Scheduling and Resource Binding Algorithm Considering Timing Variation

Centralized controllers commonly used in high-level synthesis often cause long wires and high load capacitance and that is why critical paths typically occur on paths from controllers to data registers. However, conventional high level synthesis has focused on the delay of datapaths making it difficult to solve the timing closure problem during physical synthesis. This paper presents a hardware architecture with a distributed controller, which makes the timing closure problem much easier. It also presents a novel high-level synthesis flow for synthesizing such hardware through datapath partitioning and controller optimization. According to our experimental results, the proposed approach reduces the controller and interconnect delay by 20.3-27.4% and the entire critical path delay by 6.6~10.3% with 0.2~13.3% area overhead. Even without area overhead, it reduces the critical path delay by 5.8~10%.

Group signature is one of the main theme in recent digital signature studies. Typical signature algorithm is a combination of more than 70 elliptic curve (ECC), modular (RSA), long-bit integer and hash arithmetic functions. A full H/W IP core is strongly desired for the use of group signature in SoCs in slow-clock and low-power mobile devices and embedded systems. Flexible adjustment of H/W speed and size, depending on different systems and LSI process technologies, is also required. However, for designing and verifying H/W, the group signature algorithm is too complicated to use a standard RTL (Register Transfer Level) design methodology nor any recent HLS (High Level Synthesis). Therefore, we incorporated a two-level behavioral synthesis approach, where an optimized macro-architecture is explored by a custom-made scheduler, after a database of multiple number of microarchitectures are effectively constructed by conventional HLS. We implemented the signature algorithm on a low-cost 0.25um gate-array. The H/W size is approximately 1M gates and our chip can compute a group signature at the equivalent speed (0.135 seconds@100MHz clock) with 3GHz PC S/W, while the power consumption is two orders of magnitude lower (425mW@100MHz).

High-level synthesis operates on internal models known as control/data flow graphs (CDFG) and produces a register-transfer-level (RTL) model of the hardware implementation for a given schedule. For high-level synthesis to be efficient it has to estimate the effect that a given algorithmic decision (e.g., scheduling, allocation) will have on the final hardware implementation (after logic synthesis). Currently, this effect cannot be measured accurately because the CDFGs are very distinct from the RTL/gate-level models used by logic synthesis, precluding interaction between high-level and logic synthesis. This paper presents a solution to this problem consisting of a novel internal model for synthesis which spans the domains of high-level and logic synthesis. This model is an RTL/gate-level network capable of representing all possible schedules that a given behavior may assume. This representation allows high-level synthesis algorithms to be formulated as logic transformations and effectively interleaved with logic synthesis.

An inherent performance gap between custom designs and ASICs is one of the reasons why many designers still start their designs from register transfer level (RTL) description rather than from behavioral description, which can be synthesized to RTL via high-level synthesis (HLS). Sequencing overhead is one of the factors for this performance gap; the choice between latch and flip-flop is not typically taken into account during HLS, even though it affects all the steps of HLS. HLS-l is a new design framework that employs high-performance latches during scheduling, allocation, and controller synthesis. Its main feature is a new scheduler that is based on a concept of phase step (as opposed to conventional control step), which allows us scheduling in finer granularity, register allocation that resolves the conflict of latch being read and written at the same time, and controller synthesis that exploits dual-edge triggered storage elements to support phase step based scheduling. In experiments on benchmark designs implemented in 1.2 V, 65-nm CMOS technology, HLS-l reduced latency by 16.6% on average, with 9.5% less circuit area, compared to the designs produced by conventional HLS.

With the increasing cost of on-chip global communication, high-performance designs for data-intensive applications require architectures that distribute hardware resources (computing logic, memories, interconnect, etc.) throughout the chip, while restricting computations and communications to geographic proximities. In this paper, we present a methodology for high-level synthesis (HLS) of distributed logic-memory architectures, i.e., architectures that have logic and memory distributed across several partitions in a chip. Conventional HLS tools are capable of extracting parallelism from a behavior for architectures that assume a monolithic controller/datapath communicating with a memory or memory hierarchy. This paper provides techniques to extend the synthesis frontier to more general architectures that can extract both coarse and fine-grained parallelism from data accesses and computations in a synergistic manner. Our methodology selects many possible ways of organizing data and computations, carefully examines the tradeoffs (i.e., communication overheads, synchronization costs, area overheads) in choosing one solution over another, and utilizes conventional HLS techniques for intermediate steps. We have evaluated the proposed framework on several benchmarks by generating register-transfer level (RTL) implementations using an existing commercial HLS tool with and without our enhancements, and by subjecting the resulting RTL circuits to logic synthesis and layout. The results show that circuits designed as distributed logic-memory architectures using our framework achieve significant (up to 5.3/spl times/, average of 3.5/spl times/) performance improvements over well-optimized conventional designs with small area overheads (up to 19.3%, 15.1% on average). At the same time, the reduction in the energy-delay product is by an average of 5.9/spl times/ (up to 11.0/spl times/).

In recent years, there has been a lot of research into statistical static timing analysis (SSTA) to compute the critical path delay of a circuit under timing variation. In order to compute the true critical path delay, however, false paths that cannot be sensitized by any input vector must be identified first. Since SSTA is unable to capture the dynamic timing behavior of a circuit, it is completely blind to false paths, and thus it overestimates the circuit timing. In this paper, we propose a new concept of timing analysis approach called statistical dynamic timing analysis (SDTA), which is able to precisely express the statistical behavior of dynamic transitions at the output of gate into a compact form and to directly evaluate and propagate the expressions throughout the circuit, by which the false paths can be cleaned effectively. In addition, to be practical, we propose a couple of techniques that enable a fast computation of the SDTA. We tested the proposed approach on ISCAS benchmarks and carry skip adders under timing variation to show its accuracy in computing the distribution of the true critical path delay of a circuit. In summary, compared to the previous approach of false path-aware statistical timing analysis, our timing analysis technique is able to reduce the accuracy error in the mean and standard deviation of true critical path delay distribution from 9.8% to 1.9% and from 29.4% to -3.4%, respectively.

In this paper, we propose a calculation method of gate delay for SSTA (Statistical Static Timing Analysis) considering MIS (Multiple Input Switching). In SSTA, statistical maximum/minimum operation is necessary to calculate the latest/fastest arrival time of multiple input gate. Most SSTA approaches calculate the distribution in the latest/fastest arrival time under SIS (Single Input Switching assumption), resulting in ignoring the effect of MIS on the gate delay and the output transition time. MIS occurs when multiple inputs of a gate switch nearly simultaneously. Thus, ignoring MIS causes error in the statistical maximum/minimum operation in SSTA. We propose a statistical gate delay model considering MIS. We verify the proposed method by SPICE based Monte Carlo simulations. Experimental results show that the neglect of MIS effect leads to 80% error in worst case. The error of the proposed method is less than 20%.

A methodology for the register allocation problem in high-level synthesis based on the scanline sweep algorithm is introduced. The algorithm is computationally efficient and easy to understand and to implement, and it guarantees an optimal number of registers for nonpipelined designs. In this method, a set of registers is available from which the best one can be chosen by the allocator for variable binding. The method has proved to be useful in constructive allocation strategies in which the operators and variables are simultaneously considered for binding for interconnect cost optimization. The allocation program is written in C and it is integrated into the DAGAR high-level synthesis system. >

The authors present a solution to the interface timing problem in high-level synthesis by requiring that the algorithmic specification must completely determine the interface timing on the basis of cycles. They explain the timing problem and discuss the solution, which is closely related to a specific subset of the very high-speed IC description language (VHDL). This approach has been integrated into the high-level synthesis system CALLAS. An overview is given of the synthesis techniques of CALLAS. Controller reduction, which is the key transformation for satisfying the timing constraints, is discussed. The algorithms are formally described. The timing concept is compared to other approaches. Experimental results are provided. >

Conventional High-Level Synthesis (HLS) tools exploit parallelism mostly at the Instruction Level (ILP). They statically schedule the input specifications and build centralized Finite State Machine (FSM) controllers. However, aggressive exploitation of ILP in many applications has diminishing returns and, usually, centralized approaches do not efficiently exploit coarser parallelism, because FSMs are inherently serial. In this paper we present an HLS framework able to synthesize applications that, beside ILP, also expose Task Level Parallelism (TLP). An application can expose TLP through annotations that identify the parallel functions (i.e., tasks). To generate accelerators that efficiently execute concurrent tasks, we need to solve several issues: devise a mechanism to support concurrent execution flows, exploit memory parallelism, and manage synchronization. To support concurrent execution flows, we introduce a novel adaptive controller. The adaptive controller is composed of a set of interacting control elements that independently manage the execution of a task. These control elements check dependencies and resource constraints at runtime, enabling as soon as possible execution. To support parallel access to shared memories and synchronization, we integrate with a novel Hierarchical Memory Interface (HMI). With respect to previous solutions, the proposed interface supports multi-ported memories and atomic memory operations, which commonly occur in parallel programming. Our framework can generate the hardware implementation of C functions by employing two different approaches, depending on its characteristics. If a function exposes TLP, then the framework generates hardware implementations based on the adaptive controller. Otherwise, the framework implements the function through the FSM approach, which is optimized for ILP exploitation. We evaluate our framework on a set of parallel applications and show substantial performance improvements (average speedup of 4.7) with limited area overheads (average area increase of 5.48 times)

In this paper we discuss the algorithmic aspects of uncertainty driven scheduling which is a new design paradigm. Slack oriented design flow could be used to address the uncertainty problem in high level synthesis. We formalize the concept of slack and discuss different variations of the slack driven scheduling problem. The complexity issues are studied in detail and algorithms are proposed to solve the problem. These algorithms and proofs heavily exploit the concepts and techniques of graph theory and combinatorial optimization problems.

Background and Objectives: High-level synthesis (HLS) is one of the substantial steps in designing VLSI digital circuits. The primary purpose of HLS is to minimize the digital units used in the system to improve their power, delay, and area.Methods: In the modified MFO algorithm presented in this paper, a hyperbolic spiral is chosen as the update mechanism of moths. Also, by presenting a new approach, a paramount issue involved in applying meta-heuristic methods for solving HLS problems of VLSI circuits has been disentangled.Results: By comparing the performance of the proposed method with Genetic algorithm (GA)-based method and particle swarm optimization (PSO)-based method for the synthesis of the digital filters, it is concluded that the proposed method has the higher ability in the HLS of data path in digital filters. The best improvement is 2.78% for the delay (latency), 6.51% for the occupied area of the chip and 6.93% in power consumption. Another feature of the proposed method is its high-speed in finding optimal solutions, in a manner which, more than 21.6% and 12.9% faster than the GA-based and PSO-based methods, respectively on average.Conclusion: The most important very large scale integration (VLSI) circuits are digital filters and transformers, which are widely used in audio and video processing, medical signal processing, and telecommunication systems. The complex, expansive, and discrete nature of design space in high-level synthesis problems has made them one of the most difficult problems in VLSI circuit design.

Exploring lower bound of scheduling is an important problem in high-level synthesis, In this paper, we address the problem of computing lower bound on the general minimal resource interval scheduling problem, called n-n-MRIS, in which arbitrary component selections are allowed in stead of the traditional unicomponent selection. The problem of n-n-MRIS is proved to be strongly NP hard. An efficient ILP model and a surrogate relaxation technique are proposed to produce a lower bound for the n-n-MRIS problem.

With the advent of deep sub-micron era, there is a demand to consider the design closure problem in high-level synthesis. It is well known that the slack is an effective means of tolerating the uncertainties in operation delays. Previous work ever attempted to increase the usable slack based on a given initial schedule. Instead of the post-processing approach, this paper is the first attempt to the simultaneous application of operation scheduling and slack optimization. We use a 0-1 integer linear programming (0-1 ILP) approach to formally formulate the problem. Under the design constraints (timing and resource), our approach is applicable to two different objective functions: the maximization of the total usable slack and the maximization of the number of non-zero slack operations. Compared with previous work, our approach has the following two advantages: first, our approach guarantees the optimality; second, our approach is more suitable for the design space exploration.

A novel framework for retiming using evolutionary computation for high level synthesis of digital filters