This page is organized by type of paper.
- Tutorial and Survey — good introduction starting point papers and broad surveys
- Pre-history — papers before there were FPGAs
- Classic — classic paper on a particular aspect
- Contemporary — papers covering key issues and aspects of modern FPGAs
Tutorial and Survey
- Stephen Brown and Jonathan Rose. FPGA and CPLD Architectures: A Tutorial. In IEEE Design and Test of Computers, Volume 13, Number 2, pp. 42–57, 1996. [bibtex entry] [DOI 10.1109/54.500200]
An approachable tutorial for a general audience on FPGA and CPLD architectures.
- André DeHon. The Density Advantage of Configurable Computing. In IEEE Computer, Volume 33, Number 4, pp. 41–49, 2000. [bibtex entry] [DOI 10.1109/2.839320]
Broad-audience article comparing FPGA, processor, and custom logic denstities for accelerating computing applications.
- Jan Gray. Building a RISC System in an FPGA. In Circuit Cellar Ink, Number 116–118, March, April, May, 2000. [bibtex entry] [URL]
Tutorial on building custom processors optimized for implementation on FPGAs.
- Katherine Compton and Scott Hauck. Reconfigurable Computing: a Survey of Systems and Software. In ACM Computing Surveys, Volume 34, Number 2, pp. 171—210, 2002. [bibtex entry] [DOI 10.1145/508352.508353]
An excellent survey paper on reconfigurable computing.
- T.J. Todman, G.A. Constantinides, S.J.E. Wilton, O. Mencer, W. Luk, and P.Y.K. Cheung. Reconfigurable Computing: Architectures and Design Methods. InComputers and Digital Techniques, IEE Proceedings, Volume 152, Number 2, pp. 193—207, March, 2005. [bibtex entry] [DOI 10.1049/ip-cdt:20045086]
A recent survey paper on reconfigurable computing platforms and design with a wealth of references.
- Deming Chen, Jason Cong, and Peichen Pan FPGA Design Automation: A Survey. In Foundations and Trends in Electronic Design Automation, Volume 1, Number 3, pp. 195–330, November, 2006. [bibtex entry] [DOI 10.1561/1000000003]
Modern survey of FPGA CAD algorithms.
- Ian Kuon, Russell Tessier, and Jonathan Rose. FPGA Architecture: Survey and Challenges. In Foundations and Trends in Electronic Design Automation, Volume 2, Number 2, pp. 135–253, 2007. [bibtex entry] [DOI 10.1561/1000000005]
Modern survey of FPGA Architecture.
- Scott Hauck and André DeHon. Reconfigurable Computing: The Theory and Practice of FPGA-Based Computation. Elsevier, 2008. [bibtex entry] [URL]
Comprehensive book that covers all aspects of computing with FPGAs and FPGA-like components including device architecture, programming approaches, CAD flows, design issues, and sample applications.
- Gerald Estrin, B. Bussell, R. Turn, and J. Bibb. Parallel Processing in a Restructurable Computer System. In IEEE Transactions on Electronic Computers, Volume 12, Issue 6, pp. 747–754, December, 1963. [bibtex entry] [DOI 10.1109/PGEC.1963.263558]
Early predecessor to reconfigurable computers; this was before integrated circuits and their “configurations” required physically moving wires, but the goal was the same. Modern FPGAs make this vision practical.
- Monty Denneau. The Yorktown Simulation Engine. In Proceedings of the 19th Design Automation Conference, p. 55–59, 1982. [bibtex entry] [URL]
This was a pre-FPGA logic simulation engine that was also used to simulate logic before building hardware. It includes most of the ideas behind multicontext FPGAs.
- William S. Carter, Khue Duong, Ross H. Freeman, Hung-Cheng Hsieh, Jason Y. Ja, John E. Mahoney, Luan T. Ngo, and Shelly L. Sze. A User Programmable Reconfigurable Logic Array. In Proceedings of the IEEE Custom Integrated Circuits Conference, pp. 233–235, May 1986. [bibtex entry]
First peer-review, public description of a commercial FPGA.
- Jonathan Rose and Robert Francis and David Lewis and Paul Chow. Architecture of Field-Programmable Gate Arrays: The Effect of Logic Block Functionality on Area Efficiency. In IEEE Journal of Solid-State Circuits, Volume 25, Number 5, pp. 1217–1225, October, 1990. [bibtex entry] [DOI 10.1109/4.62145]
Why did we start with 4-LUT FPGAS? But more than that, this is beautiful example of formulating a clean question about architecture, defining a parameterized space, identifying a cost model, and explorting the space to find the best option.
- Maya Gokhale, William Holmes, Andrew Kopser, Sara Lucas, Ronald Minnich, Douglas Sweely, and Daniel Lopresti. Building and Using a Highly Programmable Logic Array. In IEEE Computer, Volume 24, Number 1, pp. 81–89, 1991. [bibtex entry] [DOI 10.1109/2.67197]
One of the early FPGA Computing systems that demonstrated performance exceeding supercomputers on a specialized problem (DNA Sequence matching) using a board of FPGAs. The entire capacity of one of these boards is smaller than today’s midrange FPGAs.
- I. Page and W. Luk. Compiling Occam into FPGAs. In FPGAs, pp. 271–283, Abingdon EE&CS Books, 1991. [bibtex entry] [URL]
Describes methodology to compile from a high-level language to FPGA, a precursor to Handel-C.
- Dev C. Chen and Jan M. Rabaey. A Reconfigurable Multiprocessor IC for Rapid Prototyping of Algorithmic-Specific High-Speed DSP Data Paths. In IEEE Journal of Solid-State Circuits, Volume 27, Number 12, pp. 1895–1904, December, 1992. [bibtex entry] [DOI 10.1109/4.173120]
Early coarse-grained reconfigurable device initially intended for rapid prototyping of DSP algorithms; PADDI has 16b functional units and 8 configuration contexts which operate in VLIW fashion.
- Jonathan Babb, Russell Tessier, and Anant Agarwal. Virtual Wires: Overcoming Pin Limitations in FPGA-based Logic Emulators. In Proceedings of the IEEE Workshop on FPGAs for Custom Computing Machines, pp. 142–151, April, 1993. YEAR. [bibtex entry] [DOI 10.1109/FPGA.1993.279469]
Time-multiplexing the FPGA I/O to better balance I/O bandwith with internal FPGA capacity in multi-FPGA systems.
- Jason Cong and Yuzheng Ding. FlowMap: An Optimal Technology Mapping Algorithm for Delay Optimization in Lookup-Table Based FPGA Designs. In IEEE Transactions on Computer-Aided Design, Volume 13, Issue 1, pp. 1–12, January, 1994. [bibtex entry] [DOI 10.1109/43.273754]
How to cover logic into LUTs; nice observation that the problem can be reframed from logic packing to IO cuts. Use of dynamic programming and max flow is algorithmically elegant. There are a wealth of improvements and more sophisticated versions since this, but it’s worth starting here for the cleanness of this basic problem formulation.
- Larry McMurchie and Carl Ebeling. PathFinder: A Negotiation-Based Performance-Driven Router for FPGAs. In Proceedings of the International Symposium on Field-Programmable Gate Arrays, pp. 111–117, 1995. [bibtex entry] [DOI 10.1145/201310.201328]
The basic routing algorithm around which virtually all FPGA routing is built today. Dismisses with separate global/detail phases and uses adaptive costs to sort out congestion.
- Rick Amerson, Richard Carter, W. Bruce Culbertson, Phil Kuekes, and Greg Snider. Teramac—Configurable Custom Computing. In Proceedings of the IEEE Workshop on FPGAs for Custom Computing Machines, pp. 32–38, April, 1995. [bibtex entry] [DOI 10.1109/FPGA.1995.477406]
A reconfigurable system based on a custom FPGA-like design aimed at rapid application mapping.
- John Villasenor, and Chris Jones, and Brian Schoner. Video Communications using Rapidly Reconfigurable Hardware. In IEEE Transactions on Circuits and Systems for Video Technology, Volume 5, Number 6, pp. 565–567, December 1995. [bibtex entry] [DOI 10.1109/76.475899]
Early article articulating and demonstrating the idea of using rapid Run-Time Reconfiguration in order to run large tasks on smaller FPGA systems.
- André DeHon. DPGA Utilization and Application. In Proceedings of the International Symposium on Field-Programmable Gate Arrays, pp. 115–121, February, 1996. [bibtex entry] [DOI 10.1145/1145/228370.228387]
What would you do with a multicontext FPGA and what benefits does it offer?
- Ethan Mirsky and André DeHon. MATRIX: A Reconfigurable Computing Architecture with Configurable Instruction Distribution and Deployable Resources.In Proceedings of the IEEE Symposium on FPGAs for Custom Computing Machines, pp. 157–166, April, 1996. [bibtex entry] [DOI 10.1109/FPGA.1996.564808]
Early coarse-grained reconfigurable architecture that allows flexible organization of units and instruction distribution. The basic element is a composable 8b funcntional unit with a 256 byte memory/register file that can also be used to hold dynamic instructions.
- Carl Ebeling and Darren Cronquist and Paul Franklin. RaPiD—Reconfigurable Pipelined Datapath. In Proceedings of the International Conference on Field-Programmable Logic and Applications (published as LNCS-1142), pp. 126–135, Springer, 1997. [bibtex entry] [DOI 10.1007/3-540-61730-2_13] [PS]
Early coarse-grained, domain-specific reconfigurable architecture. RaPiD has 16b functional units arranged in a 1D linear array.
- Jean E. Vuillemin, Patrice Bertin, Didier Roncin, Mark Shand, Hervé Touati, and Philippe Boucard. Programmable Active Memories: Reconfigurable Systems Come of Age. In IEEE Transactions on Very Large Scale Integration (VLSI) Systems, Volume 4, Number 1, pp. 56–69, March, 1996. [bibtex entry] [DOI 10.1109/92.486081]
One of the earliest FPGA computing systems. PAM demonstrated impressive performance from a board of FPGAs on a range of applications; the entire capacity of a PAM board is smaller than today’s midrange FPGAs.
- Brian Von Herzen. Signal Processing at 250 MHz using High-Performance FPGAs. In Proceedings of the International Symposium on Field-Programmable Gate Arrays, pp. 62–68, 1997. [bibtex entry] [DOI 10.1145/258305.258313]
Early and inspiring demonstration that FPGAs can operate productively at very high clock rates by paying careful attention to spatial locality and pipelining.
- Jan Rabaey. Reconfigurable Computing: The Solution to Low Power Programmable DSP. In Proceedings of the 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing, Volume 1, pp. 275–278, April, 1997. [bibtex entry] [DOI 10.1109/ICASSP.1997.599622]
Early paper making the case for the energy efficiency of reconfigurable architectures and including an early comparison of energy among processors, FPGAs, and ASICs.
- W. Bruce Culbertson, Rick Amerson, Richard Carter, Phil Kuekes, and Greg Snider. Defect Tolerance on the TERAMAC Custom Computer. In Proceedings of the IEEE Symposium on FPGAs for Custom Computing Machines, pp. 116–123, April, 1997. [bibtex entry] [DOI 10.1109/FPGA.1997.624611]
Shows how reconfigurability of the FPGA can be used to to map around defects in the fabricated IC or board-level system. An early paper giving a full-system demonstration of the benefits of component-specific mapping.
- Steve Trimberger and Dean Carberry and AndersJohnson and Jennifer Wong A Time-Multiplexed FPGA. In Proceedings of the IEEE Symposium on FPGAs for Custom Computing Machines, pp. 22–28, April, 1997. [bibtex entry] [DOI 10.1109/FPGA.1997.624601]
How to add multicontext support to a mostly conventional FPGA architecture base.
- Vaughn Betz and Jonathan Rose. VPR: A New Packing, Placement, and Routing Tool for FPGA Research. In Proceedings of the International Conference on Field-Programmable Logic and Applications (published as LNCS-1304), pp. 213–222, Springer, 1997. [bibtex entry] [URL]
A good placer coupled with a good version of Pathfinder and targeted at Island-style FPGAs. The free availability of this high-quality tool has provided a baseline standard for FPGA architectural work for over a decade.
- Peichen Pan and Chih-Chang Lin. A New Retiming-based Technology Mapping Algorithm for LUT-based FPGAs. In Proceedings of the International Symposium on Field-Programmable Gate Arrays, pp. 35–42, February, 1998. [bibtex entry] [DOI 10.1145/275107.275118]
Optimally solve LUT mapping and retiming simultaneously; there are so few things we can solve optimally, and so few things we can afford to address together, it’s refreshing see formulations where you can provide optimal results across multiple traditional levels of decomposition. As with flowmap, there are later papers which take this further and provide more efficient and general solutions, but the earlier papers introduce the cleanest problems and key ideas.
- Vaughn Betz and Jonathan Rose. How Much Logic Should Go in an FPGA Logic Block? In IEEE Design and Test of Computers, Volume 15, Number 1, pp. 10–15, 1998. [bibtex entry] [DOI 10.1109/54.655177]
A paper explaining the move to “Island-Style” FPGAs. Why do we use clusters with multiple LUTs?
- Vaughn Betz, Jonathan Rose, and Alexander Marquardt. Architecture and CAD for Deep-Submicron FPGAs. Kluwer Academic Publishers, 1999. [bibtex entry] [URL]
Classic book on FPGA architecture and CAD. Describes VPR and island style FPGAs. While the technology is dated, this book provides the best single introduction to FPGA organization and implementation issues as well as a description of the popular clustering, placement, and routing algorithms using for physical mapping of designs to FPGAs.
- André DeHon. Balancing Interconnect and Computation in Reconfigurable Computing Array (or, why you don’t really want 100% LUT utilization). InProceedings of the International Symposium on Field-Programmable Gate Arrays, pp. 69–78, February, 1999. [bibtex entry] [DOI 10.1145/296399.296431]
Since interconnect is the dominant area (and delay and energy) contributor on FPGAs, architectural optimizations which try to provide adequate interconnec to use all the logic may quite inefficient; this paper turns the question around and asks how the two should be balanced together. This provides a clean, parameterized formulation of this tradeoff.
- Vaughn Betz and Jonathan Rose. FPGA Routing Architecture: Segmentation and Buffering to Optimize Speed and Density. In Proceedings of the International Symposium on Field-Programmable Gate Arrays, pp. 59–68, February, 1999. [bibtex entry] [DOI 10.1145/296399.296428]
Why FPGA tracks are segmented, and details on the tradeoffs involved.
- William Tsu, Kip Macy, Atul Joshi, Randy Huang, Norman Walker, Tony Tung, Omid Rowhani, Varghese George, John Wawrzynek, and André DeHon HSRA: High-Speed, Hierarchical Synchronous Reconfigurable Array In Proceedings of the International Symposium on Field-Programmable Gate Arrays, pp. 125–134–78, February, 1999. [bibtex entry] [DOI 10.1145/296399.296442]
Why should an FPGA run slower than a processor? Shows how adding pipelining to interconnect allows tools to target high-throughput operations.
- Timothy Callahan and John Hauser and John Wawrzynek. The Garp Architecture and C Compiler. In IEEE Computer, Volume 33, Number 4, pp. 62–69, 2000. [bibtex entry] [DOI 10.1109/2.839323]
Details one of the earliest architectures for using a reconfigurable array as a coprocessor attached to a microprocessor, including a compiler capable of automatically extracting application kernels for execution on the reconfigurable array.
- Seth C. Goldstein, Herman Schmit, Mihai Budiu, Srihari Cadambi, Matthew Moe, and R. Reed Taylor. PipeRench: a reconfigurable architecture and compiler. InIEEE Computer, Volume 33, Number 4, pp. 70–77, 2000. [bibtex entry] [DOI 10.1109/2.839324]
Coarse-grained reconfigurable with a virtual pipeline model that allows hardware scaling and fast application mapping.
- Keith Underwood. FPGAs vs. CPUs: Trends in Peak Floating-Point Performance. In Proceedings of the International Symposium on Field-Programmable Gate Arrays, pp. 171–180, February, 2004. [bibtex entry] [DOI 10.1145/968280.968305]
Article pointing out that FPGA performance on floating point was catching up with microprocessors and on track to surpass micoprocessor floating-point performance for many tasks.
- Guy Lemieux, Edmund Lee, and Marvin Tom and Anthony Yu. Directional and Single-Driver Wires in FPGA Interconnect. In Proceedings of the International Conference on Field-Programmable Technology, pp. 41–48, December, 2004. [bibtex entry] [DOI 10.1109/FPT.2004.1393249]
Why it no longer makes sense to have mult-driver, bidirectional wires.
- André DeHon, Yury Markovsky, Eylon Caspi, Michael Chu, Randy Huang, Stylianos Perissakis, Laura Pozzi, Joseph Yeh, and John Wawrzynek. Stream Computations Organized for Reconfigurable Execution In Journal of Microprocessors and Microsystems, Volume 30, Number 6, pp. 334–354, September, 2006. [bibtex entry] [DOI 10.1016/j.micpro.2006.02.009] [URL]
Scalable compute model for reconfigurable systems based on stream-connected concurrent operators. Illustrates how we can design applications at a high level and efficiently and automatically map them to physical hardware platforms with a wide-range capacities.
- P. H. W. Leong, M. P. Leong, O. Y. H. Cheung, T. Tung, C. M. Kwok, M. Y. Wong, and K. H. Lee. Pilchard—A Reconfigurable Computing Platform with Memory Slot Interface. In Proceedings of the IEEE Symposium on Field-Programmable Custom Computing Machines, pp. 170–179, April, 2001. [bibtex entry] [DOI 10.1109/FCCM.2001.36]
A reconfigurable computing platform with a memory slot interface to improve transfer latency. A similar approach is used in DRC machines.
- David Lewis, Elias Ahmed, Gregg Baeckler, Vaughn Betz, Mark Bourgeault, David Cashman, David Galloway, Mike Hutton, Chris Lane, Andy Lee, Paul Leventis, Sandy Marquardt, Cameron McClintock, Ketan Padalia, Bruce Pedersen, Giles Powell, Boris Ratchev, Srinivas Reddy, Jay Schleicher, Kevin Stevens, Richard Yuan, Richard Cliff and Jonathan Rose. The Stratix II Logic and Routing Architecture. In Proceedings of the International Symposium on Field-Programmable Gate Arrays, pp. 14–20, February, 2005. [bibtex entry] [DOI 10.1145/1046192.1046195]
A contemporary FPGA architecture.
- Chen Chang, John Wawrzynek, and Robert W. Brodersen. BEE2: A High-End Reconfigurable Computing System. In IEEE Design and Test of Computers, Volume 22, Number 2, pp. 114—125, 2005. [bibtex entry] [DOI 10.1109/MDT.2005.30]
A contemporary reconfigurable computing platform.
- C. T. Chow, L. S. M. Tsui, Philip H. W. Leong, Wayne Luk, and Steve J. E. Wilton. Dynamic voltage scaling for commercial FPGAs. In Proceedings of the International Conference on Field-Programmable Technology, pp. 173–180, December, 2005. [bibtex entry] [DOI 10.1109/FPT.2005.1568543]
Shows how to exploit dynamic voltage scaling on off-the-shelf FPGAs.
- Ian Kuon and Jonathan Rose. Measuring the Gap Between FPGAs and ASICs. In IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, Volume 26, Number 2, pp. 203–215, February, 2007. [bibtex entry] [DOI 10.1109/TCAD.2006.884574]
Modern effort to quantify the relative area, power, and delay of FPGAs compared to ASICs
- John Wawrzynek, David Patterson, Mark Oskin, Shih-Lien Lu, Christoforos Kozyrakis, James C. Hoe, Derek Chiou, and Krste Asanovic. RAMP: Research Accelerator for Multiple Processors. In IEEE Micro, Volume 27, Number 2, pp. 46—57, 2007. [bibtex entry] [DOI 10.1109/MM.2007.39]
An important, modern reconfigurable platform for emulation and simulation. With the growth in FPGA capacity, this effort can contemplate the emulation of systems containing hundreds to thousands of processor cores, where each FPGA is modeling several processors.
- Shih-Lien L. Lu and Peter Yiannacouras and Taeweon Suh and Rolf Kassa and Michael Konow. A Desktop Computer with a Reconfigurable Pentium. InTransactions on Reconfigurable Technology and Systems, Volume 1, Number 1, March, 2008. [bibtex entry] [DOI 10.1145/1331897.1331901]
Demonstration that a Pentium processor can be implemented on less than half of a modern FPGA.
- Jason Luu, Ian Kuon, Peter Jamieson, Ted Campbell, Andy Ye, Wei Mark Fang, and Jonathan Rose. VPR 5.0: FPGA CAD and Architecture Exploration Tools with Single-Driver Routing, Heterogeneity and Process Scaling. In Proceedings of the International Symposium on Field-Programmable Gate Arrays, pp. 133–142, February, 2009. [bibtex entry] [DOI 10.1145/1508128.1508150]
Update of key open-source physical design tool including updated studies on LUT size, cluster size, and segmentation.