An official website of the United States government
Recent manycore processors are kept coherent using scalable distributed directories. A paramount example is the Xeon Phi Knights Landing. It features 38 tiles packed in a single die, organized into a 2D mesh. Before accessing remote data, tiles need to query the distributed directory. The effect of this coherence traffic is poorly understood. We show that the apparent UMA behavior results from the degradation of the peak performance. We develop ways to optimize the coherence traffic, the core-to-core-affinity, and the scheduling of a set of tasks on the mesh, leveraging the unique characteristics of processor units stemming from process variations.
more »
« less
Study of Damage Development in Under-Bump Interconnects by Thermo-Mechanical Stress in Package Interconnects
- Award ID(s):
- 2122128
- PAR ID:
- 10653023
- Publisher / Repository:
- IEEE
- Date Published:
- Page Range / eLocation ID:
- 1644 to 1648
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
The development of magnetic logic devices dictates a need for a novel type of interconnect for magnetic signal transmission. Fast signal damping is one of the problems which drastically differs from conventional electric technology. Here, we describe a magnetic interconnect based on a composite multiferroic comprising piezoelectric and magnetostrictive materials. Internal signal amplification is the main reason for using multiferroic material, where a portion of energy can be transferred from electric to magnetic domains via stress-mediated coupling. The utilization of composite multiferroics consisting of piezoelectric and magnetostrictive materials offers flexibility for the separate adjustment of electric and magnetic characteristics. The structure of the proposed interconnect resembles a parallel plate capacitor filled with a piezoelectric, where one of the plates comprises a magnetoelastic material. An electric field applied across the plates of the capacitor produces stress, which, in turn, affects the magnetic properties of the magnetostrictive material. The charging of the capacitor from one edge results in the charge diffusion accompanied by the magnetization change in the magnetostrictive layer. This enables the amplitude of the magnetic signal to remain constant during the propagation. The operation of the proposed interconnects is illustrated by numerical modeling. The model is based on the Landau–Lifshitz–Gilbert equation with the electric field-dependent anisotropy term included. A variety of magnetic logic devices and architectures can benefit from the proposed interconnects, as they provide reliable and low-energy-consuming data transmission. According to the estimates, the group velocity of magnetic signals may be up to 105 m/s with energy dissipation less than 10−18 J per bit per 100 nm. The physical limits and practical challenges of the proposed approach are also discussed.more » « less
-
The size of transistors has drastically reduced over the years. Interconnects have likewise also been scaled down. Today, conventional copper (Cu)-based interconnects face a significant impediment to further scaling since their electrical conductivity decreases at smaller dimensions, which also worsens the signal delay and energy consumption. As a result, alternative scalable materials such as semi-metals and 2D materials were being investigated as potential Cu replacements. In this paper, we experimentally showed that CoPt can provide better resistivity than Cu at thin dimensions and proposed hybrid poly-Si with a CoPt coating for local routing in standard cells for compactness. We evaluated the performance gain for DRAM/eDRAM, and area vs. performance trade-off for D-Flip-Flop (DFF) using hybrid poly-Si with a thin film of CoPt. We gained up to a 3-fold reduction in delay and a 15.6% reduction in cell area with the proposed hybrid interconnect. We also studied the system-level interconnect design using NbAs, a topological semi-metal with high electron mobility at the nanoscale, and demonstrated its advantages over Cu in terms of resistivity, propagation delay, and slew rate. Our simulations revealed that NbAs could reduce the propagation delay by up to 35.88%. We further evaluated the potential system-level performance gain for NbAs-based interconnects in cache memories and observed an instructions per cycle (IPC) improvement of up to 23.8%.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1109/ECTC51529.2024.00272
