

“…Verbauwhede et al consider a production axis for each array to model the relative production and consumption time (or date) of the individual array accesses [9]. The difference between these two dates equals the number of array elements produced between them, while the maximum difference gives the storage requirement for the considered array.…”

Section: Introductionmentioning

confidence: 99%

“…The maximal window size gives the storage requirement for the corresponding array. Treating the arrays separately, this technique (and [9], as well) does not consider the possibility of inter-array in-place mapping [3].…”

Section: Introductionmentioning

confidence: 99%

“…The authors employ a data dependence analysis similar to [1], their major improvement being to add the capability of taking into account available execution ordering information, based mainly on loop interchanges. Different from the previous works which are only approximate methods, this paper presents a non-scalar technique for computing exactly the minimum memory size in multi-dimensional signal processing algorithms, when the specifications are procedural, i.e., the execution ordering is induced by the loop structure and it is thus fixed (like in several previous works [9,10,7]). This assumption is based on the fact that in present industrial design, the design entry usually includes a full fixation of the execution ordering.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Computation of the Minimum Data Storage for Multi-Dimensional Signal Processing

Luican

Zhu

Balasa

2007

2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07

View full text Add to dashboard Cite

-The amount of the data storage in signal processing systems, whose behavior is described by loop-organized algorithmic specifications, has an important impact on the overall energy consumption, chip area, as well as system performance. This paper presents a non-scalar approach for computing the minimum storage requirements in high-level procedural specifications, where the main data structures are multi-dimensional arrays. This methodology uses both algebraic techniques specific to the data-flow analysis used in modern compilers and, also, more recent advances in the theory of polyhedra. In contrast with all the previous works which are only estimation methods, this approach can perform the exact computation of the minimum data storage even for applications with numerous loop nests and complex array references.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Computation of the Minimum Data Storage for Multi-Dimensional Signal Processing

Luican

Zhu

Balasa

2007

2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07

View full text Add to dashboard Cite

show abstract

“…In [30], a production time axis is created for each array. This models the relative production and consumption time, or date, of the individual array accesses.…”

mentioning

confidence: 99%

Computation of the minimum data storage and applications in memory management for multimedia signal processing

Luican

Zhu

Balasa

2008

ICA

View full text Add to dashboard Cite

Abstract. The amount of the data storage in signal processing systems, whose behavior is described by loop-organized algorithmic specifications, has an important impact on the overall energy consumption, chip area, as well as system performance. This paper presents a methodology based on lattices [25] which can be used to address several memory management tasks for applications with high-level specifications, where the main data structures are multidimensional arrays. This methodology was used in the past for the exact computation of the minimum data storage in applications with procedural, affine specifications [2]. The paper discusses two applications of that technique in the memory management of data-dominated signal processing systems: (1) the evaluation of the impact of loop transformations on the data storage, and (2) the assessment and efficient implementation of models of mapping multidimensional signals into the physical memory.

show abstract

Guidance of Loop Ordering for Reduced Memory Usage in Signal Processing Applications

Kjeldsberg

Catthoor

Verdoolaege

et al. 2008

J Sign Process Syst Sign Image Video Technol

View full text Add to dashboard Cite

Data dominated signal processing applications are typically described using large and multidimensional arrays and loop nests. The order of production and consumption of array elements in these loop nests has huge impact on the amount of memory required during execution. This is essential since the size and complexity of the memory hierarchy is the dominating factor for power, performance and chip size in these applications. This paper presents a number of guiding principles for the ordering of the dimensions in the loop nests. They enable the designer, or design tools, to find the optimal ordering of loop nest dimensions for individual data dependencies in the code. We prove the validity of the guiding principles when no prior restrictions are given regarding fixation of dimensions. If some dimensions are already fixed at given nest levels, this is taken into account when fixing the remaining dimensions. In most cases an optimal ordering is found for this situation as well. The guiding principles can be used in the early design phases in order to enable minimization of the memory requirement through in-place mapping. We use real life examples to show how they can be applied to reach a cost optimized end product. The results show orders of magnitude improvement in memory requirement compared to using the declared array sizes, and similar penalties for choosing the suboptimal ordering of loops when in-place mapping is exploited. AbbreviationsDP dependency part LR length ratio DV dependency vector ND nonspanning dimension DVP dependency vector polytope SD spanning dimension ID iteration domain UB upper bound LB lower bound 302 P.G. Kjeldsberg et al.

show abstract

Memory estimation for high level synthesis

Cited by 51 publications

References 15 publications

Computation of the Minimum Data Storage for Multi-Dimensional Signal Processing

Computation of the Minimum Data Storage for Multi-Dimensional Signal Processing

Computation of the minimum data storage and applications in memory management for multimedia signal processing

Guidance of Loop Ordering for Reduced Memory Usage in Signal Processing Applications

Contact Info

Product

Resources

About