Nachiket Kapre
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Nachiket Kapre has a total of 43 co-authors. Here are the top co-authors by number of shared publications:
Andre DeHon
University of Pennsylvania, Philadelphia, PA 19104
7 shared publicationsMichael Delorimier
University of Pennsylvania, Philadelphia
3 shared publicationsNikil Mehta
University of Pennsylvania, Philadelphia
3 shared publicationsAbid Rafique
Electrical and Electronic Engineering Department, Imperial College London, UK
2 shared publicationsRaphael Rubin
California Institute of Technology
2 shared publicationsA. DeHon
1 shared publication
Latest Publications
- 1 September 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 165-173
Jack of All Scales: A Versatile FPGA Tensor Block for MXFP Precisions
- 1 September 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 165-173
Jack of All Scales: A Versatile FPGA Tensor Block for MXFP Precisions
- Open Access
- 13 February 2020
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Reconfigurable Technology and Systems
- Vol. 13 (2) , 1-35
HopliteBuf
HopliteBuf is a deflection-free, low-cost, and high-speed FPGA overlay Network-on-chip (NoC) with stall-free buffers. It is an FPGA-friendly 2D unidirectional torus topology built on top of HopliteRT overlay NoC. The stall-free buffers in HopliteBuf are supported by static analysis tools based on network calculus that help determine worst-case FIFO occupancy bounds for a prescribed workload. We implement these FIFOs using cheap LUT SRAMs (Xilinx SRL32s and Intel MLABs) to reduce cost. HopliteBuf is a hybrid microarchitecture that combines the performance benefits of conventional buffered NoCs by using stall-free buffers with the cost advantages of deflection-routed NoCs by retaining the lightweight unidirectional torus topology structure. We present two design variants of the HopliteBuf NoC: (1) single corner-turn FIFO ( W → S ) and (2) dual corner-turn FIFO ( W → S + N ). The single corner-turn ( W → S ) design is simpler and only introduces a buffering requirement for packets changing dimension from the X ring to the downhill Y ring (or West to South). The dual corner-turn variant requires two FIFOs for turning packets going downhill ( W → S ) as well as uphill ( W → N ). The dual corner-turn design overcomes the mathematical analysis challenges associated with single corner-turn designs for communication workloads with cyclic dependencies between flow traversal paths at the expense of a small increase in resource cost. Our static analysis delivers bounds that are not only better (in latency) than HopliteRT but also tighter by 2−3×. Across 100 randomly generated flowsets mapped to a 5×5 system size, HopliteBuf is able to route a larger fraction of these flowsets with W → S bkp backpressure-based router, we observe that our HopliteBuf solution offers 25--30% better feasibility at 30--40% lower LUT cost.
- 13 February 2020
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Reconfigurable Technology and Systems
- Vol. 13 (2) , 1-35
Open AccessHopliteBuf
HopliteBuf is a deflection-free, low-cost, and high-speed FPGA overlay Network-on-chip (NoC) with stall-free buffers. It is an FPGA-friendly 2D unidirectional torus topology built on top of HopliteRT overlay NoC. The stall-free buffers in HopliteBuf are supported by static analysis tools based on network calculus that help determine worst-case FIFO occupancy bounds for a prescribed workload. We implement these FIFOs using cheap LUT SRAMs (Xilinx SRL32s and Intel MLABs) to reduce cost. HopliteBuf is a hybrid microarchitecture that combines the performance benefits of conventional buffered NoCs by using stall-free buffers with the cost advantages of deflection-routed NoCs by retaining the lightweight unidirectional torus topology structure. We present two design variants of the HopliteBuf NoC: (1) single corner-turn FIFO ( W → S ) and (2) dual corner-turn FIFO ( W → S + N ). The single corner-turn ( W → S ) design is simpler and only introduces a buffering requirement for packets changing dimension from the X ring to the downhill Y ring (or West to South). The dual corner-turn variant requires two FIFOs for turning packets going downhill ( W → S ) as well as uphill ( W → N ). The dual corner-turn design overcomes the mathematical analysis challenges associated with single corner-turn designs for communication workloads with cyclic dependencies between flow traversal paths at the expense of a small increase in resource cost. Our static analysis delivers bounds that are not only better (in latency) than HopliteRT but also tighter by 2−3×. Across 100 randomly generated flowsets mapped to a 5×5 system size, HopliteBuf is able to route a larger fraction of these flowsets with W → S bkp backpressure-based router, we observe that our HopliteBuf solution offers 25--30% better feasibility at 30--40% lower LUT cost.
- 1 December 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 64-71
HopliteRT: An efficient FPGA NoC for real-time applications
Overlay NoCs, such as Hoplite, are cheap to implement on an FPGA but provide no bounds on worst-case routing latency of packets traversing the NoC due to deflection routing. In this paper, we show how to adapt Hoplite to enable calculation of precise upper bounds on routing latency by modifying the routing function to prioritize deflections, and by regulating the injection of packets to meet certain throughput and burstiness constraints. We provide an analytical model for computing end-to-end latency in the form of (1) in-flight time in the network T f , and (2) waiting time at the source node T s . To bound in-flight time in an m × m NoC, we modify the routing function and switching crossbar richness in the Hoplite router to deliver T f = ΔX + ΔΥ + (AY × m) + 2 where AX and ΔΥ are differences of the source and destination address co-ordinates of the packet. To bound the waiting time at the source, we add a Token Bucket regulator with rate ρ i and burstiness σ i for each flow f i of node (x, y) to deliver (⌈1/ρ i ⌉ - 1) + T s : T s = ⌈σ(Γ C f )/1-σ(Γ C f );⌉ which depends on the regulator period 1/ρ i , burstiness a and the rate ρ of all interfering flows Γ C f . A 64b implementation of our HopliteRT router requires ≈4% fewer LUTs, and similar number of FFs compared to the original Hoplite router. We also need two small counters at each client port for regulating injection. We evaluate our model and RTL implementation across synthetic traffic patterns and observe behavior that conforms with the analytical bounds.
- 1 December 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 64-71
HopliteRT: An efficient FPGA NoC for real-time applications
Overlay NoCs, such as Hoplite, are cheap to implement on an FPGA but provide no bounds on worst-case routing latency of packets traversing the NoC due to deflection routing. In this paper, we show how to adapt Hoplite to enable calculation of precise upper bounds on routing latency by modifying the routing function to prioritize deflections, and by regulating the injection of packets to meet certain throughput and burstiness constraints. We provide an analytical model for computing end-to-end latency in the form of (1) in-flight time in the network T f , and (2) waiting time at the source node T s . To bound in-flight time in an m × m NoC, we modify the routing function and switching crossbar richness in the Hoplite router to deliver T f = ΔX + ΔΥ + (AY × m) + 2 where AX and ΔΥ are differences of the source and destination address co-ordinates of the packet. To bound the waiting time at the source, we add a Token Bucket regulator with rate ρ i and burstiness σ i for each flow f i of node (x, y) to deliver (⌈1/ρ i ⌉ - 1) + T s : T s = ⌈σ(Γ C f )/1-σ(Γ C f );⌉ which depends on the regulator period 1/ρ i , burstiness a and the rate ρ of all interfering flows Γ C f . A 64b implementation of our HopliteRT router requires ≈4% fewer LUTs, and similar number of FFs compared to the original Hoplite router. We also need two small counters at each client port for regulating injection. We evaluate our model and RTL implementation across synthetic traffic patterns and observe behavior that conforms with the analytical bounds.
- 1 December 2015
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
G-DMA: improving memory access performance for hardware accelerated sparse graph computation
Scatter-gather direct memory access (DMA) transfers can be used to efficiently fetch graph memory data for onchip processing of graph applications. We present a hardware controlled graph DMA engine which can operate autonomously without the need for CPU interaction. Graph processing algorithms can asynchronously request graph data which is fetched from memory and streamed to the processing core. An implementation of Dijkstra's shortest path algorithm operating on real-world social network graph data is used to demonstrate the operation of our hardware controlled graph DMA engine. Our system shows a 38% decrease in processing time over a standard CPU implementation and up to an 11% improvement over a hardware implementation using standard AXI4 read requests.
- 1 December 2015
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
G-DMA: improving memory access performance for hardware accelerated sparse graph computation
Scatter-gather direct memory access (DMA) transfers can be used to efficiently fetch graph memory data for onchip processing of graph applications. We present a hardware controlled graph DMA engine which can operate autonomously without the need for CPU interaction. Graph processing algorithms can asynchronously request graph data which is fetched from memory and streamed to the processing core. An implementation of Dijkstra's shortest path algorithm operating on real-world social network graph data is used to demonstrate the operation of our hardware controlled graph DMA engine. Our system shows a 38% decrease in processing time over a standard CPU implementation and up to an 11% improvement over a hardware implementation using standard AXI4 read requests.
- Open Access
- 1 August 2012
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 7705 (1946147X) , 443-450
Enhancing performance of Tall-Skinny QR factorization using FPGAs
Communication-avoiding linear algebra algorithms with low communication latency and high memory bandwidth requirements like Tall-Skinny QR factorization (TSQR) are highly appropriate for acceleration using FPGAs. TSQR parallelizes QR factorization of tall-skinny matrices in a divide-and-conquer fashion by decomposing them into sub-matrices, performing local QR factorizations and then merging the intermediate results. As TSQR is a dense linear algebra problem, one would therefore imagine GPU to show better performance. However, the performance of GPU is limited by the memory bandwidth in local QR factorizations and global communication latency in the merge stage. We exploit the shape of the matrix and propose an FPGA-based custom architecture which avoids these bottlenecks by using high-bandwidth on-chip memories for local QR factorizations and by performing the merge stage entirely on-chip to reduce communication latency. We achieve a peak double-precision floating-point performance of 129 GFLOPs on Virtex-6 SX475T. A quantitative comparison of our proposed design with recent QR factorization on FPGAs and GPU shows up to 7.7× and 12.7× speed up respectively. Additionally, we show even higher performance over optimized linear algebra libraries like Intel MKL for multi-cores, CULA for GPUs and MAGMA for hybrid systems.
- 1 August 2012
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 7705 (1946147X) , 443-450
Open AccessEnhancing performance of Tall-Skinny QR factorization using FPGAs
Communication-avoiding linear algebra algorithms with low communication latency and high memory bandwidth requirements like Tall-Skinny QR factorization (TSQR) are highly appropriate for acceleration using FPGAs. TSQR parallelizes QR factorization of tall-skinny matrices in a divide-and-conquer fashion by decomposing them into sub-matrices, performing local QR factorizations and then merging the intermediate results. As TSQR is a dense linear algebra problem, one would therefore imagine GPU to show better performance. However, the performance of GPU is limited by the memory bandwidth in local QR factorizations and global communication latency in the merge stage. We exploit the shape of the matrix and propose an FPGA-based custom architecture which avoids these bottlenecks by using high-bandwidth on-chip memories for local QR factorizations and by performing the merge stage entirely on-chip to reduce communication latency. We achieve a peak double-precision floating-point performance of 129 GFLOPs on Virtex-6 SX475T. A quantitative comparison of our proposed design with recent QR factorization on FPGAs and GPU shows up to 7.7× and 12.7× speed up respectively. Additionally, we show even higher performance over optimized linear algebra libraries like Intel MKL for multi-cores, CULA for GPUs and MAGMA for hybrid systems.
- 1 April 2012
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 77-84
FX-SCORE: A Framework for Fixed-Point Compilation of SPICE Device Models Using Gappa++
Automated, offline precision-analysis of dataflow computation containing elementary functions (e.g. exp) and if-then-else control flow operations enables accurate fixed-point FPGA implementation of SPICE device equations. We perform interval analysis of these equations using Gappa++ to statically compare error bounds of fixed-point and double-precision implementations. This is possible due to the limited dynamic range of physical voltage, current and conductance quantities in a SPICE simulation of real-world circuits. In contrast to previous custom-precision SPICE device mappings, our fixed-point implementation has the same accuracy as double-precision implementation when compared to ideal arithmetic (reals). To deliver these implementations we develop FX-SCORE, a high-level framework based on the SCORE streaming FPGA framework, that automatically generates Gappa++ scripts and AutoESL circuits to explore the cost-quality tradeoffs of Fixed-point FPGA implementations. Using our methodology, we can determine whether fixed-point is always better than a double-precision implementation at the same relative error. We demonstrate 35% geometric mean area improvement for different SPICE device models such as Diode, Level-1 MOSFET and an Approximate MOSFET when comparing custom fixed-point implementations with standard double-precision realizations.
- 1 April 2012
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 77-84
FX-SCORE: A Framework for Fixed-Point Compilation of SPICE Device Models Using Gappa++
Automated, offline precision-analysis of dataflow computation containing elementary functions (e.g. exp) and if-then-else control flow operations enables accurate fixed-point FPGA implementation of SPICE device equations. We perform interval analysis of these equations using Gappa++ to statically compare error bounds of fixed-point and double-precision implementations. This is possible due to the limited dynamic range of physical voltage, current and conductance quantities in a SPICE simulation of real-world circuits. In contrast to previous custom-precision SPICE device mappings, our fixed-point implementation has the same accuracy as double-precision implementation when compared to ideal arithmetic (reals). To deliver these implementations we develop FX-SCORE, a high-level framework based on the SCORE streaming FPGA framework, that automatically generates Gappa++ scripts and AutoESL circuits to explore the cost-quality tradeoffs of Fixed-point FPGA implementations. Using our methodology, we can determine whether fixed-point is always better than a double-precision implementation at the same relative error. We demonstrate 35% geometric mean area improvement for different SPICE device models such as Diode, Level-1 MOSFET and an Approximate MOSFET when comparing custom fixed-point implementations with standard double-precision realizations.
- 1 January 2012
- Published by Springer Nature
- p. 239-250
A High Throughput FPGA-Based Implementation of the Lanczos Method for the Symmetric Extremal Eigenvalue Problem
- 1 January 2012
- Published by Springer Nature
- p. 239-250
A High Throughput FPGA-Based Implementation of the Lanczos Method for the Symmetric Extremal Eigenvalue Problem
- 1 December 2011
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-9
VLIW-SCORE: Beyond C for sequential control of SPICE FPGA acceleration
Many stand-alone, FPGA-based accelerators separate the implementation of a computation into two components - (1) a large parallel component that is realized as hardware on spatial FPGA fabric and (2) a small control and co-ordination component that is realized as software on embedded soft-core processors like an off-the-shelf Xilinx Microblaze (or host offchip CPU). While this hardware-software partitioning methodology allows the designer to lower design effort when composing the accelerator system, it introduces unnecessary Amdahl's Law bottlenecks and limits scalability. In this paper, we show how to avoid these limitations with VLIW-SCORE: a combination of a high-level parallel programming framework called SCORE and a custom, hybrid VLIW hardware organization. We demonstrate the benefits of this methodology for the SPICE circuit simulator when implementing the simulation control algorithms. With our spatial mapping flow we are able to improve performance by ≈30% (mean across circuit benchmarks) when compared to the Microblaze implementation for the Xilinx Virtex-6 LX760 FPGA. For complete application acceleration, we see an improved speedup from 1.9× for the Microblaze-based design to 2.6× for the hybrid, custom VLIW implementation when comparing a Xilinx Virtex-6 LX760 FPGA (40nm) with an Intel Core i7 965 CPU (45nm).
- 1 December 2011
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-9
VLIW-SCORE: Beyond C for sequential control of SPICE FPGA acceleration
Many stand-alone, FPGA-based accelerators separate the implementation of a computation into two components - (1) a large parallel component that is realized as hardware on spatial FPGA fabric and (2) a small control and co-ordination component that is realized as software on embedded soft-core processors like an off-the-shelf Xilinx Microblaze (or host offchip CPU). While this hardware-software partitioning methodology allows the designer to lower design effort when composing the accelerator system, it introduces unnecessary Amdahl's Law bottlenecks and limits scalability. In this paper, we show how to avoid these limitations with VLIW-SCORE: a combination of a high-level parallel programming framework called SCORE and a custom, hybrid VLIW hardware organization. We demonstrate the benefits of this methodology for the SPICE circuit simulator when implementing the simulation control algorithms. With our spatial mapping flow we are able to improve performance by ≈30% (mean across circuit benchmarks) when compared to the Microblaze implementation for the Xilinx Virtex-6 LX760 FPGA. For complete application acceleration, we see an improved speedup from 1.9× for the Microblaze-based design to 2.6× for the hybrid, custom VLIW implementation when comparing a Xilinx Virtex-6 LX760 FPGA (40nm) with an Intel Core i7 965 CPU (45nm).
- Open Access
- 29 September 2011
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Autonomous and Adaptive Systems
- Vol. 6 (3) , 1-20
Spatial hardware implementation for sparse graph algorithms in GraphStep
How do we develop programs that are easy to express, easy to reason about, and able to achieve high performance on massively parallel machines? To address this problem, we introduce GraphStep, a domain-specific compute model that captures algorithms that act on static, irregular, sparse graphs. In GraphStep, algorithms are expressed directly without requiring the programmer to explicitly manage parallel synchronization, operation ordering, placement, or scheduling details. Problems in the sparse graph domain are usually highly concurrent and communicate along graph edges. Exposing concurrency and communication structure allows scheduling of parallel operations and management of communication that is necessary for performance on a spatial computer. We study the performance of a semantic network application, a shortest-path application, and a max-flow/min-cut application. We introduce a language syntax for GraphStep applications. The total speedup over sequential versions of the applications studied ranges from a factor of 19 to a factor of 15,000. Spatially-aware graph optimizations (e.g., node decomposition, placement and route scheduling) delivered speedups from 3 to 30 times over a spatially-oblivious mapping.
- 29 September 2011
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Autonomous and Adaptive Systems
- Vol. 6 (3) , 1-20
Open AccessSpatial hardware implementation for sparse graph algorithms in GraphStep
How do we develop programs that are easy to express, easy to reason about, and able to achieve high performance on massively parallel machines? To address this problem, we introduce GraphStep, a domain-specific compute model that captures algorithms that act on static, irregular, sparse graphs. In GraphStep, algorithms are expressed directly without requiring the programmer to explicitly manage parallel synchronization, operation ordering, placement, or scheduling details. Problems in the sparse graph domain are usually highly concurrent and communicate along graph edges. Exposing concurrency and communication structure allows scheduling of parallel operations and management of communication that is necessary for performance on a spatial computer. We study the performance of a semantic network application, a shortest-path application, and a max-flow/min-cut application. We introduce a language syntax for GraphStep applications. The total speedup over sequential versions of the applications studied ranges from a factor of 19 to a factor of 15,000. Spatially-aware graph optimizations (e.g., node decomposition, placement and route scheduling) delivered speedups from 3 to 30 times over a spatially-oblivious mapping.
- Open Access
- 7 March 2011
- Published by Hindawi Limited in International Journal of Reconfigurable Computing
- Vol. 2011, 1-14
An NoC Traffic Compiler for Efficient FPGA Implementation of Sparse Graph-Oriented Workloads
Parallel graph-oriented applications expressed in the Bulk-Synchronous Parallel (BSP) and Token Dataflow compute models generate highly-structured communication workloads from messages propagating along graph edges. We can statially expose this structure to traffic compilers and optimization tools to reshape and reduce traffic for higher performance (or lower area, lower energy, lower cost). Such offline traffic optimization eliminates the need for complex, runtime NoC hardware and enables lightweight, scalable NoCs. We perform load balancing, placement, fanout routing, and fine-grained synchronization to optimize our workloads for large networks up to 2025 parallel elements for BSP model and 25 parallel elements for Token Dataflow. This allows us to demonstrate speedups between 1.2× and 22× (3.5× mean), area reductions (number of Processing Elements) between 3× and 15× (9× mean) and dynamic energy savings between 2× and 3.5× (2.7× mean) over a range of real-world graph applications in the BSP compute model. We deliver speedups of 0.5–13× (geomean 3.6×) for Sparse Direct Matrix Solve (Token Dataflow compute model) applied to a range of sparse matrices when using a high-quality placement algorithm. We expect such traffic optimization tools and techniques to become an essential part of the NoC application-mapping flow.
- 7 March 2011
- Published by Hindawi Limited in International Journal of Reconfigurable Computing
- Vol. 2011, 1-14
Open AccessAn NoC Traffic Compiler for Efficient FPGA Implementation of Sparse Graph-Oriented Workloads
Parallel graph-oriented applications expressed in the Bulk-Synchronous Parallel (BSP) and Token Dataflow compute models generate highly-structured communication workloads from messages propagating along graph edges. We can statially expose this structure to traffic compilers and optimization tools to reshape and reduce traffic for higher performance (or lower area, lower energy, lower cost). Such offline traffic optimization eliminates the need for complex, runtime NoC hardware and enables lightweight, scalable NoCs. We perform load balancing, placement, fanout routing, and fine-grained synchronization to optimize our workloads for large networks up to 2025 parallel elements for BSP model and 25 parallel elements for Token Dataflow. This allows us to demonstrate speedups between 1.2× and 22× (3.5× mean), area reductions (number of Processing Elements) between 3× and 15× (9× mean) and dynamic energy savings between 2× and 3.5× (2.7× mean) over a range of real-world graph applications in the BSP compute model. We deliver speedups of 0.5–13× (geomean 3.6×) for Sparse Direct Matrix Solve (Token Dataflow compute model) applied to a range of sparse matrices when using a high-quality placement algorithm. We expect such traffic optimization tools and techniques to become an essential part of the NoC application-mapping flow.