An official website of the United States government
Full-system simulation of computer systems is critical for capturing the complex interplay between various hard-ware and software components in future systems. Modeling the network subsystem is indispensable for the fidelity of full-system simulations due to the increasing importance of scale-out systems. Over the last decade, the network software stack has undergone major changes, with userspace networking stacks and data-plane networks rapidly replacing the conventional kernel network stack. Nevertheless, the current state-of-the-art architectural simulator, gem5, still employs kernel networking, which precludes realistic network application scenarios. In this work, we first demonstrate the limitations of gem5's current network stack in achieving high network bandwidth. Then, we enable a userspace networking stack on gem5. We extend gem5's NIC hardware model and device driver to sup-port userspace device drivers running the DPDK framework. Additionally, we implement a network load generator hardware model in gem5 to generate various traffic patterns and per-form per-packet timestamp and latency measurements without introducing packet loss. We develop a suite of six network-intensive benchmarks for stress testing the host network stack. These applications, based on DPDK, can run on both gem5 and real systems. Our experimental results show that enabling userspace networking improves gem5's network bandwidth by 6.3× compared with the current Linux kernel software stack. We characterize the performance of DPDK benchmarks running on both a real system and gem5, and evaluate the sensitivity of the applications to various system and microarchitecture parameters. This work marks the first step in refactoring the networking subsystem in gem5.
more »
« less
MetaSys: A Practical Open-source Metadata Management System to Implement and Evaluate Cross-layer Optimizations
This article introduces the first open-source FPGA-based infrastructure, MetaSys, with a prototype in a RISC-V system, to enable the rapid implementation and evaluation of a wide range of cross-layer techniques in real hardware. Hardware-software cooperative techniques are powerful approaches to improving the performance, quality of service, and security of general-purpose processors. They are, however, typically challenging to rapidly implement and evaluate in real hardware as they require full-stack changes to the hardware, system software, and instruction-set architecture (ISA). MetaSys implements a rich hardware-software interface and lightweight metadata support that can be used as a common basis to rapidly implement and evaluate new cross-layer techniques. We demonstrate MetaSys’s versatility and ease-of-use by implementing and evaluating three cross-layer techniques for: (i) prefetching in graph analytics; (ii) bounds checking in memory unsafe languages, and (iii) return address protection in stack frames; each technique requiring only ~100 lines of Chisel code over MetaSys. Using MetaSys, we perform the first detailed experimental study to quantify the performance overheads of using a single metadata management system to enable multiple cross-layer optimizations in CPUs. We identify the key sources of bottlenecks and system inefficiency of a general metadata management system. We design MetaSys to minimize these inefficiencies and provide increased versatility compared to previously proposed metadata systems. Using three use cases and a detailed characterization, we demonstrate that a common metadata management system can be used to efficiently support diverse cross-layer techniques in CPUs. MetaSys is completely and freely available at https://github.com/CMU-SAFARI/MetaSys .
more »
« less
- Award ID(s):
- 2028949
- PAR ID:
- 10414819
- Date Published:
- Journal Name:
- ACM Transactions on Architecture and Code Optimization
- Volume:
- 19
- Issue:
- 2
- ISSN:
- 1544-3566
- Page Range / eLocation ID:
- 1 to 29
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Over the past decade, machine learning model complexity has grown at an extraordinary rate, as has the scale of the systems training such large models. However there is an alarmingly low hardware utilization (5-20%) in large scale AI systems. The low system utilization is a cumulative effect of minor losses across different layers of the stack, exacerbated by the disconnect between engineers designing different layers spanning across different industries. To address this challenge, in this work we designed a cross-stack performance modelling and design space exploration framework. First, we introduce CrossFlow, a novel framework that enables cross-layer analysis all the way from the technology layer to the algorithmic layer. Next, we introduce DeepFlow (built on top of CrossFlow using machine learning techniques) to automate the design space exploration and co-optimization across different layers of the stack. We have validated CrossFlow’s accuracy with distributed training on real commercial hardware and showcase several DeepFlow case studies demonstrating pitfalls of not optimizing across the technology-hardware-software stack for what is likely, the most important workload driving large development investments in all aspects of computing stack.more » « less
-
Energy-harvesting designs typically include highly entangled app-lication-level and energy-management subsystems that span both hardware and software. This tight integration makes developing sophisticated energy-harvesting systems challenging, as developers have to consider both embedded system development and intermit-tent energy management simultaneously. Even when successful, solutions are often monolithic, produce suboptimal performance, and require substantial effort to translate to a new design. Instead, we propose a new energy-harvesting power management architecture, Altair that offloads all energy-management operations to the power supply itself while making the power supply programmable. Altair introduces an energy supervisor and a standard interface to enable an abstraction layer between the power supply hardware and the running application, making both replaceable and recon-figurable. To ensure minimal resource conflict on the application processor, while running resource-hungry optimization techniques in the supervisor, we implement the Altair design in a lower power microcontroller that runs in parallel with the application. We also develop a programmable power supply module and a software library for seamless application development with Altair. We evaluate the versatility of the proposed architecture across a spectrum of IoT devices and demonstrate the generality of the plat-form. We also design and implement an online energy-management technique using reinforcement learning on top of the platform and compare the performance against fixed duty-cycle baselines. Results indicate that sensors running the online energy-manager perform similar to continuously powered sensors, have a l0x higher event generation rate than the intermittently powered ones, 1.8-7x higher event detection accuracy, experience 50% fewer power failures, and are 44% more available than the sensors that maintain a constant duty-cycle.more » « less
-
In recent years, Quantum Computing (QC) has progressed to the point where small working prototypes are available for use. Termed Noisy Intermediate-Scale Quantum (NISQ) computers, these prototypes are too small for large benchmarks or even for Quantum Error Correction, but they do have sufficient resources to run small benchmarks, particularly if compiled with optimizations to make use of scarce qubits and limited operation counts and coherence times. QC has not yet, however, settled on a particular preferred device implementation technology, and indeed different NISQ prototypes implement qubits with very different physical approaches and therefore widely-varying device and machine characteristics. Our work performs a full-stack, benchmark-driven hardware-software analysis of QC systems. We evaluate QC architectural possibilities, software-visible gates, and software optimizations to tackle fundamental design questions about gate set choices, communication topology, the factors affecting benchmark performance and compiler optimizations. In order to answer key cross-technology and cross-platform design questions, our work has built the first top-to-bottom toolflow to target different qubit device technologies, including superconducting and trapped ion qubits which are the current QC front-runners. We use our toolflow, TriQ, to conduct real-system measurements on 7 running QC prototypes from 3 different groups, IBM, Rigetti, and University of Maryland. From these real-system experiences at QC's hardware-software interface, we make observations about native and software-visible gates for different QC technologies, communication topologies, and the value of noise-aware compilation even on lower-noise platforms. This is the largest cross-platform real-system QC study performed thus far; its results have the potential to inform both QC device and compiler design going forward.more » « less
-
Graphics processing units (GPUs) manufactured by NVIDIA continue to dominate many fields of research, including real-time GPU-management. NVIDIA’s status as a key enabling technology for deep learning and image processing makes this unsurprising, especially when combined with the company’s push into embedded, safety-critical domains like autonomous driving. NVIDIA’s primary competitor, AMD, has received comparatively little attention, due in part to few embedded offerings and a lack of support from popular deep-learning toolkits. Recently, however, AMD’s ROCm (Radeon Open Compute) software platform was made available to address at least the second of these two issues, but is ROCm worth the attention of safety-critical software developers? In order to answer this question, this paper explores the features and pitfalls of AMD GPUs, focusing on contrasting details with NVIDIA’s GPU hardware and software. We argue that an open software stack such as ROCm may be able to provide much-needed flexibility and reproducibility in the context of real-time GPU research, where new algorithmic or analysis techniques should typically remain agnostic to the underlying GPU architecture. In support of this claim, we summarize how closed-source platforms have obstructed prior research using NVIDIA GPUs, and then demonstrate that AMD may be a viable alternative by modifying components of the ROCm software stack to implement spatial partitioning. Finally, we present a case study using the PyTorch deep-learning framework that demonstrates the impact such modifications can have on complex real-world software.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1145/3505250
