Giga-Scale System-On-A-Chip International Center on System-on-a-Chip (ICSOC) Jason Cong University of California, Los Angeles Tel: , (Other participants are listed inside)

Jason Cong2 Project Summary Develop new design methodology to enable efficient giga-scale integration for system-on-a-chip (SOC) designs Project includes three major components –SOC synthesis tools and methodologies –SOC verification, test, and diagnosis –SOC design driver – network processor

Jason Cong3 Research Team by Institutions  US  UCLA: Jason Cong  UC Santa Barbara: Tim Cheng  Taiwan  NTHU: Shi-Yu Huang, Tingting Hwang, J. K. Lee, Youn-Long Lin, C. L. Liu, Cheng-Wen Wu, Allen Wu  NCTU: Jing-Yang Jou  China  Tsinghua Univ.: Jinian Bian, Xianlong Hong, Zeyi Wang, Hongxi Xue  Peking Univ.: Xu Cheng  Zhejiang Univ.: Xiaolang Yan

Jason Cong4 Current Research Team  US  UCLA: Jason Cong  UC Santa Barbara: Tim Cheng  Taiwan  NTHU: Shi-Yu Huang, Tingting Hwang, J. K. Lee, Youn-Long Lin, C. L. Liu, Cheng-Wen Wu, Allen Wu  NCTU: Jing-Yang Jou  China  Tsinghua Univ.: Jinian Bian, Xianlong Hong, Zeyi Wang, Hongxi Xue  Peking Univ.: Xu Cheng  Zhejiang Univ.: Xiaolang Yan  Several new faculty members in the 7 institutions  Guest members from National University of Singapore, Purdue Univ., and UCLA (EE Dept)

Jason Cong5 Thrust 1 -- SOC Synthesis Environment/Methodology (Led by Jason Cong) Code Generation for Retargetable Compiler and Assembler Generator Design Spec VHDL/C Co-Simulation Design Partitioning DSP Synthesis and Optimization FPGA Synthesis and Technology Mapping ASIC Synthesis Interconnect-Driven High-level Synthesis Synthesis for IP Reuse Physical Synthesis for Full-Chip Assembly Embedded Processors DSPs Embedded FPGAs Customized Logic

Jason Cong clock 2 clock 3 clock 4 clock 5 clock n ITRS’ um Tech n 5.63 G Hz across-chip clock n 800 mm 2 (28.3mm x 28.3mm) n IPEM BIWS estimations u Buffer size: 100x u Driver/receiver size: 100x n On semi-global layer (tier 3) : u Can travel up to 11.4 mm in one cycle u Need 5 clock cycles from corner to corner Interconnect Bottleneck in Nanometer Designs u 2nd challenge: Single-cycle full chip synchronization is no longer possible u Not supported by the current CAD toolset u About to happen soon

Jason Cong7 Regular Distributed Register Architecture (2) Global Interconnect … LCC Reg. file … LCC Reg. file … LCC Reg. file … LCC Reg. file … LCC Reg. file … LCC Reg. file FSM Local Computational Cluster (LCC) …. Register File WiWi HiHi Island FSM ADD MUX MUL Cluster with area constraint  Use register banks:  Registers in each island are partitioned to k banks for 1 cycle, 2 cycle, … k cycle interconnect communication in each island  Highly regular 1 cycle 2 cycle k cycle

Jason Cong8 MCAS: Placement-Driven Architectural Synthesis Using RDR Architecture Register and port binding Datapath & FSM generation Floorplan constraints RTL VHDL files Multi-cycle path constraints CDFG C / VHDL CDFG generation + 2 * 3* * 7 * * 11 * RDR Arch. Spec. Target clock period Resource allocation Resource constraints -+ ** -- * * - * - * Interconnected Component Graph (ICG) Functional unit binding Mult1 Alu2 Mult2 Alu1 Interconnected Component Graph (ICG) Location information Scheduling-driven placement Reg. file … Alu1 1,5,10 … Reg. file … Mul2 3,7,12 … Alu2 2,6,9 Mul1 4,8,11 Placement-driven rebinding & scheduling Cycle1 Cycle2 Cycle3 Cycle4 Cycle5 Cycle6 Cycle7 * * * +- * -- * - * - Reg. file … Alu1 1,5,10 … Reg. file … Mul2 3,7,11 … Alu2 2,6,9 Mul1 4,8,12

Jason Cong9 Experimental Results (3)  Synopsys Behavioral Compiler setting: default (optimizing latency)  Average latency ratio of MCAS vs. BC: 69% n MCAS basic flow vs. Synopsys’ Behavioral Compiler (on Virtex-II) LatencyResource

Jason Cong10 Optimality Study of Large-Scale Circuit Placement Construction of Placement Example with Known Optimal (PEKO) [C. Chang et al, 2003] ? n Construct instances with known optimal using the characteristic of the original problem n First quantitative evaluation of the optimality of circuit placement problem n Existing placement algorithms can be 70% to 150% away from the optimal

Jason Cong11 High Interest in the Community Two EE Times articles coverage –Placement tools criticized for hampering IC designs [Feb’03] –IC placement benchmarks needed, researchers say [April’03] More than 60 downloads from our website –Cadence, IBM, Intel, Magma, Mentor Graphics, Synopsys, etc –CMU, SUNY, UCB, UCSB, UCSD, UIC, UMichgan, UWaterloo, etc Used in every placement since its publication

Jason Cong12 1. Synthesis & Verification  Hardware/Software Partition:  Propose a SSS based H/S partition algorithm (ASICON2003)  better solution than SA and less runtime than Tabu  High-level Synthesis:  Re-synthesis algorithm after floorplanning for timing optimization (ASICON2003)  Based on initial scheduling do floorplanning  After floorplanning do re-scheduling and re-allocation by force- balance method  Controller Synthesis:  A Heuristic State Minimization Algorithm For Incompletely Specified Finite State Machine (ASICON2003, JCST)

Jason Cong13 2. Floorplanning & Interconnect Planning  Based on proposed Corner Block List (CBL) representation propose several Extended Corner Block List, ECBL, CCBL and SUB-CBL to speed up floorplanning and handle more complicate L/T shaped and rectilinear shaped blocks.  Propose floorplanning algorithms with some geometric constraints, such as boundary, abutment, L/T shaped blocks.  Propose integrated floorplanning and buffer planning algorithms with consideration of congestion.  Using research results from UCLA on interconnect planning  About 30 papers published in DAC, ICCAD, ISPD, ASPDAC, ISCAS and Transactions.

Jason Cong14 3. P/G Network Analysis & Optimization  Propose an Area Minimization of Power Distribution Network Using Efficient Nonlinear Programming Techniques (ICCAD2001, accepted by IEEE Trans. On CAD)  Propose a decoupling capacitance optimization algorithm for Robust On-Chip Power Delivery (ASPDAC2004, ASICON2003) 4. Global Routing & Special Routing  Propose several congestion, timing, and both timing and congestion optimization global routing algorithms  Papers were published in ASPDAC, ISCAS, and IEEE Transactions.

Jason Cong15 5. Parasitic R/L/C Etraction  3-D R/C Extraction using Boundary Element Method (BEM)  Quasi-Multiple Medium (QMM) BEM algorithms  Hierarchical Block BEM (HBBEM) technique  Fast 3-D Inductance Extraction (FIE)  Papers were published in ASPDAC, ASICON and IEEE Transaction on MTT

Jason Cong16 Thrust 2 -- SOC Verification, Test, and Diagnosis (Led by Tim Cheng) Verification and Testing Enabling techniques for semi-formal functional verification Integrated framework for simulation, vector generation and model checking Testing and diagnosis for heterogeneous SOC Self-testing using on- chip programmable components Self-testing for on-chip analog/mixed-signal components New test techniques for deep-submicron embedded memories Scalable constraint-solving techniques Automatic/semi-automatic functional vector generation from HDL code

Tim Cheng17 Key Results - Verification Developed and released ATPG-based SAT solvers for circuits (Univ. of California, Santa Barbara) –Integrating structural ATPG and SAT techniques with new conflict learning –CSAT: Fast combinational solver (released on March 2003) Demonstrated X speedup over state-of-the-art SAT solvers on industrial test cases (reported by Intel and Calypto) Has been integrated into Intel’s FV verification system and a startup’s verification engine Publications: DATE2003 and DAC2003 –Satori2: Fast sequential solver (released on Dec. 2003) Demonstrated 10X-200X speedup over a commercial, sequential ATPG engine on public benchmark circuits Publications: ICCAD2003, HLDVT2003 and ASPDAC2004

Tim Cheng18 Key Results - Testing A new Statistical Delay Testing and Diagnosis framework consisting of five major components (UCSB): Defect Injection & Simulation Statistical Timing Analysis Framework (Cell-based characterization) Static Timing Analysis Dynamic Timing Simulator Path Filtering Critical Path Selection Diagnosis ATPG/Pattern Selection Selection/Generation of high quality tests for target paths [ITC’01][DATE 2004] Selection/Generation of high quality tests for target paths [ITC’01][DATE 2004]  Identifying tests that activate longer delay along the target path Delay fault diagnosis based on statistical timing model [DATE’03, VTS’03, DAC’03] Delay fault diagnosis based on statistical timing model [DATE’03, VTS’03, DAC’03]  Ref: Krstic, Wang, Cheng,& Abadir, DATE’03–Best Paper Award in Test Statistical timing analysis Statistical critical path selection [DAC’02,ICCAD’02]  Selecting statistical long & true paths whose tests maximize detection of parametric failures Path coverage metric [ASPDAC’03]  Estimating the quality of a path set

Tim Cheng19 Key Results - Testing On-Chip Jitter Extraction for Bit-Error-Rate (BER) Testing of Multi- GHz Signal (UCSB) –Using on-chip, single-shot measurement unit to sample signal periods for spectral analysis –Demonstrated, through simulation, accurate extraction of multiple sinusoids and random jitter components for a 3GHz signal –Publications: ASPDAC2004 and DATE2004

Jason Cong20 Thrust 3 – Design Driver: Network Security Processor (Led by Prof. C. W. Wu) Applications: IPSec, SSL, VPN, etc. Functionalities: –Public key: RSA, ECC –Secret key: AES –Hashing (Message authentication): HMAC (SHA-1/MD5) –Truly random number generator (FIPS 140-1,140-2 compliant) Target technology: 0.18  m or below Clock rate: 200MHz or higher (internal) 32-bit data and instruction word 10Gbps (OC192) Power: 1 to 10mW/MHz at 3V (LP to HP) Die size: 50mm 2 On-chip bus: AMBA (Advanced Microcontroller Bus Architecture)

Jason Cong21 Encryption Modules (PKEM) Public key encryption module –Operations: 32-bit word-based modular multiplication Multiplication over GF(p) and GF(2 m ) An RSA cryptography engine with small area overhead and high speed Scalable word-width TSMC 0.35μm 34K gates (1.7×1.8 mm 2 ) 100MHz clock Scalable key length Throughput –512-bit key: 1.79Kbps/MHz –1024-bit key: 470bps/MHz

Jason Cong22 Encryption Modules (SKEM) Secret key encryption module –Operations: Matrix operations, manipulation AES cryptography 32-bit external interface 58K gates Over 200MHz clock Throughput: 2Gbps Support key length of 128/192/256 bits Technology TSMC 0.25  m CMOS Package128CQFP Core Size 1,279 x 1,271  m 2 Gate Count63.4K Max. Freq.250MHz Throughput Gbps (128-bit key) Gbps (196-bit key) Gbps (256-bit key)

Jason Cong23 Journal Publications C.-T. Huang and C.-W. Wu, ``High-speed easily testable Galois-field inverter'', IEEE Trans. Circuits and Systems II: Analog and Digital Signal Processing, vol. 47, no. 9, pp , Sept S.-A. Hwang and C.-W. Wu, ``Unified VLSI systolic array design for LZ data compression'', IEEE Trans. VLSI Systems, vol. 9, no. 4, pp , Aug C.-H. Wu, J.-H. Hong, and C.-W. Wu, ``VLSI design of RSA cryptosystem based on the Chinese Remainder Theorem'', J. Inform. Science and Engineering, vol. 17, no. 6, pp , Nov J.-H. Hong and C.-W. Wu, ``Cellular array modular multiplier for the RSA public- key cryptosystem based on modified Booth's algorithm'', IEEE Trans. VLSI Systems, vol. 11, no. 3, pp , June C.-P. Su, T.-F. Lin, C.-T. Huang, and C.-W. Wu, ``A high-throughput low-cost AES processor'', IEEE Communications Magazine, vol. 41, no. 12, pp , Dec

Jason Cong24 Conference Publications J.-H. Hong and C.-W. Wu, ``Radix-4 modular multiplication and exponentiation algorithms for the RSA public-key cryptosystem'', in Proc. Asia and South Pacific Design Automation Conf. (ASP-DAC), Yokohama, Jan. 2000, pp J.-H. Hong, P.-Y. Tsai, and C.-W. Wu, ``Interleaving schemes for a systolic RSA public-key cryptosystem based on an improved Montgomery's algorithm'', in Proc. 11th VLSI Design/CAD Symp., Pingtung, Aug. 2000, pp C.-H. Wu, J.-H. Hong, and C.-W. Wu, ``An RSA cryptosystem based on the Chinese Remainder Theorem'', in Proc. 11th VLSI Design/CAD Symp., Pingtung, Aug. 2000, pp C.-H. Wu, J.-H. Hong, and C.-W. Wu, ``RSA cryptosystem design based on the Chinese Remainder Theorem'', in Proc. Asia and South Pacific Design Automation Conf. (ASP-DAC), Yokohama, Jan. 2001, pp Y.-C. Lin, C.-P. Su, C.-W. Wang, and C.-W. Wu, ``A word-based RSA public-key crypto-procesoor core'', in Proc. 12th VLSI Design/CAD Symp., Hsinchu, Aug T.-F. Lin, C.-P. Su, C.-T. Huang, and C.-W. Wu, ``A high-throughput low-cost AES cipher chip'', in Proc. 3rd IEEE Asia- Pacific Conf. ASIC, Taipei, Aug. 2002, pp Y.-T. Lin, C.-P. Su, C.-T. Huang, C.-W. Wu, S.-Y. Huang, and T.-Y. Chang, ``Low-power embedded memory architecture design for SOC'', in Proc. 13th VLSI Design/CAD Symp., Taitung, Aug. 2002, pp M.-C. Sun, C.-P. Su, C.-T. Huang, and C.-W. Wu, ``Design of a scalable RSA and ECC crypto-processor'', in Proc. Asia and South Pacific Design Automation Conf. (ASP-DAC), Kitakyushu, Jan. 2003, pp , (Best Paper Award). C.-P. Su, T.-F. Lin, C.-T. Huang, and C.-W. Wu, ``A highly efficient AES cipher chip'', in Proc. Asia and South Pacific Design Automation Conf. (ASP-DAC), Kitakyushu, Jan. 2003, pp , (Design Contest Special Feature Award). J.-H. Hong, C.-L. Liu, B.-Y. Tsai, and C.-W. Wu, ``A radix-4 modular multiplier for fast RSA public-key cryptosystem'', in Proc. 14th VLSI Design/CAD Symp., Hualien, Aug. 2003, pp M.-Y. Wang, C.-P. Su, C.-T. Huang, and C.-W. Wu, ``An HMAC processor with integrated SHA-1 and MD5 algorithms'', in Proc. Asia and South Pacific Design Automation Conf. (ASP-DAC), Yokohama, Jan (to appear).