Current developments in microprocessor design favor increased core counts over frequency scaling to improve processor performance and energy efficiency. Coupling this architectural trend with a message-passing protocol helps realize a data-center-on-a-die. The prototype chip (Figs. 5.7.1 and 5.7.7) described in this paper integrates 48 Pentium ™ class IA-32 cores [1] on a 6×4 2D-mesh network of tiled core clusters with high-speed I/Os on the periphery. The chip contains 1.3B transistors. Each core has a private 256KB L2 cache (12MB total on-die) and is optimized to support a message-passing-programming model whereby cores communicate through shared memory. A 16KB message-passing buffer (MPB) is present in every tile, giving a total of 384KB on-die shared memory, for increased performance. Power is kept at a minimum by transmitting dynamic, fine-grained voltage-change commands over the network to an on-die voltage-regulator controller (VRC). Further power savings are achieved through active frequency scaling at the tile granularity. Memory accesses are distributed over four on-die DDR3 controllers for an aggregate peak memory bandwidth of 21GB/s at 4× burst. Additionally, an 8-byte bidirectional system interface (SIF) provides 6.4GB/s of I/O bandwidth. The die area is 567mm 2 and is implemented in 45nm high-κ metal-gate CMOS [2].The design is organized in a 6×4 2D-array of tiles [3] to increase scalability. Each tile is a cluster of two enhanced IA-32 cores sharing a router for inter-tile communication. Cores operate in-order and are two-way superscalar. A 256 entry lookup table (LUT) extension of the 64-entry TLB translates 32-bit virtual addresses to 36-bit physical addresses. The separate L1 instruction and data caches are upsized to 16KB and support both write-through and write-back. Each L1 cache is reinforced by a unified 256KB 4-way write-back L2 cache. The L2 uses a 32-byte line size, matching the cache line size internal to the core, and has a 10-cycle hit latency. The L2 also uses in-line double-error-detection and single-error-correction for improved performance and several programmable sleep modes for power reduction. The L2 cache controller features a time-outand-retry mechanism for increased system reliability.Shared memory coherency is maintained through software protocols, such as MPI and OpenMP [4], in an effort to eliminate the communication and hardware overhead required for a memory coherent 2D-mesh. A new message-passing memory type (MPMT) is introduced as an architectural enhancement to optimize data sharing using these software procedures. A single bit in a core's TLB designates MPMT cache lines. The MPMT retains all the performance benefits of a conventional cache line, but distinguishes itself by addressing non-coherent shared memory. All MPMT cache lines are invalidated before reads/writes to the shared memory to prevent a core from working on stale data. A new instruction, MBINV, is added to the core to invalidate all MPMT cache entries in a single cycle. Subsequent reads/writes to inval...
A method based on energy minimization is used to determine the spacing and penetration of a regular array of cracks in a slab that is shrinking due to a changing temperature field. The results show a range of different crack propagation behavior dependent on a single dimensionless parameter, being the ratio of the slab thickness and a characteristic length for the material. At low parameter values the minimum energy state can be achieved by continually adding more cracks until a steady state is achieved. At higher values, a minimum crack spacing is reached at finite time, beyond which the cracks are constrained to propagate with the minimum spacing. In the latter case, the uniform propagation is potentially unstable to a spatial period doubling, leading to increasingly complex crack penetration patterns. The energy minimization combined with the period doubling instability provides a means of determining the minimum energy state of cracks for all time. The problem considered here can be seen as a paradigm for cracking phenomena that occur on a large range of scales, from planetary to microscopic.
No abstract
We consider three-dimensional finite-amplitude thermal convection in a fluid layer with boundaries of finite conductivity. Busse & Riahi (1980) and Proctor (1981) showed that the preferred planform of convection in such a system is a square-cell tesselation provided that the boundaries are much poorer conductors than the fluid, in contrast to the roll solutions which are obtained for perfectly conducting boundaries. We determine here the conductivity of the boundaries at which the preferred planform changes from roll to square-cell type. We show that, for low-Prandtl-number fluids (e.g. mercury), square-cell solutions are realized only when the boundaries are almost insulating; while, for high-Prandtl-number fluids (e.g. silicone oils), square-cell solutions are stable when the boundaries have conductivity comparable to that of the fluid.
We consider finite-amplitude thermal convection, in a horizontal fluid layer. The viscosity of the fluid is dependent upon its temperature. Using a weakly nonlinear expansion procedure, we examine the stability of two-dimensional roll and three-dimensional square planforms, in order to determine which should be preferred in convection experiments. The analysis shows that the roll planform is preferred for low values of the ratio of the viscosities at the top and bottom boundaries, but the square planform is preferred for larger values of the ratio. At still larger values, subcritical convection is predicted. We also include the effects of boundaries having finite thermal conductivity, which enables favourable comparison to be made with experimental studies. A discrepancy between the present work and a previous study of this problem (Busse & Frick 1985) is discussed.
The Durham Adaptive Optics Simulation Platform (DASP) is a Monte-Carlo modelling tool used for the simulation of astronomical and solar adaptive optics systems. In recent years, this tool has been used to predict the expected performance of the forthcoming extremely large telescope adaptive optics systems, and has seen the addition of several modules with new features, including Fresnel optics propagation and extended object wavefront sensing. Here, we provide an overview of the features of DASP and the situations in which it can be used. Additionally, the user tools for configuration and control are described.The Durham adaptive optics (AO) simulation platform (DASP) has been under development since the early 1990s. Its current framework was established in 2006 to meet the challenges of modelling the forthcoming extremely large telescopes, with primary mirror diameters of over 20 m. Since 2006, DASP has been regularly developed to improve computational performance, increase simulation fidelity, and expand the number of features that can be modelled. It uses a modular design, allowing new developments and algorithms to be added whilst maintaining compatibility. DASP is developed primarily in Python and C, and uses pthreads and MPI for parallelization enabling modelling of the largest proposed telescopes on reasonable timescales.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.