Presentation is loading. Please wait.

Presentation is loading. Please wait.

20031 Janusz Starzyk and Yongtao Guo School of Electrical Engineering and Computer Science Ohio University, Athens, OH 45701, U.S.A. September,

Similar presentations


Presentation on theme: "20031 Janusz Starzyk and Yongtao Guo School of Electrical Engineering and Computer Science Ohio University, Athens, OH 45701, U.S.A. September,"— Presentation transcript:

1 SOLAR @ 20031 Janusz Starzyk and Yongtao Guo School of Electrical Engineering and Computer Science Ohio University, Athens, OH 45701, U.S.A. September, 2003

2 SOLAR @ 20032 ONTLINE 1.Introduction 2.SOLAR Principle 3.Simulation Results 4.HW/SW Co-Simulation 5.Hardware Organization 6.Conclusion

3 SOLAR @ 20033 Self Organizing Learning Array SOLAR New learning algorithm –Multi layer structure and on-line learning; –local and sparse interconnections; –entropy based self-organized learning Superior performance –Parallel computing organization; –Low power dissipation; –Efficient communication; –High chip utilization rate; Potential to be a leading technology in machine learning –pave the way to machine intelligence application areas including pattern recognition, intelligent control, signal processing, robotics and biological research.

4 SOLAR @ 20034 DARPA: Cognitive Information Processing Technology  Wanted: machine that can reason, using substantial amounts of knowledge  Can learn from its experiences so that its performance improves with knowledge and experience  Can explain itself and can accept direction  Is aware of its own behavior and reflects on its own capabilities  Responds in a robust manner to a surprise

5 SOLAR @ 20035 Self-Organizing Learning ARray (SOLAR ) Dowling, 1998, p. 17

6 SOLAR @ 20036 Here,,, represent the probabilities of each class, attribute probability and joint probability respectively. Self-organizing Principle Neuron self-organization includes:  Selection of inputs  Choosing transformation function  Setting threshold  Providing output probabilities  Setting output control

7 SOLAR @ 20037 Self-Organizing Process and Neuron Structure

8 Self-organizing Process Matlab Simulation Initial interconnection Learning process

9 SOLAR @ 20039 Synthetic Data Classification

10 SOLAR @ 200310 Credit Card Data Set Method Error Rate Cal5 0.131 SOLAR 0.133 Itrule 0.137 Discrim 0.141 Logdisc 0.141 DIPOL92 0.141 CART 0.145 RBF 0.145 CASTLE 0.148 NaiveBay 0.151 Backprop 0.154 C4.5 0.155 SMART 0.158 Baytree 0.171 k-NN 0.181 NewID 0.181 LVQ 0.197 ALLOC80 0.201 Quadisc 0.207 Default 0.440 Kohonen Failed SOLAR self organizing structure

11 SOLAR @ 200311 SW/HW codesign of SOLAR JTAG Programming Software run in PC PCI Bus Hardware Board Virtex XCV800FPGA dynamic configuration

12 SOLAR @ 200312 Cosimulation - What and Why? Cosimulation –Simulation of heterogeneous systems whose hardware and software components are interacting Benefits of cosimulation –Verifying correct functionality of the target even before hardware is built –Profiling the dynamic behavior –Identifying the performance bottleneck –Preventing problems such as over-design or under- design related to system integration –Saving the system development cost and cycle

13 SOLAR @ 200313 Traditional Cosimulation Environment –A software process Written in high-level language, such as C/C++ –A simulation process of hardware model Hardware description language, such as VHDL –Inter-process communication (IPC) routine Connect the hardware process and software process Software Model (C-program) Hardware Model (VHDL) IPC routines Foreign IPC procedures IPC Two simulators

14 SOLAR @ 200314 Traditional Cosimulation  To perform cosimulation, two simulators should be combined and complex IPC should be developed. These IPCs are error-prone routines requiring to handle various formats of data and processed signals  Especially, when focusing on hardware part, we hope that the software part is minimized and the HW/SW communication is simple and reliable

15 SOLAR @ 200315 SOLAR Cosimulation –A software process Written in behavioral VHDL which is not synthesizable –A hardware process Written in RTL VHDL which is synthesizable –HW/SW communication FSM and FIFOs Software Model (Behavioral VHDL) Hardware Model (RTL VHDL) One simulator FSM and FIFOs

16 SOLAR @ 200316 SOLAR Cosimulation  To perform SOLAR cosimulation, one single VHDL simulator is applied. So complex error- prone IPC is avoided. Data formats and other problems can be easily handled.  The interface between HW/SW is implemented by several FIFOs controlled by a FSM, which is simple, reliable and easily modified.  File I/O functions are used to simplify software part design when focusing on hardware part implementation.

17 SOLAR @ 200317 Co-simulation System Decomposition Interface modeling (RTL VHDL Main Initialization File I/O SOLAR Training Over No Yes System architecture modelling (Behavioral VHDL) Input FIFO Output FIFO FSMFSM Interface Control OP EBE REG FIFO MEM Self-organizing learning architecture (Structural VHDL)

18 SOLAR @ 200318 SW Organization VHDL Model All functions and signal variables in the packages are shared, and program execution is functionally interleaved. lower level package is the description for system input and output, initialization and update of the memory element in the network. The higher level packages encapsulate new system functions based on the functions described by the lower level packages. The highest design level function representing the software part in the overall system implements the system organization and management.

19 SOLAR @ 200319 Single Neuron’s Hardware Architecture Figure 4: Single neuron’s learning architecture D REG CTRL R R R R FIFO/DMA CTRL MAIN CONTROLLER OP 1024X32 FIFO INTERFACEINTERFACE INTERFACEINTERFACE M ALU M

20 SOLAR @ 200320 Interface Process SW HW time configuration send data Receive data conf done start wait command send command over read registers dma request … … time

21 SOLAR @ 200321 Interface Modeling class other 1 2 3 4 5 6 Software (behavioral VHDL) Interface FIFOs memory module Ctrl Others Figure 5: Interface modeling using FSM&FIFO Hardware (structural VHDL) training

22 SOLAR @ 200322 Interface Simulation Small Training Data Set

23 SOLAR @ 200323 System Synchronized Work Software Work Hardware Work Interface Work Time

24 SOLAR @ 200324 Incremental Prototyping Overall system design can be accelerated by replacing HW subcomponent with real hardware once successfully simulated. HW function is completely defined and prototyped t HW function VHDL- simulated (incremental part)

25 SOLAR @ 200325 EBE Simulation Main Procedures contain:  Sending data from software to Chip Memory  Trigger start signal  ALU calculation for all data  Moving calculated results to intermediate memory  Threshold scanning & ID calculation  Updating the intermediate values  Data Movement if the current ID is optimal  Repeating from 3 to 6 untill all functions are scanned  Sending data from Chip to software In this simulation waveform, the signal “Opt_Threshold” and “ID” represent the optimal threshold and the corresponding information index deficiency for this particular training neuron in its learning subspace.

26 SOLAR @ 200326 EBE Prototyping SOLAR Training SOLAR Training Map onto Virtex (57.8% logic, 60.3% route) Minimum period: 23.140ns (Maximum Frequency: 43.215MHz) Minimum input arrival time before clock: 11.036ns Maximum output required time after clock: 13.758ns

27 SOLAR @ 200327 For instance, a particular neuron has 1024 subspace data. PC to Chip: 38x1024 = 38912 CLKs ALU calculation: 16x1024=16384 CLKs Threshold scan & ID calculation (maximum): (4x1024+7) x1024=4201472 CLKs Data Movement (Maximum) 1x1024=1024 CLKs Chip to PC: 1x1024=1024 CLKs Other: (starting sequence, wait, handshaking, etc.) 20x1024 =20480 CLKs Total: 38912+(16384+4201472+1024)x7+1024+20480= 29592576 CLKs Run Time Main Operations CLK Number per DATA PC data to in- chip memory 38 ALU Calculation 16 Threshold Scanning 4 ID calculation7 Memory data Movement 1 In-chip FIFO to PC 1 x7 functions

28 SOLAR @ 200328 Prototyping Board

29 SOLAR @ 200329 Future Work - System SOLAR

30 SOLAR @ 200330 SOLAR will grow Rack (4 boards,1x4) 1 Million gates 6 Million gates 24 Million gates Half of a billion gates Board (6 chips,2x3)System (16 cabinets, 4X4)Chip VIRTEXCV1000

31 SOLAR @ 200331 Questions


Download ppt "20031 Janusz Starzyk and Yongtao Guo School of Electrical Engineering and Computer Science Ohio University, Athens, OH 45701, U.S.A. September,"

Similar presentations


Ads by Google