Superscalar Research Articles

Advanced architecture optimisation and performance analysis of a reconfigurable grid ALU processor

In the billion transistor era only a few architectural approaches propose new paths to improve the execution of conventional sequential instruction streams. Many legacy applications could profit from processors that are able to speed-up the execution of sequential applications beyond the performance of current superscalar processors. The Grid arithmetic logic unit (ALU) Processor (GAP) accelerates conventional sequential instruction streams without the need for recompilation. The GAP comprises a processor front-end similar to that of a superscalar processor extended by a configuration unit and a two-dimensional array of functional units that forms the execution unit. Instruction sequences are mapped dynamically into the array by the configuration unit so that they form the dataflow graph of the sequence. This study shows a performance evaluation of the GAP architecture with different array dimensions as well as its performance using a simplified interconnection network. GAP outperforms an out-of-order superscalar processor by a maximum of factor 2 with a complete crossbar interconnect between two array rows. Reducing the interconnection network to the minimum shows a maximum performance drawback of 10% for only a particular configuration and a single benchmark. In general, the slowdown is less than 2% for the minimum interconnect (two buses) and about 0.02% if three interconnection buses are used.

Empirical model for cooperative resizing of processor structures to exploit power-performance efficiency at runtime

Power consumption has become a major cause of concern spanning from data centres to handheld devices. Traditionally, improvement in power-performance efficiency of a modern superscalar processor came from technology scaling. However, that is no longer the case. Many of the current systems deploy coarse grain voltage and/or frequency scaling for power management. These techniques are attractive, but limited because of their granularity of control and effectiveness in nano-complementary metal-oxide-semiconductor (CMOS) technologies. This study proposes a novel architecture-level mechanism to exploit intra-thread variations for power-performance efficiency in modern superscalar processors. This class of processors implement several buffer/queue structures to support speculative out-of-order execution for performance enhancement. Applications may not need full capabilities of such structures at all times. A mechanism that collaboratively adapts a finite set of key hardware structures to the changing programme behaviour can allow the processor to operate with heterogeneous power-performance capabilities. This study presents a novel offline regression-based empirical model to estimate structure resizing for a selected set of structures. It is shown that using a few processor runtime events, the system can dynamically estimate structure resizing to exploit power-performance efficiency. Results show that using the proposed empirical model, a selective set of key structures can be resized at runtime to deliver on average 40% power-performance efficiency over a baseline design, with only 5% loss of performance.

Disjoint out-of-order execution processor

High-performance superscalar architectures used to exploit instruction level parallelism in single-thread applications have become too complex and power hungry for the multicore processors era. We propose a new architecture that uses multiple small latency-tolerant out-of-order cores to improve single-thread performance. Improving single-thread performance with multiple small out-of-order cores allows designers to place more of these cores on the same die. Consequently, emerging highly parallel applications can take full advantage of the multicore parallel hardware without sacrificing performance of inherently serial and hard to parallelize applications. Our architecture combines speculative multithreading (SpMT) with checkpoint recovery and continual flow pipeline architectures. It splits single-thread program execution into disjoint control and data threads that execute concurrently on multiple cooperating small and latency-tolerant out-of-order cores. Hence we call this style of execution Disjoint Out-of-Order Execution (DOE). DOE uses latency tolerance to overcome performance issues of SpMT caused by interthread data dependences. To evaluate this architecture, we have developed a microarchitecture performance model of DOE based on PTLSim, a simulation infrastructure of the x86 instruction set architecture. We evaluate the potential performance of DOE processor architecture using a simple heuristic to fork control independent threads in hardware at the target addresses of future procedure return instructions. Using applications from SpecInt 2000, we study DOE under ideal as well as realistic architectural constraints. We discuss the performance impact of key DOE architecture and application variables such as number of cores, interthread data dependences, intercore data communication delay, buffers capacity, and branch mispredictions. Without any DOE specific compiler optimizations, our results show that DOE outperforms conventional SpMT architectures by 15%, on average. We also show that DOE with four small cores can perform on average equally well to a large superscalar core, consuming about the same power. Most importantly, DOE improves throughput performance by a significant amount over a large superscalar core, up to 2.5 times, when running multitasking applications.

Multi-objective optimisations for a superscalar architecture with selective value prediction

This work extends an earlier manual design space ex ploration of our developed Selective Load Value Pre diction based superscalar architecture to the L2 unified cache. A fter that we perform an automatic design space expl oration using a special developed software tool by varying several architectural parameters. Our goal is to find optim al configurations in terms of CPI (Cycles per Instruction) and energy consumption. By varying 19 architectural parameter s, as we proposed, the design space is over 2.5 millions of billions configurations which obviously means that only heuristic search can be considered. Therefore, we propose dif ferent methods of automatic design space exploratio n based on our developed FADSE tool which allow us to evaluate only 2500 configurations of the above mentioned huge design space! The experimental results show that our automatic de sign space exploration (DSE) provides significantly better configurations than our previous manual DSE approach, considering the proposed multi-objective approac h.

멀티코어 프로세서의 명령어 자취형 모의실험에 대한 연구

In order to overcome the complexity and power problems of superscalar processors, the multi-core architecture has been prevalent recently. Although the execution-driven simulation is wide spread, the trace-driven simulation has speed advantages over the execution-driven simulation. We present a methodology to simulate multi-core architecture using trace-driven simulator. Using SPEC 2000 benchmarks as input, the trace-driven simulation has been performed for the cores ranging from 2 to 16 extensively. As a result, the 16-core processor resulted in 4.1 IPC and 13.3 times speed up over single-core processor on the average.

A hyperscalar dual-core architecture for embedded systems

This paper proposes a lightweight reconfigurable dual-core architecture for embedded systems, called hyperscalar dual-core architecture. The proposed architecture can play three different roles (a 2-issue statically scheduled superscalar processor, a homogeneous dual-core processor, or a standalone single-core processor), allowing embedded systems to accommodate diverse workloads. The proposed design uses four extended instructions to enable programmers to dynamically switch the roles of the proposed architecture. This paper also presents an instruction analyzer (IA) that connects two scalar in-order cores to handle role switching. The design of IA makes it possible for the two cores to work together like a 2-issue statically scheduled superscalar processor. Based on the proposed dispatching rules, the IA dispatches instructions with data dependencies to the same core. Since two cores act like a statically scheduled superscalar processor, they can resolve data dependencies using existing forwarding paths without introducing the high-area-cost inter-core operand-switching crossbars. Simulation results show that when the proposed architecture works in a statically scheduled superscalar manner, it achieves a 26% higher instructions per cycle (IPC) averaged across all 29 benchmarks from the MiBench suite than a scalar in-order core. The increases in area and power to extend a homogeneous dual-core processor to a hyperscalar dual-core processor are only 1.8% and 1.75%, respectively, using 90nm CMOS technology.

Delay Evaluation of Issue Queue in Superscalar Processors with Banking Tag RAM and Correct Critical Path Identification

Delay Evaluation of Issue Queue in Superscalar Processors with Banking Tag RAM and Correct Critical Path Identification

Exploring branch target buffer access filtering for low-energy and high-performance microarchitectures

Powerful branch predictors along with a large branch target buffer (BTB) are employed in superscalar and simultaneous multi-threading (SMT) processors for instruction-level parallelism and thread-level parallelism exploitation. However, the large BTB not only dominates the predictor energy consumption, but also becomes a major roadblock in achieving faster clock frequencies at deep sub-micron technologies. The authors propose here a filtering scheme to dramatically reduce the accesses to the BTB to achieve significantly reduced energy consumption in the BTB while maintaining the performance. For a simulated superscalar microprocessor, the experimental evaluation shows that the BTB access filtering (BAF) design achieves an 88.5% dynamic energy reduction with negligible performance loss. The authors also study the leakage behaviour and its control in the BAF design. The results show that by applying a drowsy strategy, very effective leakage control can be achieved. For the high-performance design, the BAF can also improve BTB's performance scalability at new technologies. For the simultaneous multi-threading environment, the authors evaluate the effectiveness of the BAF design and propose a banked BAF (BK-BAF) scheme to further reduce the energy consumption and performance overhead. The experimental results confirm that the BK-BAF scheme can be an energy/performance-effective design for next generation SMT processors.

SYRANT: SYmmetric resource allocation on not-taken and taken paths

In the multicore era, achieving ultimate single process performance is still an issue e.g. for single process workload or for sequential sections in parallel applications. Unfortunately, despite tremendous research effort on branch prediction, substantial performance potential is still wasted due to branch mispredictions. On a branch misprediction resolution, instruction treatment on the wrong path is essentially thrown away. However, in most cases after a conditional branch, the taken and the not-taken paths of execution merge after a few instructions. Instructions that follow the reconvergence point are executed whatever the branch outcome is. We present SYRANT (SYmmetric Resource Allocation on Not-taken and Taken paths), a new technique for exploiting control independence. SYRANT essentially uses the same pipeline structure as a conventional processor. SYRANT tries to enforce the allocation of the exact same resources on the out-of-order execution mechanism (physical register, load/store queue and reorder buffer) for both the taken and not-taken paths. Thus, on a branch misprediction, the result of an instruction already executed on the wrong path after the reconvergence point can be conserved in the same structure when it is data independent. Adding SYRANT on top of an aggressive superscalar execution core allows to improve performance for applications suffering a significant branch misprediction rate. As a side, but important extra contribution, we introduce ABL/SBL a simple and non-intrusive hardware reconvergence detection mechanism. ABL/SBL can be used in a conventional superscalar processor to improve branch prediction accuracy by exploiting the execution of branches along the wrong path.

Energy savings in simultaneous multi-threaded processors through dynamic resizing of datapath resources

Nowadays, all the designers of systems from high-performance servers to battery-operated handheld devices aim for reliability, high-performance and longevity. Central within these aims the issue of processor power consumption is becoming increasingly important. In this study, we aim to adapt our already-proven method for single-threaded superscalar processors to simultaneous multi-threaded (SMT) processors for energy savings. The original method focused on resizing datapath resources according to the demands of running applications. To achieve this, the targeted resources are physically divided into multiple partitions, and turned on and off according to the needs of the applications. Since, the energy consumption of the turned-off datapath resources is quite low, as a result, it becomes possible to have great amount of energy savings within a processor. However, special care must be taken when there are multiple threads racing against each other to gain access to shared datapath resources. As a result, our proposed microarchitectural technique achieves 0.5% Instructions Per Cycle (IPC) and 3.2% Total number of instructions Per Cycle (TPC) improvement, while it turns off 45% of the Reorder Buffer (ROB), 59% of the Load-Store Queue (LSQ), 43% of the Issue Queue (IQ), 30% of the integer Physical Register Files (PRF) and, finally, 48% of the floating PRF, on the average across all simulated benchmarks. According to our estimates, the total processor power is reduced by 12%, on the average.

CoreSymphony

This paper describes CoreSymphony , a cooperative and reconfigurable superscalar processor architecture that improves single-thread performance in chip multiprocessor. CoreSymphony enables some narrow-issue cores to be fused into a single wide-issue core. In this paper, we describe the problems associated with achieving the cooperative superscalar processor. We then describe techniques by which to overcome these problems. The evaluation results obtained using SPEC2006 benchmarks indicate that four-core fusion achieves 88% higher IPC than an individual core.

A Unitable Computing Architecture for Chip Multiprocessors

This paper proposes a unitable multi-core architecture, called hyperscalar, that can dynamically unite many scalar cores as a larger superscalar processor to accelerate a thread. To accomplish this, this paper proposes the virtual shared register files (VSRFs) that help the instructions of a thread in different cores can logically face a uniform set of register files. We also propose an instruction analyzer that can detect and tag the dependence information to the newly fetched instructions. With the tags, instructions in the united cores can issue requests to obtain their remote operands via the VSRF. Thus, the dependences arising among instructions in different cores can be resolved. Moreover, some extended instructions are defined for programmers to grow or shrink the number of united cores to match the available instruction level parallelism for different applications. The reconfigurable feature of hyperscalar covers a spectrum of workloads well, providing high single-thread performance when thread level parallelism (TLP) is low and high throughput when TLP is high. Simulation results show that the eight-core hyperscalar chip multiprocessor's two-, four-and eight-core-united configurations archive 93, 80 and 76% of the performance of the monolithic two-, four-and eight-issue out-of-order superscalar processors with lower area costs and better support for software diversity.

Researches on Design and Implementations of Two 2-Bit Predictors

High-performance superscalar processors, which can be used in building base-stations of cell phones and cloud-based web servers, rely heavily on aggressive branch predictors to avoid stalls no matter the branch is taken or not. Dynamic branch predictor which is based on the historical records of the previously executed branches always outputs good performance. Two-bit predictor discussed in this paper is one of the most popular ones that always practically employed. This paper demonstrates two implementations of two-bit predictors. Using BTB to process solo conditional branches is the first one, which is normally used. Other categories of branch instructions may results in several bubbles as the penalty. While the other implementation employs BTB dealing with all kinds of branch instructions, generating target addresses without any delay cycle. Simulation results show that the second implementation has much better performance than the former one. It decreases the mis-prediction rate from 12.26% to 11.48%, and also has much higher prediction accuracy on indirect jumps. With these results, we have our predictor re-designed accordingly and implemented successfully in superscalar processors.

Novel Low Overhead Post-Silicon Self-Correction Technique for Parallel Prefix Adders Using Selective Redundancy and Adaptive Clocking

In this paper, we present a post-silicon self-correction technique to leverage the redundancy present in parallel prefix adders (PPA). Our technique is based on the fact that a set of carries in PPAs can be made mutually exclusive. Therefore, defects in a set of bits can only corrupt the corresponding set of Sum outputs whereas the remaining Sums are computed correctly. To efficiently utilize the above property of PPAs in presence of defects, we perform addition in multiple clock cycles. In cycle-1, one of the correct set of bits are computed and stored at the output registers. In the subsequent cycles, the operands are shifted by one bit at a time and the remaining sets of bits are recovered. This allows us to compute the correct output at the cost of throughput degradation and minor area and delay overhead while maintaining high frequency and yield. Finally, the proposed technique is used in a superscalar processor, whereby the self-correcting adder is assigned lower priority than fault-free adders to reduce the overall throughput degradation.

Parallelization and performance comparison of the conjugate gradient equation solver on multicore Cell and Xeon computers

Multicore accelerators are used today to supplement traditional superscalar processors in massively parallel computer nodes with extra floating-point computation power. This paper presents our parallelization and performance enhancement and evaluation of the conjugate gradient (CG) linear equation solver with enhanced matrix multiplication on the Cell Broadband Engine accelerator. The paper also compares the CG performance results on the Cell and two CG implementations on a computer with two quadcore Xeon processors, one with OpenMP and the other with OpenMPI. We also report the enhancements made on the CG code and performance analysis of CG on single and dual Cell Broadband Engine packages with 8 and 16 synergistic processing elements and on Xeon for heptadiagonal matrices, in particular to matrix multiplication and synchronization. We also report the communication and computation time breakdowns and the floating point operations per second ratio. Our parallel CG solver is shown to scale well with data size, grid dimensionality, and number of cores. Copyright © 2011 John Wiley & Sons, Ltd.

FabScalar

A growing body of work has compiled a strong case for the single-ISA heterogeneous multi-core paradigm. A single-ISA heterogeneous multi-core provides multiple, differently-designed superscalar core types that can streamline the execution of diverse programs and program phases. No prior research has addressed the 'Achilles' heel of this paradigm: design and verification effort is multiplied by the number of different core types. This work frames superscalar processors in a canonical form, so that it becomes feasible to quickly design many cores that differ in the three major superscalar dimensions: superscalar width, pipeline depth, and sizes of structures for extracting instruction-level parallelism (ILP). From this idea, we develop a toolset, called FabScalar, for automatically composing the synthesizable register-transfer-level (RTL) designs of arbitrary cores within a canonical superscalar template. The template defines canonical pipeline stages and interfaces among them. A Canonical Pipeline Stage Library (CPSL) provides many implementations of each canonical pipeline stage, that differ in their superscalar width and depth of sub-pipelining. An RTL generation tool uses the template and CPSL to automatically generate an overall core of desired configuration. Validation experiments are performed along three fronts to evaluate the quality of RTL designs generated by FabScalar: functional and performance (instructions-per-cycle (IPC)) validation, timing validation (cycle time), and confirmation of suitability for standard ASIC flows. With FabScalar, a chip with many different superscalar core types is conceivable.

Architecture of SIMD Type Vector Processor

Throughput and performance are the major constraints in designing system level models. As vector processor used deeply pipelined functional unit, the operation on elements of vector was performed concurrently. It means the elements were processed one by one. Improvement can be made in vector processing by incorporating parallelism in execution of these concurrent operations so that these operations can be performed simultaneously. This paper presents a design and implementation of SIMD-Vector processor that implements this parallelism on short vectors having 4 words. The operation on these words is performed simultaneously i.e. the operation on these words is performed in one cycle. This reduces the clock cycles per instruction (CPI). To implement parallelism in vector processing requires parallel issue and execution of vector instructions. Vector processor operates on a vector and superscalar processor issues multiple instructions at a time. This means parallel pipelines are implemented and then made these to support vector data. SIMD-Vector processor will operate on short vector say 4 words vector in a superscalar fashion i.e. 4 words will be fetched at a time and then executed in parallel. This requires redundant functional units e.g. if addition is to be performed on two vectors multiple adders are needed. We have designed the architecture of SIMD type Vector processor. All the designing parameters are explained.

Exploiting dynamic micro-architecture usage in gate sizing

Modern high performance microprocessors incorporate an abundance of replicated structural components. Many of these components often experience substantially lower utilization while executing a diverse pool of applications. To recover energy efficiency from the lower utilization, system architects resort to dynamic voltage frequency scaling (DVFS). In this paper, we demonstrate that dynamic adaptations using DVFS are markedly energy inefficient than techniques that design circuits ground up for lower performance. We propose a novel microarchitecture aware gate sizing and threshold voltage assignment algorithm to mitigate this current limitation. Our technique is the first of its kind that exploits architectural slack in gate sizing, and leverages on-chip redundancy and slack. We evaluate this circuit-architectural co-optimization framework in a superscalar processor by combining standard cell based gate sizing flows with state-of-the-art architectural simulation. Our results show 17–46% improvement in the datapath energy efficiency over traditional circuit designs incorporating DVFS schemes.

A New Compiler for Space-Time Scheduling of ILP Processors

the exploitation of potential performance of superscalar processors has shown that processor is fed with sufficient instruction bandwidth. The fetcher and the Instruction Stream Buffer (ISB) are the key elements to achieve this target. Beyond the basic blocks, the instruction stream is not supported by current ISBs. The split line instruction problem depreciates this situation for x86 processors. With the implementation of Line Weighted Branch Target Buffer (LWBTB), the advance branch information and reassembling of cache lines can be predicted by the ISB. The code generation for parallel register share architecture involves some issues that are not present in sequential code compilation and is inherently complex. To resolve such issues, a consistency contract between the code and the machine can be defined and a compiler is required to preserve the contract during the transformation of code. We want to achieve high level parallelism at faster clock speed it require distribution of processor resource and avoiding primitive that require single cycle global communication. Distribution of its resources, including instruction stream, register files, memory port and ALUs, over a pipelined two dimensional mesh interconnect are done by raw microprocessor (4). In this paper, we propose a compiler RPCC for general purpose sequential programs on the raw machine.

Performance-optimum superscalar architecture for embedded applications

Embedded applications are widely used in portable devices such as wireless phones, personal digital assistants, laptops, etc. High throughput and real time requirements are especially important in such data-intensive tasks. Therefore, architectures that provide the required performance are the most desirable. On the other hand, processor performance is severely related to the average memory access delay, number of processor registers and also size of the instruction window and superscalar parameters. Therefore, cache, register file and superscalar parameters are the major architectural concerns in designing a superscalar architecture for embedded processors. Although increasing cache and register file size leads to performance improvements in high performance embedded processors, the increased area, power consumption and memory delay are the overheads of these techniques. This paper explores the effect of cache, register file and superscalar parameters on the processor performance to specify the optimum size of these parameters for embedded applications. Experimental results show that although having bigger size of these parameters is one of the performance improvement approaches in embedded processors, however, by increasing the size of some parameters over a threshold value, performance improvement is saturated and especially in cache size, increments over this threshold value decrease the performance.