Search

Computer architects extensively use simulation to steer future processor research and development. Simulating large-scale multicore processors is extremely time-consuming and is sometimes impossible because of simulation infrastructure constraints and/or simulation host compute and memory limitations. This paper proposes scale-model simulation, a novel methodology to predict large-scale multicore system performance. Scale-model simulation first constructs and simulates a scale model of the target system with reduced core count and shared resources. Target system performance is then predicted through machine-learning (ML) based extrapolation. Scale-model simulation predicts 32-core target system performance based on a single-core scale model with an average error of 8.0% and 15.8% for homogeneous and heterogeneous multiprogram workloads, respectively, while yielding a 28×

Generic multi-threaded sampled simulation has been a long-standing, challenging problem with the potential to help change how researchers study modern, complex computing systems. Yet, a practical solution for reducing complex multi-threaded applications into a representative sample has been elusive. Existing techniques either do not provide significant speedups to be useful (Time-Based Sampling techniques can show less than a 10× speedup compared to a fully-detailed simulation) or apply only to particular synchronization types (BarrierPoint for barrier-based workloads or SMARTS for server workloads). In addition, workload-specific solutions, like BarrierPoint, are rigid with respect to region selection, which can limit the overall simulation speedup when intra-barrier regions are large. A solution is needed that both supports generic multi-threaded applications, no matter the synchronization primitives used, as well as allows for ease of deployment and fast evaluation. In this work, we aim to solve these challenges and propose a novel sampling technique for multi-threaded applications, called LoopPoint, that is both agnostic to the type of synchronization primitives used and scales by the similarity exhibited by the application. The proposed methodology combines several vital features, including (1) repeatable, up-front application analysis, (2) a novel clustering approach to take into account run-time parallelism, and (3) the use of loop-based simulation markers to divide the work into measurable chunks, even in the presence of spin-loops. LoopPoint identifies representative simulation regions that can be simulated in parallel to achieve speedups of up to 801x for the train input set of the multi-threaded SPEC CPU2017 benchmarks with an average simulation error of just 2.33%. For the ref inputs of CPU2017, we calculate the speedup with LoopPoint to be 11,587x on average (for parallel simulation), and up to 31,253x, demonstrating how the identification of application regularity and loops can lead to significant simulation improvements compared to state-of-the-art solutions.

We introduce statistical sampling for the record manager/archivist, its various options and how to correctly apply these techniques with modern aids. Using the mathematical concept of confidence intervals, we propose a statistically sound method for determining sample size and quality. Illustrated by means of examples and graphs, this guide aims to show statistical sampling as a valuable option when defining a successful record and archive management strategy.

Single-ISA heterogeneous multi-cores consisting of small (e.g., in-order) and big (e.g., out-of-order) cores dramatically improve energy- and power-efficiency by scheduling workloads on the most appropriate core type. A significant body of recent work has focused on improving system throughput through scheduling. However, none of the prior work has looked into fairness. Yet, guaranteeing that all threads make equal progress on heterogeneous multi-cores is of utmost importance for both multi-threaded and multi-program workloads to improve performance and quality-of-service. Furthermore, modern operating systems affinitize workloads to cores (pinned scheduling) which dramatically affects fairness on heterogeneous multi-cores. In this paper, we propose fairness-aware scheduling for single-ISA heterogeneous multi-cores, and explore two flavors for doing so. Equal-time scheduling runs each thread or workload on each core type for an equal fraction of the time, whereas equal-progress scheduling strives at getting equal amounts of work done on each core type. Our experimental results demonstrate an average 14% (and up to 25%) performance improvement over pinned scheduling through fairness-aware scheduling for homogeneous multi-threaded workloads; equal-progress scheduling improves performance by 32% on average for heterogeneous multi-threaded workloads. Further, we report dramatic improvements in fairness over prior scheduling proposals for multi-program workloads, while achieving system throughput comparable to throughput-optimized scheduling, and an average 21% improvement in throughput over pinned scheduling.

Co-optimizing hardware and software can lead to substantial performance and energy benefits, and is becoming an increasingly important design paradigm. In scientific computing, power constraints increasingly necessitate the return to specialized chips such as Intel’s MIC or IBM’s Blue-Gene architectures. To enable hardware/software co-design in early stages of the design cycle, we propose a simulation infrastructure methodology by combining high-abstraction performance simulation using Sniper with power modeling using McPAT and custom DRAM power models. Sniper/McPAT is fast — simulation speed is around 2 MIPS on an 8-core host machine — because it uses analytical modeling to abstract away core performance during multi-core simulation. We demonstrate Sniper/McPAT’s accuracy through validation against real hardware; we report average performance and power prediction errors of 22.1% and 8.3%, respectively, for a set of SPEComp benchmarks.

Sniper is a next generation parallel, high-speed and accurate x86 simulator. This multi-core simulator is based on the interval core model and the Graphite simulation infrastructure, allowing for fast and accurate simulation and for trading off simulation speed for accuracy to allow a range of flexible simulation options when exploring different homogeneous and heterogeneous multi-core architectures. The Sniper simulator allows one to perform timing simulations for both multi-programmed workloads and multi-threaded, shared-memory applications running on 10s to 100+ cores, at a high speed when compared to existing simulators. The main feature of the simulator is its core model which is based on interval simulation, a fast mechanistic core model. Interval simulation raises the level of abstraction in architectural simulation which allows for faster simulator development and evaluation times; it does so by ’jumping’ between miss events, called intervals. Sniper has been validated against multi-socket Intel Core2 and Nehalem systems and provides average performance prediction errors within 25% at a simulation speed of up to several MIPS. This simulator, and the interval core model, is useful for uncore and system-level studies that require more detail than the typical one-IPC models, but for which cycle-accurate simulators are too slow to allow workloads of meaningful sizes to be simulated. As an added benefit, the interval core model allows the generation of CPI stacks, which show the number of cycles lost due to different characteristics of the system, like the cache hierarchy or branch predictor, and lead to a better understanding of each component’s effect on total system performance. This extends the use for Sniper to application characterization and hardware/software co-design. The Sniper simulator is available for download at http://snipersim.org and can be used freely for academic research.

Writing well-performing parallel programs is challenging in the multi-core processor era. In addition to achieving good per-thread performance, which in itself is a balancing act between instruction-level parallelism, pipeline effects and good memory performance, multi-threaded programs complicate matters even further. These programs require synchronization, and are affected by the interactions between threads through sharing of both processor resources and the cache hierarchy. At the Intel Exascience Lab, we are developing an architectural simulator called Sniper for simulating future exascale-era multi-core processors. Its goal is twofold: Sniper should assist hardware designers to make design decisions, while simultaneously providing software designers with a tool to gain insight into the behavior of their algorithms and allow for optimization. By taking architectural features into account, our simulator can provide more insight into parallel programs than what can be obtained from existing performance analysis tools. This unique combination of hardware simulator and software performance analysis tool makes Sniper a useful tool for a simultaneous exploration of the hardware and software design space for future high-performance multi-core systems.

While the number of cores in both general purpose chip-multiprocessors (CMPs) and embedded Multi-Processor Systems-on-Chip (MPSoCs) keeps rising, on-chip communication becomes more and more important. In order to write efficient programs for these architectures, it is therefore necessary to have a good idea of the communication behavior of an application. We present a communication profiler that extracts this behavior from compiled, sequential C/C++ programs, and constructs a dynamic data-flow graph at the level of major functional blocks. It can also be used to view differences between program phases, such as different video frames, which allows both input- and phase-specific optimizations to be made. Finally, PinComm can visualize inter-thread communication in parallel programs, which can help in optimizing communication behavior and spotting communication-related performance bottlenecks.

While the number of cores in both embedded MultiProcessor Systems-on-Chip and general purpose processors keeps rising, on-chip communication becomes more and more important. In order to write efficient programs for these architectures, it is therefore necessary to have a good idea of the communication behavior of an application. We present a communication profiler that extracts this behavior from compiled sequential C/C++ programs, and constructs a dynamic dataflow graph at the level of major functional blocks. In contrast to existing methods of measuring inter-program communication, our tool automatically generates the program's dataflow graph and is less demanding for the developer. It can also be used to view differences between program phases (such as different video frames), which allows both input- and phase-specific optimizations to be made. We also look at how this information can subsequently be used to guide the effort of parallelizing the application, to co-design the software, memory hierarchy and communication hardware, and to provide new sources of communication-related runtime optimizations.