TFTim Fischer

Tim Fischer

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Tim Fischer has a total of 75 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    EPAC: The Last Dance

    by Filippo Mantovani, Fabio Banchelli, Pablo Vizcaino, Roger Ferrer, Oscar Palomar, Francesco Minervini, Jesus Labarta, Mauro Olivieri, Sebastiano Pomata, Pedro Marcuello, et al.


  • Open Access

    Toward Open-Source Chiplets for HPC and AI: Occamy and Beyond

    by Paul Scheffler, Thomas Benz, Tim Fischer, Lorenzo Leone, Sina Arjmandpour, Luca Benini

    We present a roadmap for open-source chiplet-based RISC-V systems targeting high-performance computing and artificial intelligence, aiming to close the performance gap to proprietary designs. Starting with Occamy, the first open, silicon-proven dual-chiplet RISC-V manycore in 12nm FinFET, we scale to Ramora, a mesh-NoC-based dual-chiplet system, and to Ogopogo, a 7nm quad-chiplet concept architecture achieving state-of-the-art compute density. Finally, we explore possible avenues to extend openness beyond logic-core RTL into simulation, EDA, PDKs, and off-die PHYs.

  • ControlPULPlet: A Flexible Real-time Multicore RISC-V Controller for 2.5-D Systems-in-Package

    by Alessandro Ottaviano, Robert Balas, Tim Fischer, Thomas Benz, Andrea Bartolini, Luca Benini

    The growing complexity of real-time (RT) control algorithms with increasing performance demands along with the shift to 2.5-D technology drive the need for scalable controllers to manage chiplets’ coupled operation in 2.5-D systems-in-package (SiPs). These controllers must offer RT computing capabilities, as well as SiP-compatible IO interfaces for communicating with the controlled dies. Due to RT constraints, a key challenge is minimizing the performance penalty of die-to-die (D2D) communication with respect to native on-chip control interfaces. We address this challenge with ControlPULPlet, an open-source, RT multicore RISC-V controller designed specifically for SiP integration. ControlPULPlet features a 32-bit CV32RT core for fast interrupt handling and a specialized direct memory access engine to automate periodic sensor readout. A tightly coupled programmable multicore cluster for acceleration of advanced control algorithms is integrated through a dedicated AXI4 port. A flexible AXI4-compatible D2D link enables efficient communication in 2.5-D SiPs. We implemented and fabricated ControlPULPlet as a silicon demonstrator called Kairos in TSMC’s 65-nm CMOS. Kairos runs model predictive control algorithms at up to 290 MHz in a 30 mW power envelope. The D2D link attains a peak duplex transfer rate of 51 Gbit/s at 200 MHz, at the minimal costs of just $\mathrm {7.6~kGE }$ in PHY area per channel, adding just 2.9% to the total system area.


  • Open Access

    TeraNoC: A Multi-Channel 32-bit Fine-Grained, Hybrid Mesh-Crossbar NoC for Efficient Scale-up of 1000+ Core Shared-L1-Memory Clusters

    by Yichao Zhang, Zexin Fu, Tim Fischer, Yinrong Li, Marco Bertuletti, Luca Benini

    A key challenge in on-chip interconnect design is to scale up bandwidth while maintaining low latency and high area efficiency. 2D-meshes scale with low wiring area and congestion overhead; however, their end-to-end latency increases with the number of hops, making them unsuitable for latency-sensitive core-to-L1-memory access. On the other hand, crossbars offer low latency, but their routing complexity grows quadratically with the number of I/Os, requiring large physical routing resources and limiting area-efficient scalability. This two-sided interconnect bottleneck hinders the scale-up of many-core, low-latency, tightly coupled shared-memory clusters, pushing designers toward instantiating many smaller and loosely coupled clusters, at the cost of hardware and software overheads. We present TeraNoC, an open-source, hybrid mesh-crossbar on-chip interconnect that offers both scalability and low latency, while maintaining very low routing overhead. The topology, built on 32bit word-width multi-channel 2D-meshes and crossbars, enables the area-efficient scale-up of shared-memory clusters. A router remapper is designed to balance traffic load across interconnect channels. Using TeraNoC, we build a cluster with 1024 single-stage, single-issue cores that share a 4096-banked L1 memory, implemented in 12nm technology. The low interconnect stalls enable high compute utilization of up to 0.85 IPC in compute-intensive, data-parallel key GenAI kernels. TeraNoC only consumes 7.6\% of the total cluster power in kernels dominated by crossbar accesses, and 22.7\% in kernels with high 2D-mesh traffic. Compared to a hierarchical crossbar-only cluster, TeraNoC reduces die area by 37.8\% and improves area efficiency (GFLOP/s/mm2) by up to 98.7\%, while occupying only 10.9\% of the logic area.

  • Top Cited Papers

    Occamy: A 432-Core Dual-Chiplet Dual-HBM2E 768-DP-GFLOP/s RISC-V System for 8-to-64-bit Dense and Sparse Computing in 12-nm FinFET

    by Paul Scheffler, Thomas Benz, Viviane Potocnik, Tim Fischer, Luca Colagrande, Nils Wistoff, Yichao Zhang, Luca Bertaccini, Gianmarco Ottavi, Manuel Eggimann, et al.

    Machine learning (ML) and high-performance computing (HPC) applications increasingly combine dense and sparse memory access computations to maximize storage efficiency. However, existing central processing units (CPUs) and graphics processing units (GPUs) struggle to flexibly handle these heterogeneous workloads with consistently high compute efficiency. We present Occamy, a 432-core, 768-DP-GFLOP/s, dual-HBM2E, dual-chiplet RISC-V system with a latency-tolerant hierarchical interconnect and in-core streaming units (SUs) designed to accelerate dense and sparse FP8-to-FP64 ML and HPC workloads. We implement Occamy’s compute chiplets in 12-nm FinFET and its passive interposer, Hedwig, in a 65-nm node. On dense linear algebra (LA), Occamy achieves a competitive floating-point unit (FPU) utilization of 89%. On stencil codes, Occamy reaches an FPU utilization of 83% and a technology-node-normalized compute density of 11.1 DP-GFLOP/s/mm2, leading state-of-the-art (SoA) processors by 1.7and 1.2, respectively. On sparse-dense LA, it achieves 42% FPU utilization and a normalized compute density of 5.95 DP-GFLOP/s/mm2, surpassing the SoA by 5.2and 11, respectively. On sparse–sparse LA, Occamy reaches a throughput of up to 187 GCOMP/s at 17.4 GCOMP/s/W and a compute density of 3.63 GCOMP/s/mm2. Finally, we reach up to 75% and 54% FPU utilization on and dense (large language model) and graph-sparse (graph convolutional network) ML inference workloads. Occamy’s register transfer level (RTL) description is freely available under a permissive open-source license.


  • Top Cited Papers

    FlooNoC: A 645-Gb/s/link 0.15-pJ/B/hop Open-Source NoC With Wide Physical Links and End-to-End AXI4 Parallel Multistream Support

    by Tim Fischer, Michael Rogenmoser, Thomas Benz, Frank K. Gürkaynak, Luca Benini

    The new generation of domain-specific AI accelerators is characterized by rapidly increasing demands for bulk data transfers, as opposed to small, latency-critical cache line transfers typical of traditional cache-coherent systems. In this article, we address this critical need by introducing the FlooNoC network-on-chip (NoC), featuring very wide, fully advanced extensible interface (AXI4) compliant links designed to meet the massive bandwidth needs at high energy efficiency. At the transport level, nonblocking transactions are supported for latency tolerance. In addition, a novel end-to-end ordering approach for AXI4, enabled by a multistream capable direct memory access (DMA) engine, simplifies network interfaces (NIs) and eliminates interstream dependencies. Furthermore, dedicated physical links are instantiated for short, latency-critical messages. A complete end-to-end reference implementation in 12-nm FinFET technology demonstrates the physical feasibility and power performance area (PPA) benefits of our approach. Using wide links on high levels of metal, we achieve a bandwidth of 645 Gb/s/link and a total aggregate bandwidth of 103 Tb/s for an $8\times 4$ mesh of processors’ cluster tiles, with a total of 288 RISC-V cores. The NoC imposes a minimal area overhead of only 3.5% per compute tile and achieves a leading-edge energy efficiency of 0.15 pJ/B/hop at 0.8 V. Compared with state-of-the-art (SoA) NoCs, our system offers three times the energy efficiency and more than double the link bandwidth. Furthermore, compared with a traditional AXI4-based multilayer interconnect, our NoC achieves a 30% reduction in area, corresponding to a 47% increase in GFLOPSDP within the same floorplan.


  • Open Access

    AraXL: A Physically Scalable, Ultra-Wide RISC-V Vector Processor Design for Fast and Efficient Computation on Long Vectors

    by Navaneeth Kunhi Purayil, Matteo Perotti, Tim Fischer, Luca Benini

    The ever-growing scale of data parallelism in today's HPC and ML applications presents a big challenge for computing architectures' energy efficiency and performance. Vector processors address the scale-up challenge by decoupling Vector Register File (VRF) and datapath widths, allowing the VRF to host long vectors and increase register-stored data reuse while reducing the relative cost of instruction fetch and decode. However, even the largest vector processor designs today struggle to scale to more than 8 vector lanes with double-precision Floating Point Units (FPUs) and 256 64-bit elements per vector register. This limitation is induced by difficulties in the physical implementation, which becomes wire-dominated and inefficient. In this work, we present AraXL, a modular and scalable 64-bit RISC-V V vector architecture targeting long-vector applications for HPC and ML. AraXL addresses the physical scalability challenges of state-of-the-art vector processors with a distributed and hierarchical interconnect, supporting up to 64 parallel vector lanes and reaching the maximum Vector Register File size of 64 Kibit/vreg permitted by the RISC-V V 1.0 ISA specification. Implemented in a 22-nm technology node, our 64-lane AraXL achieves a performance peak of 146 GFLOPs on computation-intensive HPC/ML kernels (>99% FPU utilization) and energy efficiency of 40.1 GFLOPs/W (1.15 GHz, TT, 0.8V), with only 3.8x the area of a 16-lane instance.

  • Open Access

    Occamy: A 432-Core Dual-Chiplet Dual-HBM2E 768-DP-GFLOP/s RISC-V System for 8-to-64-bit Dense and Sparse Computing in 12nm FinFET

    by Paul Scheffler, Thomas Benz, Viviane Potocnik, Tim Fischer, Luca Colagrande, Nils Wistoff, Yichao Zhang, Luca Bertaccini, Gianmarco Ottavi, Manuel Eggimann, et al.

    ML and HPC applications increasingly combine dense and sparse memory access computations to maximize storage efficiency. However, existing CPUs and GPUs struggle to flexibly handle these heterogeneous workloads with consistently high compute efficiency. We present Occamy, a 432-Core, 768-DP-GFLOP/s, dual-HBM2E, dual-chiplet RISC-V system with a latency-tolerant hierarchical interconnect and in-core streaming units (SUs) designed to accelerate dense and sparse FP8-to-FP64 ML and HPC workloads. We implement Occamy's compute chiplets in 12 nm FinFET, and its passive interposer, Hedwig, in a 65 nm node. On dense linear algebra (LA), Occamy achieves a competitive FPU utilization of 89%. On stencil codes, Occamy reaches an FPU utilization of 83% and a technology-node-normalized compute density of 11.1 DP-GFLOP/s/mm2,leading state-of-the-art (SoA) processors by 1.7x and 1.2x, respectively. On sparse-dense linear algebra (LA), it achieves 42% FPU utilization and a normalized compute density of 5.95 DP-GFLOP/s/mm2, surpassing the SoA by 5.2x and 11x, respectively. On, sparse-sparse LA, Occamy reaches a throughput of up to 187 GCOMP/s at 17.4 GCOMP/s/W and a compute density of 3.63 GCOMP/s/mm2. Finally, we reach up to 75% and 54% FPU utilization on and dense (LLM) and graph-sparse (GCN) ML inference workloads. Occamy's RTL is freely available under a permissive open-source license.

  • Optimizing Foundation Model Inference on a Many-Tiny-Core Open-Source RISC-V Platform

    by Viviane Potocnik, Luca Colagrande, Tim Fischer, Luca Bertaccini, Daniele Jahier Pagliari, Alessio Burrello, Luca Benini

    Transformer-based foundation models have become crucial for various domains, most notably natural language processing (NLP) or computer vision (CV). These models are predominantly deployed on high-performance GPUs or hardwired accelerators with highly customized, proprietary instruction sets. Until now, limited attention has been given to RISC-V-based general-purpose platforms. In our work, we present the first inference results of transformer models on an open-source many-tiny- core RISC-V platform implementing distributed Softmax primitives and leveraging ISA extensions for SIMD floating-point operand streaming and instruction repetition, as well as specialized DMA engines to minimize costly main memory accesses and to tolerate their latency. We focus on two foundational transformer topologies, encoder-only and decoder-only models. For encoder-only models, we demonstrate a speedup of up to 12.8× between the most optimized implementation and the baseline version. We reach over 79% FPU utilization and 294 GFLOPS/W, outperforming State-of-the-Art (SoA) accelerators by more than 2× utilizing the HW platform while achieving comparable throughput per computational unit. For decoder-only topologies, we achieve 16.1× speedup in the Non-Autoregressive (NAR) mode and up to 35.6× speedup in the Autoregressive (AR) mode compared to the baseline implementation. Compared to the best SoA dedicated accelerator, we achieve 2.04× higher FPU utilization.


  • MiniFloats on RISC-V Cores: ISA Extensions With Mixed-Precision Short Dot Products

    by Luca Bertaccini, Gianna Paulin, Matheus Cavalcante, Tim Fischer, Stefan Mach, Luca Benini

    Low-precision floating-point (FP) formats have recently been intensely investigated in the context of machine learning inference and training applications. While 16-bit formats are already widely used, 8-bit FP data types have lately emerged as a viable option for neural network training when employed in a mixed-precision scenario and combined with rounding methods increasing the precision in compound additions, such as stochastic rounding. So far, hardware implementations supporting FP8 are mostly implemented within domain-specific accelerators. We propose two RISC-V instruction set architecture (ISA) extensions, enhancing respectively scalar and vector general-purpose cores with low and mixed-precision capabilities. The extensions support two 8-bit and two 16-bit FP formats and are based on dot-product instructions accumulating at higher precision. We develop a hardware unit supporting mixed-precision dot products and stochastic rounding and integrate it into an open-source floating-point unit (FPU). Finally, we integrate the enhanced FPU into a cluster of scalar cores, as well as a cluster of vector cores, and implement them in a 12 nm FinFET technology. The former achieves 575 GFLOPS/W on FP8-to-FP16 matrix multiplications at 0.8 V, 1.26 GHz; the latter reaches 860 GFLOPS/W at 0.8 V, 1.08 GHz, 1.93x higher efficiency than computing on FP16-to-FP32.