Scalable Computing Systems Laboratory - Useful Tools
Downloads Available from SCALE

ABACUS AND DRUM released on Github


1. Thermal Characterization of Real Processors


Hot spots are a major concern in high-end processors as they constrain performance and limit the lifetime of semiconductor chips. To illustrate the magnitude of thermal variations, we use our infrared imaging setup to capture thermal traces of a dual-core AMD Athlon II 240 and a dual-core Intel Core 2 Duo processor while running various CPU SPEC2006. The traces demonstrate that within die thermal gradients can reach up to 16 C, and that differences in workloads can lead to strong variations in hotspot locations.We also use these data to find different sensor allocation techniques and various full thermal map characterization methods. Detailed methods can be found in following publications.




Chip configuration

Test Chip AMD Athlon II x2 240 – Dual Core
Technology 45 nm
Die Dimension 14 mm x 8.5 mm
Frequency 2.1 GHz
Benchmark SPEC CPU 2006



Infrared Imaging Thermal Data
Related publications
  • S. Reda, R. Cochran, and A. N. Nowroz, "Improved Thermal Tracking for Processors Using Hard and Soft Sensor Allocation Techniques," IEEE Transactions on Computers, vol. 60(6), pp. 841 – 861, 2011.
  • A. N. Nowroz, R. Cochran and S. Reda, “Thermal Monitoring of Real Processors: Techniques for Sensor Allocation and Full Characterization,” Design Automation Conference , pp. 56 – 61, 2010.
  • S. Reda, “Thermal and Power Characterization of Real Computing Devices,” IEEE Journal on Emerging Topics in Circuits and Systems, vol. 1(2), pp. 76 – 87, 2011.


2. Power Characterization of FPGA and Multi-core Processor


Design-time power analysis is one of the most critical tasks conducted by chip architects and circuit designers. While computer aided power analysis tools can provide power consumption estimates for various circuit blocks, these estimates can substantially deviate from the actual power consumption of working silicon chips. We propose a novel methodology that provides accurate, detailed post-silicon spatial power estimates using the thermal infrared emissions from the backside of silicon die. We have performed post-silicon spatial power characterization on two different chips, Stratix II FPGA and AMD Athlon II x4 quad-core chip.

2.1. Stratix II FPGA



Chip configuration

Test Chip Stratix II FPGA
Technology 90 nm
Die Dimension 23 mm x 24 mm
Test Area Dimension 6.4 mm x 7 mm
Frequency 50 MHz clock
Benchmark Micro heater block based


The post-silicon spatial power estimates for Stratix II FPGA is divided into two different sections, one DC analysis when the blocks are excited with DC source and AC analysis when the blocks are excited with AC source. Please refer to following related publications for detailed framework.

Data for downloads - Bi-level

Data for downloads - Multi-level

Chip configuration

Test Chip AMD Athlon II x4 Quad-core
Technology 45 nm
Die Dimension 12 mm x 14 mm
Frequency 1.7GHz - 2.4 GHz
Benchmark SPEC CPU 2006


Data for downloads
  • AMDquadcore DATA – ISLPED2013 [Download here]


  • Related publications - AMD Quad-core
    • • Kapil Dev, Abdullah Nazma Nowroz and Sherief Reda, “Power Mapping and Modeling of Multi-core Processors,” in IEEE ISLPED,pp. 39-44, 2013.


    Non-therma/power related downloads:

    • Jan'08. Silicon measurements of the within-die variations of 10 chips are now available for direct download.
    • My work on APLACE [C18, C20] is now available.
    • Some of my placement research work (J5, C12, C13, C14) is integrated and available in the publicly available placer Capo.
    • The code for the TSP approach Match Twice and Stitch (J2) is available [download]. The input format is the TSPLIB format.
    • A CAD tool for physical design of DNA arrays  (J1, J3, C7, C8, C9, C10) [download]. Besides using simple academic input and output format, the tool also accepts input files from Affymetrix and outputs their layout format.
    • GCHECK: An equivalent checker that uses BDDs and SATs as internal verification engines. This tool has not been maintained since 2001.