skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.


Search for: All records

Creators/Authors contains: "Richardson, Stephen"

Note: When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external site maintained by the publisher. Some full text articles may not yet be available without a charge during the embargo (administrative interval).
What is a DOI Number?

Some links on this page may take you to non-federal websites. Their policies may differ from this site.

  1. Amber is a system-on-chip (SoC) with a coarse-grained reconfigurable array (CGRA) for acceleration of dense linear algebra applications, such as machine learning (ML), image processing, and computer vision. It is designed using an agile accelerator-compiler co-design flow; the compiler updates automatically with hardware changes, enabling continuous application-level evaluation of the hardware-software system. To increase hardware utilization and minimize reconfigurability overhead, Amber features the following: 1) dynamic partial reconfiguration (DPR) of the CGRA for higher resource utilization by allowing fast switching between applications and partitioning resources between simultaneous applications; 2) streaming memory controllers supporting affine access patterns for efficient mapping of dense linear algebra; and 3) low-overhead transcendental and complex arithmetic operations. The physical design of Amber features a unique clock distribution method and timing methodology to efficiently layout its hierarchical and tile-based design. Amber achieves a peak energy efficiency of 538 INT16 GOPS/W and 483 BFloat16 GFLOPS/W. Compared with a CPU, a GPU, and a field-programmable gate array (FPGA), Amber has up to 3902x, 152x, and 107x better energy-delay product (EDP), respectively. 
    more » « less
  2. null (Ed.)
    Subducting tectonic plates carry water and other surficial components into Earth’s interior. Previous studies suggest that serpentinized peridotite is a key part of deep recycling, but this geochemical pathway has not been directly traced. Here, we report Fe-Ni–rich metallic inclusions in sublithospheric diamonds from a depth of 360 to 750 km with isotopically heavy iron (δ 56 Fe = 0.79 to 0.90‰) and unradiogenic osmium ( 187 Os/ 188 Os = 0.111). These iron values lie outside the range of known mantle compositions or expected reaction products at depth. This signature represents subducted iron from magnetite and/or Fe-Ni alloys precipitated during serpentinization of oceanic peridotite, a lithology known to carry unradiogenic osmium inherited from prior convection and melt depletion. These diamond-hosted inclusions trace serpentinite subduction into the mantle transition zone. We propose that iron-rich phases from serpentinite contribute a labile heavy iron component to the heterogeneous convecting mantle eventually sampled by oceanic basalts. 
    more » « less