Energy efficiency of electronic digital processors is primarily limited by the energy consumption of electronic communication and interconnects. The industry is almost unanimously pushing towards replacing both long-h...
详细信息
Energy efficiency of electronic digital processors is primarily limited by the energy consumption of electronic communication and interconnects. The industry is almost unanimously pushing towards replacing both long-haul, as well as local chip interconnects, using optics to drastically increase efficiency. In this paper, we explore what comes after the successful migration to optical interconnects, as with this inefficiency solved, the main source of energy consumption will be electronic digital computing, memory and electro-optical conversion. Our approach attempts to address all these issues by introducing efficient all-optical digital computing and memory, which in turn eliminates the need for electro-optical conversions. Here, we demonstrate for the first time a scheme to enable general-purpose digital data processing in an integrated form and present our photonic integrated circuit (PIC) implementation. For this demonstration we implemented a URISC architecture capable of running any classical piece of software all-optically and present a comprehensive architectural framework for all-optical computing to go beyond.
Implementation of a computer architecture simulator prototype written in Turbo Prolog, which is suitable for educational purposes, is described. Besides adapting to teaching needs for conventional use, the simulator d...
详细信息
Implementation of a computer architecture simulator prototype written in Turbo Prolog, which is suitable for educational purposes, is described. Besides adapting to teaching needs for conventional use, the simulator design incorporates features for an easy modifiability of computer architectural structures appropriate to the trainee's learning level. Thus, this simulator is not limited to a particular architecture but permits the student to vary dynamically the architectural characteristics of the computer under study in order to verify the consequences of his decisions and to experiment with families of computers and alternatives of design.
This paper describes the Connection Machine, a programmable computer based on cellular automata. The essential idea behind the connection machine is that a regular locally-connected cellular array can be made to behav...
详细信息
This paper describes the Connection Machine, a programmable computer based on cellular automata. The essential idea behind the connection machine is that a regular locally-connected cellular array can be made to behave as if the processing cells are connected into any desired topology. When the topology of the machine is chosen to match the topology of the application program, the result is a fast, powerful computing engine. The Connection Machine was originally designed to implement knowledge retrieval operations in artificial intelligence programs, but the hardware and the programming techniques are apparently applicable to a much larger class of problems. A machine with 100,000 processing cells is currently being constructed.
Energy efficient architecture is essential to improve both the performance and power consumption of a computer system. However, modern computers suffer from the severe "memory wall" problem due to the signif...
详细信息
Energy efficient architecture is essential to improve both the performance and power consumption of a computer system. However, modern computers suffer from the severe "memory wall" problem due to the significant performance gap between the processor technology and the memory technology. Thus, the computer architecture community is evolving from compute-centric to memory-centric designs to reduce the data movement overhead. This paper presents a comprehensive survey of the main challenges and recent advances in memory-centric energy efficient computer architecture. We summarize two research directions: improving the memory technology and processing closer to memory. The former focuses on optimizing the conventional memory technology and exploiting emerging non-volatile memory (NVM) technology. The latter talks about currently popular processing in memory (PIM) technology, including near-memory processing (NMP) and in-memory processing (IMP). Moreover, some other topics like hardware for machine learning (ML), ML for hardware, security, privacy, and reliability are gaining increasing attention and should be considered seriously in the design phase of a computer system. The community is facing various challenges and opportunities simultaneously, requiring researchers to have a more comprehensive understanding of this field which is also the goal of this paper.
In this invited work we consider Petri Nets and some interpretations to describe systems in the computer architecture domain. Petri Nets allow to model complex relations of synchronization and concurrency in a simple ...
详细信息
In this invited work we consider Petri Nets and some interpretations to describe systems in the computer architecture domain. Petri Nets allow to model complex relations of synchronization and concurrency in a simple way. A Petri net model can be used to understand a given design by simulation as well as to analyze functional and performance properties without changing the representation of the modelled system. A very simple example is used to illustrate the approach together with brief references to a few works from the literature where other architectural problems have been considered from a Petri Net perspective.
To bridge the gap between learning and doing in traditional computer architecture courses, students require a different educational approach, such as the integrated environment the authors describe.
To bridge the gap between learning and doing in traditional computer architecture courses, students require a different educational approach, such as the integrated environment the authors describe.
Reasoning about today's tremendously complex computer systems is difficult and developing them is expensive. Detailed software simulations are thus essential for evaluating computer architecture ideas. Industry us...
详细信息
Reasoning about today's tremendously complex computer systems is difficult and developing them is expensive. Detailed software simulations are thus essential for evaluating computer architecture ideas. Industry uses simulation extensively during processor and system design as the easiest and least expensive way to explore design options. Unfortunately, constructing accurate models of modern computer systems is becoming harder and more time-consuming, while the effort required to develop high-fidelity simulation tools typically yields few academic rewards. Without funding and promising prospects for academic recognition, research and development in these areas will likely languish.
MinneSPEC proposes reduced input sets that microprocessor designers can use to model representative short-running workloads. A four-step methodology verifies the program behavior similarity of these input sets to refe...
详细信息
MinneSPEC proposes reduced input sets that microprocessor designers can use to model representative short-running workloads. A four-step methodology verifies the program behavior similarity of these input sets to reference sets.
The fields of robotics and game consoles offer an interesting and broad range of lab platforms with appropriate characteristics for teaching computer architecture concepts. This work analyzes the impact of one approac...
详细信息
The fields of robotics and game consoles offer an interesting and broad range of lab platforms with appropriate characteristics for teaching computer architecture concepts. This work analyzes the impact of one approach based on game consoles and another one based on robotics from a triple dimension: student motivation, acquired knowledge, and perception of the employed platform. The study has been carried out on a sample of 96 students using the Arduino-based robot and 75 students using the Nintendo-DS console. A mixed methodology is employed encompassing quantitative and qualitative approaches. Five instruments are used to measure the three aforementioned dimensions. Results show that despite both platforms performing similarly in the three considered dimensions (student motivation, acquired knowledge, and perception of the employed platform), the robotics platform does it slightly better than game console, based on the obtained average scores for the considered instruments. Despite this outperforming, motivation and perception decrease for the students using the robotics platform as result of some identified constraint. This suggests that changes are required in the organization of the lab sessions to promote teamwork skills and to overcome the lack of simulators to remove the obstacles hinting motivation and performance. However, a clear correlation between motivation and perception and acquired knowledge has not been identified on computer architecture. Implications of affordances and constraints of both platforms, types of activities, and their impact on results have been discussed.
Understanding the architecture of a processor can be uninteresting and deterring for computer science students, since low-level details of computer architecture are often perceived to lack real-world impact. These cou...
详细信息
Understanding the architecture of a processor can be uninteresting and deterring for computer science students, since low-level details of computer architecture are often perceived to lack real-world impact. These courses typically have a strong practical component where students learn the fundamentals of the computer architecture and the handling of input/output operations through the development of simple programs in a low-level assembly programming language. Since these practical sessions require a strong involvement, student attendance and withdrawal rates are poor, lowering academic results and introducing a negative feedback loop that preconditions students to dislike them. This article introduces a new methodology for the practical sessions of computer Organization and Design courses. This methodology disavows the use of simulators and focuses on actual hardware to promote a feeling of proximity to the execution and outcome of the programs. The proposed setup uses Raspberry Pi devices to encourage students to work autonomously, due to their low cost, capability of running an OS, and rich ecosystem of simple hardware devices. The setup is completed with RISC OS, which combines a simple window-based graphical interface with a low-level management of the hardware without requiring software abstraction layers. The article presents the methodology and the UCDebug tool, developed to help students debug their codes in RISC OS. After the introduction of the new setup at the University of Cantabria, academic results and student satisfaction have improved. The setup has also allowed to sustain a similar organization of the courses throughout the COVID-19 pandemic.
暂无评论