TAU '02Pub Date : 2002-12-02DOI: 10.1145/589411.589414
J. Dambre, D. Stroobandt, J. V. Campenhout
{"title":"A probabilistic approach to clock cycle prediction","authors":"J. Dambre, D. Stroobandt, J. V. Campenhout","doi":"10.1145/589411.589414","DOIUrl":"https://doi.org/10.1145/589411.589414","url":null,"abstract":"When developing new technologies, it is important to have an indication of the gain that can be achieved by exploring different research directions. Part of this gain is measured by achievable system performance. In this paper, we focus on the a priori prediction of clock speed as a measure for system performance.Previous approaches to clock cycle prediction were based on the summation of a number of (predicted) wire delays, equal to the maximum logic depth in a circuit. However, these methods do not consider the fact that the minimum clock cycle is determined by the largest combinatorial delay that occurs in a very complex and parallel interconnection topology. Indeed, in most circuits, there are a large number of paths with maximum or almost maximum logic depth. When implementing those circuits, any of these paths might become the path with maximal delay.In this paper, we present a new probabilistic model for predicting the maximal combinatorial path delay that partially captures the impact of the parallelism, present in real circuits. Our model is based on the distributions of the sum and of the maximum of a number of independent random variables. We experimentally validate our model using measured wire delay distributions.","PeriodicalId":338381,"journal":{"name":"TAU '02","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125809231","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
TAU '02Pub Date : 2002-12-02DOI: 10.1145/589411.589420
A. Efrati, Moshe Kleyner
{"title":"Timing analysis challenges for high speed CPUs at 90nm and below","authors":"A. Efrati, Moshe Kleyner","doi":"10.1145/589411.589420","DOIUrl":"https://doi.org/10.1145/589411.589420","url":null,"abstract":"Advances of the VLSI technology into the sub-90nm processes, enabling complex CPU designs that work at GHz frequencies pose numerous design and verification challenges.In this invited presentation, we focus on challenges in timing analysis of CPUs working at GHz speeds and sub-90nm processes.We start by brief overview of Timing Analysis tool used for intel CPUs and the \"shell\" timing models used for large blocks and how they integrate into full-chip model. Hierarchical timing is emphasized as key enabler for handling full-chip timing.Next, short-term challenges are presented:Xtalk impact on timingActive interconnectMixed abstraction, device to full-chipUse of domino as characterized cells.An intermediate accuracy model for Xtalk is introduced, called SMCF, which adaptively adjust equivalent MCF of attackers based on slope relationship and an empiric formula.We go into some detail with an example of timing checks that need to be applied for domino cells, which are different from checks applied on static cells.Finally some mid-term challenges are described:Multiple Input SwitchingProcess and environment variabilitySleep transistors.As variability is covered in more depth in other papers at this conference we mention it briefly but bring some examples of MIS and sleep transistor issues.","PeriodicalId":338381,"journal":{"name":"TAU '02","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126976221","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
TAU '02Pub Date : 2002-12-02DOI: 10.1145/589411.589413
L. Scheffer
{"title":"Explicit computation of performance as a function of process variation","authors":"L. Scheffer","doi":"10.1145/589411.589413","DOIUrl":"https://doi.org/10.1145/589411.589413","url":null,"abstract":"Each manufactured chip is a little bit different, and designers want as many as possible of these chips to work. Process variation is a function of many variables, as the width, thickness, and inter-layer thickness can vary independently for each layer on a chip, as can temperature and voltage. Currently designers cope with this by picking a few subsets of these conditions, called process corners, and analyzing at these conditions. However, it's easy to show this approach is both too conservative (the specified conditions will seldom occur) and not conservative enough (it misses errors that can occur due to process variation). We present a unified theory of process variation that includes inter-chip variation, intra-chip deterministic variation (such as caused by proximity effects and metal density), and intra-chip statistical variation. Using this mechanism, we can explicitly compute performance as a function of process variation. This allows us to compute less pessimistic timing numbers and address yield optimization in the design process.","PeriodicalId":338381,"journal":{"name":"TAU '02","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133658405","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
TAU '02Pub Date : 2002-12-02DOI: 10.1145/589411.589426
V. Rao, J. Soreff, Ravichander Ledalla, Fred L. Yang
{"title":"Aggressive crunching of extracted RC netlists","authors":"V. Rao, J. Soreff, Ravichander Ledalla, Fred L. Yang","doi":"10.1145/589411.589426","DOIUrl":"https://doi.org/10.1145/589411.589426","url":null,"abstract":"This paper presents a short-and-update technique for resistors (possibly connected to sinks) that can further crunch the RC network effectively after eliminating internal nodes [1,14]. Our method produces a realizable RC circuit and preserves the total capacitance in the network. While our technique cannot guarantee preserving the Elmore delay at each network sink node, the maximum delay error can be controlled by the user. Our method provides a smooth tradeoff between run time and delay accuracy, ranging from full retention of all resistors to complete elimination.","PeriodicalId":338381,"journal":{"name":"TAU '02","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130231311","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
TAU '02Pub Date : 2002-12-02DOI: 10.1145/589411.589430
Jun Chen, Lei He
{"title":"Determination of worst-case crosstalk noise for non-switching victims in GHz+ buses","authors":"Jun Chen, Lei He","doi":"10.1145/589411.589430","DOIUrl":"https://doi.org/10.1145/589411.589430","url":null,"abstract":"Considering RLC interconnect model and multiple switching aggressors, we study switching pattern generation and switching time alignment that leads to worst-case crosstalk noise for a quiet victim or a noisy one. We assume that aggressors can have arbitrary switching patterns and can switch at arbitrary times. We show that the commonly used superposition algorithm results in 15% underestimation on average, and propose a new algorithm that has virtually the same complexity as the superposition algorithm but approximates the exhaustive search very well with only 4% underestimation on average. Further, we show that applying RC model to GHz+ interconnects in IRTS 0.10um technology underestimates crosstalk noise by up to 80%, and convincingly conclude that RLC model is necessary to analyze such interconnects.","PeriodicalId":338381,"journal":{"name":"TAU '02","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128879425","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
TAU '02Pub Date : 2002-12-02DOI: 10.1145/589411.589423
Ali Dasdan
{"title":"Efficient algorithms for debugging timing constraint violations","authors":"Ali Dasdan","doi":"10.1145/589411.589423","DOIUrl":"https://doi.org/10.1145/589411.589423","url":null,"abstract":"A system of binary linear constraints or difference constraints (SDC) consists of a set of variables that are constrained by a set of unary or binary linear inequalities. In such diverse applications as scheduling, interface timing verification, real-time systems, multimedia systems, layout compaction, and constraint satisfaction, SDCs have successfully been used to model systems of both temporal and spatial constraints. Formally, an SDC is modeled by a weighted, directed graph called a constraint graph. The consistency (or feasibility) of an SDC means that there is at least one instantiation (or solution) of its variables that satisfies all its constraints. It is well known that the absence of positive cycles in a constraint graph implies the consistency of the corresponding SDC, so the consistency can be decided in strongly polynomial time. If the system is consistent, a solution can also be found in strongly polynomial time. However, if the system is inconsistent, there is no solution unless the system is repaired (or debugged). The debugging task is equivalent to freeing the corresponding constraint graph from all its positive cycles. All the previous algorithms for this task take time proportional to the number of positive cycles in the graph, which can grow exponentially. We have recently proposed a provably strongly polynomial-time algorithm for this task, i.e., an algorithm whose time complexity is polynomial in the size of the input constraint graph. In this paper, we propose extensions of this algorithm for different debugging scenarios. We theoretically and experimentally justify the efficiency and efficacy of our algorithms.","PeriodicalId":338381,"journal":{"name":"TAU '02","volume":"91 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125893853","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
TAU '02Pub Date : 2002-12-02DOI: 10.1145/589411.589429
Bhavana Thudi, D. Blaauw
{"title":"Efficient switching window computation for cross-talk noise","authors":"Bhavana Thudi, D. Blaauw","doi":"10.1145/589411.589429","DOIUrl":"https://doi.org/10.1145/589411.589429","url":null,"abstract":"In this paper, we present an efficient method for computing switching windows in the presence of delay noise. In static timing analysis, delay noise has traditionally been modeled using a simple switch-factor based noise model and the computation of switching windows is performed using an iterative algorithm where timing window propagation and switch factor updates are computed repeatedly until convergence. It was shown that the worst-case number of iterations required for convergence is O(n), where n is the number of gates in the circuit, resulting in an overall run time of O(n2). It was also shown that the iterations converge to different solutions, depending on the initial assumptions, making it unclear which solution is correct. In this paper, we show that the iterative nature of the problem is due to the switching-factor noise model and the order in which events are evaluated. Based on superposition model, we propose a time-sort based algorithm to compute the impact of delay noise on timing windows. We prove that the proposed algorithm has a run time that is linear with the circuit size. Since the algorithm is non-iterative and does not require initial assumptions, it eliminates the multiple solution problem. We tested the algorithm on a number of designs and show that it achieves significant speedup over the iterative approach.","PeriodicalId":338381,"journal":{"name":"TAU '02","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127839049","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
TAU '02Pub Date : 2002-12-02DOI: 10.1145/589411.589419
K. Keutzer, M. Orshansky
{"title":"From blind certainty to informed uncertainty","authors":"K. Keutzer, M. Orshansky","doi":"10.1145/589411.589419","DOIUrl":"https://doi.org/10.1145/589411.589419","url":null,"abstract":"The accuracy, computational efficiency, and reliability of static timing analysis have made it the workhorse for verifying the timing of synchronous digital integrated circuits for more than a decade. In this paper we charge that the traditional deterministic approach to analyzing the timing of circuits is significantly undermining its accuracy and may even challenge its reliability. We argue that computation of the static timing of a circuit requires a dramatic rethinking in order to continue serving its role as an enabler of high-performance designs. More fundamentally we believe that for circuits to be reliably designed the underlying probabilistic effects must be brought to the forefront of design and no longer hidden under conservative approximations. The reasons that justify such a radical transition are presented together with directions for solutions.","PeriodicalId":338381,"journal":{"name":"TAU '02","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127850230","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
TAU '02Pub Date : 2002-12-02DOI: 10.1145/589411.589416
Min Zhao, Kaushik Gala, V. Zolotov, Yuhong Fu, R. Panda, R. Ramkumar, B. K. Agrawal
{"title":"Worst case clock skew under power supply variations","authors":"Min Zhao, Kaushik Gala, V. Zolotov, Yuhong Fu, R. Panda, R. Ramkumar, B. K. Agrawal","doi":"10.1145/589411.589416","DOIUrl":"https://doi.org/10.1145/589411.589416","url":null,"abstract":"On-chip power distribution networks are resistive in nature and hence create large variations in the voltage levels across the chip. These variations have significant impact on the delays of global signals, such as the clock, which span the entire chip. A clock network that is balanced without considering delay variations induced by power supply variations can suffer significant degradation of its skew during the chips operation. In this paper, we describe a practical approach to determining the worst-case skew in a clock network in the presence of power supply variations. Experiments using the proposed approach on the clock nets of several processors show that clock skew can nearly double in the worst case. The proposed methodology is easily extendable for studying the impact of process variations and on-chip parasitic inductance on clock skew.","PeriodicalId":338381,"journal":{"name":"TAU '02","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133386608","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
TAU '02Pub Date : 2002-12-02DOI: 10.1145/589411.589428
H. Zhou
{"title":"Clock schedule verification with crosstalk","authors":"H. Zhou","doi":"10.1145/589411.589428","DOIUrl":"https://doi.org/10.1145/589411.589428","url":null,"abstract":"Delay variation due to crosstalk has made timing analysis a hard problem. In sequential circuits with transparent latches, crosstalk makes the clock schedule verification even harder. In this paper, we point out a false negative problem in current clock schedule verification techniques and propose a new approach based on switching windows. In this approach, coupling delay calculations are naturally combined with latch iterations. A novel algorithm is given for clock schedule verification in the presence of crosstalk and primary experiments show promising results.","PeriodicalId":338381,"journal":{"name":"TAU '02","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126030015","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}