Krste Asanovic
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Krste Asanovic has a total of 158 co-authors. Here are the top co-authors by number of shared publications:
- Borivoje Nikolic
University of California, Berkeley, CA, USA
10 shared publications - Jerry Zhao
University of California, Berkeley, Berkeley, CA, USA
5 shared publications - Sagar Karandikar
University of California, Berkeley, Berkeley, California, USA
5 shared publications - Yakun Sophia Shao
University of California, Berkeley, Berkeley, California, USA
5 shared publications - Alon Amid
University of California, Berkeley, Berkeley, CA, USA
4 shared publications - Colin Schmidt
Department of Electrical Engineering and Computer Sciences, University of California at Berkeley, Berkeley, CA, USA
3 shared publications
Latest Publications
- Open Access
- 19 May 2026
- Published by Association for Computing Machinery (ACM)
- p. 133-136
Matrix Extensions for the RISC-V ISA
- 19 May 2026
- Published by Association for Computing Machinery (ACM)
- p. 133-136
Open AccessMatrix Extensions for the RISC-V ISA
- 31 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. PP (00189200) , 1-15
Cygnus: A Heterogeneous Octa-Core RISC-V Vector Processor for Digital Signal Processing Applications
We introduce Cygnus, an energy-efficient octa-core RISC-V vector processor, compliant with the RISC-V vector extension (RVV) 1.0 specification, specifically designed for digital signal processing (DSP) and robotics applications. This system-on-chip (SoC) features dynamic instruction scheduling with short-vector lengths at the vector core level and employs a big/little architecture with tightly coupled memory (TCM) at the SoC level. The TCM design enables efficient data streaming across the cores, bypassing the cache hierarchy for streaming applications. Cygnus achieves a high average utilization of 90% across general matrix multiply (GEMM) and convolution (CONV) kernels, demonstrating energy efficiencies of 414 GOPS/W for INT8 GEMM and 109 GFLOPS/W for FP32 GEMM. In a typical computer vision kernel that implements denoising using a non-local means (NLMs) algorithm, the SoC provides a latency of 3.7 ms with an energy consumption of $518~\mu $ J per frame. Moreover, we demonstrate the adaptability of our approach using our SoC in a drone control application that employs a model-predictive control (MPC) algorithm based on TinyMPC. Cygnus, running the Zephyr real-time operating system (RTOS), manages control for three drones while optimizing performance through vectorization, achieving over $3.5\times $ improvement in solve time and significant reductions in position-tracking error at higher clock frequencies. The implementation demonstrates enhanced the real-time performance and accuracy compared with baseline scalar methods.
- 31 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. PP (00189200) , 1-15
Cygnus: A Heterogeneous Octa-Core RISC-V Vector Processor for Digital Signal Processing Applications
We introduce Cygnus, an energy-efficient octa-core RISC-V vector processor, compliant with the RISC-V vector extension (RVV) 1.0 specification, specifically designed for digital signal processing (DSP) and robotics applications. This system-on-chip (SoC) features dynamic instruction scheduling with short-vector lengths at the vector core level and employs a big/little architecture with tightly coupled memory (TCM) at the SoC level. The TCM design enables efficient data streaming across the cores, bypassing the cache hierarchy for streaming applications. Cygnus achieves a high average utilization of 90% across general matrix multiply (GEMM) and convolution (CONV) kernels, demonstrating energy efficiencies of 414 GOPS/W for INT8 GEMM and 109 GFLOPS/W for FP32 GEMM. In a typical computer vision kernel that implements denoising using a non-local means (NLMs) algorithm, the SoC provides a latency of 3.7 ms with an energy consumption of $518~\mu $ J per frame. Moreover, we demonstrate the adaptability of our approach using our SoC in a drone control application that employs a model-predictive control (MPC) algorithm based on TinyMPC. Cygnus, running the Zephyr real-time operating system (RTOS), manages control for three drones while optimizing performance through vectorization, achieving over $3.5\times $ improvement in solve time and significant reductions in position-tracking error at higher clock frequencies. The implementation demonstrates enhanced the real-time performance and accuracy compared with baseline scalar methods.
- 13 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Micro
- Vol. 44 (4) , 97-105
AuRORA: A Full-Stack Solution for Scalable and Virtualized Accelerator Integration
To meet the increasingly demanding compute requirements of modern workloads, SoCs must provide an accelerator-rich hardware architecture and software programming interface. However, scalability remains a first-order concern, as introducing additional unmanageable complexity to either physical design or software integration may prohibit the deployment of new accelerators. To address these challenges, this work presents AuRORA, an accelerator integration methodology that provides a scalable physical accelerator interface, while preserving software semantics with minimal overhead for accelerator access. AuRORA provides a new accelerator integration methodology that preserves the software and hardware interface of a tightly-CPU-coupled accelerator, while physically disaggregating the accelerators away from a host CPU. To address software scalability, AuRORA also includes a lightweight software runtime for an SoC with heterogeneous accelerators, providing low-overhead access to these accelerators for multi-tenant applications.
- 13 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Micro
- Vol. 44 (4) , 97-105
AuRORA: A Full-Stack Solution for Scalable and Virtualized Accelerator Integration
To meet the increasingly demanding compute requirements of modern workloads, SoCs must provide an accelerator-rich hardware architecture and software programming interface. However, scalability remains a first-order concern, as introducing additional unmanageable complexity to either physical design or software integration may prohibit the deployment of new accelerators. To address these challenges, this work presents AuRORA, an accelerator integration methodology that provides a scalable physical accelerator interface, while preserving software semantics with minimal overhead for accelerator access. AuRORA provides a new accelerator integration methodology that preserves the software and hardware interface of a tightly-CPU-coupled accelerator, while physically disaggregating the accelerators away from a host CPU. To address software scalability, AuRORA also includes a lightweight software runtime for an SoC with heterogeneous accelerators, providing low-overhead access to these accelerators for multi-tenant applications.
- Open Access
- 27 April 2024
- Published by Association for Computing Machinery (ACM)
- p. 1048-1062
Zoomie: A Software-like Debugging Tool for FPGAs
FPGA prototyping has long been an indispensable technique in pre-silicon verification as well as enabling early-stage software development. FPGAs themselves have also gained popularity as hardware accelerators deployed in datacenters. However, FPGA development brings a plethora of problems. These issues constitute a high barrier towards mass adoption of agile development surrounding FPGA-based projects. To address these problems, we have built Zoomie for fast incremental compilation, reusing verification infrastructure, and a software-inspired approach towards open-source emulation. We show that Zoomie achieves 18× speedup over the vendor toolchain in incremental compilation time for million-gate designs. At the same time, Zoomie also provides a software-like debugging experience with breakpoints, stepping the design, and forcing values in a running design.
- 27 April 2024
- Published by Association for Computing Machinery (ACM)
- p. 1048-1062
Open AccessZoomie: A Software-like Debugging Tool for FPGAs
FPGA prototyping has long been an indispensable technique in pre-silicon verification as well as enabling early-stage software development. FPGAs themselves have also gained popularity as hardware accelerators deployed in datacenters. However, FPGA development brings a plethora of problems. These issues constitute a high barrier towards mass adoption of agile development surrounding FPGA-based projects. To address these problems, we have built Zoomie for fast incremental compilation, reusing verification infrastructure, and a software-inspired approach towards open-source emulation. We show that Zoomie achieves 18× speedup over the vendor toolchain in incremental compilation time for million-gate designs. At the same time, Zoomie also provides a software-like debugging experience with breakpoints, stepping the design, and forcing values in a running design.
- Open Access
- 28 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 62-76
AuRORA: Virtualized Accelerator Orchestration for Multi-Tenant Workloads
With the widespread adoption of deep neural networks (DNNs) across applications, there is a growing demand for DNN deployment solutions that can seamlessly support multi-tenant execution. This involves simultaneously running multiple DNN workloads on heterogeneous architectures with domain-specific accelerators. However, existing accelerator interfaces directly bind the accelerator’s physical resources to user threads, without an efficient mechanism to adaptively re-partition available resources. This leads to high programming complexities and performance overheads due to sub-optimal resource allocation, making scalable many-accelerator deployment impractical. To address this challenge, we propose AuRORA, a novel accelerator integration methodology that enables scalable accelerator deployment for multi-tenant workloads. In particular, AuRORA supports virtualized accelerator orchestration via co-designing the hardware-software stack of accelerators to allow adaptively binding current workloads onto available accelerators. We demonstrate that AuRORA achieves 2.02 higher overall SLA satisfaction, 1.33 overall system throughput, and 1.34 overall fairness compared to existing accelerator integration solutions with less than 2.7% area overhead.
- 28 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 62-76
Open AccessAuRORA: Virtualized Accelerator Orchestration for Multi-Tenant Workloads
With the widespread adoption of deep neural networks (DNNs) across applications, there is a growing demand for DNN deployment solutions that can seamlessly support multi-tenant execution. This involves simultaneously running multiple DNN workloads on heterogeneous architectures with domain-specific accelerators. However, existing accelerator interfaces directly bind the accelerator’s physical resources to user threads, without an efficient mechanism to adaptively re-partition available resources. This leads to high programming complexities and performance overheads due to sub-optimal resource allocation, making scalable many-accelerator deployment impractical. To address this challenge, we propose AuRORA, a novel accelerator integration methodology that enables scalable accelerator deployment for multi-tenant workloads. In particular, AuRORA supports virtualized accelerator orchestration via co-designing the hardware-software stack of accelerators to allow adaptively binding current workloads onto available accelerators. We demonstrate that AuRORA achieves 2.02 higher overall SLA satisfaction, 1.33 overall system throughput, and 1.34 overall fairness compared to existing accelerator integration solutions with less than 2.7% area overhead.
- Open Access
- 17 June 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-17
CDPU: Co-designing Compression and Decompression Processing Units for Hyperscale Systems
- 17 June 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-17
Open AccessCDPU: Co-designing Compression and Decompression Processing Units for Hyperscale Systems
- Open Access
- 17 June 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-16
Profiling Hyperscale Big Data Processing
Computing demand continues to grow exponentially, largely driven by "big data" processing on hyperscale data stores. At the same time, the slowdown in Moore's law is leading the industry to embrace custom computing in large-scale systems. Taken together, these trends motivate the need to characterize live production traffic on these large data processing platforms and understand the opportunity of acceleration at scale. This paper addresses this key need. We characterize three important production distributed database and data analytics platforms at Google to identify key hardware acceleration opportunities and perform a comprehensive limits study to understand the trade-offs among various hardware acceleration strategies. We observe that hyperscale data processing platforms spend significant time on distributed storage and other remote work across distributed workers. Therefore, optimizing storage and remote work in addition to compute acceleration is critical for these platforms. We present a detailed breakdown of the compute-intensive functions in these platforms and identify dominant key data operations related to datacenter and systems taxes. We observe that no single accelerator can provide a significant benefit but collectively, a sea of accelerators, can accelerate many of these smaller platform-specific functions. We demonstrate the potential gains of the sea of accelerators proposal in a limits study and analytical model. We perform a comprehensive study to understand the trade-offs between accelerator location (on-chip/off-chip) and invocation model (synchronous/asynchronous). We propose and evaluate a chained accelerator execution model where identified compute-intensive functions are accelerated and pipelined to avoid invocation from the core, achieving a 3x improvement over the baseline system while nearly matching identical performance to an ideal fully asynchronous execution model.
- 17 June 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-16
Open AccessProfiling Hyperscale Big Data Processing
Computing demand continues to grow exponentially, largely driven by "big data" processing on hyperscale data stores. At the same time, the slowdown in Moore's law is leading the industry to embrace custom computing in large-scale systems. Taken together, these trends motivate the need to characterize live production traffic on these large data processing platforms and understand the opportunity of acceleration at scale. This paper addresses this key need. We characterize three important production distributed database and data analytics platforms at Google to identify key hardware acceleration opportunities and perform a comprehensive limits study to understand the trade-offs among various hardware acceleration strategies. We observe that hyperscale data processing platforms spend significant time on distributed storage and other remote work across distributed workers. Therefore, optimizing storage and remote work in addition to compute acceleration is critical for these platforms. We present a detailed breakdown of the compute-intensive functions in these platforms and identify dominant key data operations related to datacenter and systems taxes. We observe that no single accelerator can provide a significant benefit but collectively, a sea of accelerators, can accelerate many of these smaller platform-specific functions. We demonstrate the potential gains of the sea of accelerators proposal in a limits study and analytical model. We perform a comprehensive study to understand the trade-offs between accelerator location (on-chip/off-chip) and invocation model (synchronous/asynchronous). We propose and evaluate a chained accelerator execution model where identified compute-intensive functions are accelerated and pipelined to avoid invocation from the core, achieving a 3x improvement over the baseline system while nearly matching identical performance to an ideal fully asynchronous execution model.
- 11 May 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Micro
- Vol. 43 (3) , 7-8
Special Issue on Hot Chips 34
The Hot Chips conference has served as a leading venue for presenting architectural details of new chips and chip-related technologies, from established industry leaders, startups, and academia. This article presents an introduction to the seven articles that are chosen to be in the IEEE Micro HotChips Special Issue of 2023.
- 11 May 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Micro
- Vol. 43 (3) , 7-8
Special Issue on Hot Chips 34
The Hot Chips conference has served as a leading venue for presenting architectural details of new chips and chip-related technologies, from established industry leaders, startups, and academia. This article presents an introduction to the seven articles that are chosen to be in the IEEE Micro HotChips Special Issue of 2023.
- 9 November 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 57 (1) , 140-152
An Eight-Core 1.44-GHz RISC-V Vector Processor in 16-nm FinFET
This work presents a RISC-V system-on-chip (SoC) with eight application cores containing programmable-precision vector accelerators. The SoC is built by using a generator-based design methodology, which enables the integration of open-source and project-specific building blocks to develop differentiated functionality. The digital component generators use Chisel, the analog component generators use the Berkeley Analog Generator (BAG), and the physical design flow is implemented with Hammer. The chip totals 125 M gates and is implemented in a 16-nm finFET process. The vector accelerator achieves peak energy efficiency per task of 209 half-precision, 92 single-precision, and 56 double-precision GFLOPs/W for a matrix multiplication kernel at 0.55 V and 339 MHz.
- 9 November 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 57 (1) , 140-152
An Eight-Core 1.44-GHz RISC-V Vector Processor in 16-nm FinFET
This work presents a RISC-V system-on-chip (SoC) with eight application cores containing programmable-precision vector accelerators. The SoC is built by using a generator-based design methodology, which enables the integration of open-source and project-specific building blocks to develop differentiated functionality. The digital component generators use Chisel, the analog component generators use the Berkeley Analog Generator (BAG), and the physical design flow is implemented with Hammer. The chip totals 125 M gates and is implemented in a 16-nm finFET process. The vector accelerator achieves peak energy efficiency per task of 209 half-precision, 92 single-precision, and 56 double-precision GFLOPs/W for a matrix multiplication kernel at 0.55 V and 339 MHz.
- 1 June 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Micro
- Vol. 41 (4) , 58-66
Accessible, FPGA Resource-Optimized Simulation of Multiclock Systems in FireSim
Given the complexity of modern systems-on-chip, hardware-assisted verification is an integral part of the chip-design process. However, chip designers often need to choose between richly featured but expensive emulation platforms or faster, cheaper, but less debuggable FPGA prototyping solutions. FireSim, an open-source, FPGA-accelerated hardware emulation platform hosted in the public cloud, attempts to accessibly offer the best of both worlds. This article highlights two new FireSim capabilities that help realize this goal: multicycle resource optimizations, which can enable an eight-fold increase emulated core count, and FPGA-agnostic support for multiclock systems. These supplement existing FireSim features which provide a foundation for productive emulation, including a cloud manager to automatically scale out experiments and a rich debug toolkit.
- 1 June 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Micro
- Vol. 41 (4) , 58-66
Accessible, FPGA Resource-Optimized Simulation of Multiclock Systems in FireSim
Given the complexity of modern systems-on-chip, hardware-assisted verification is an integral part of the chip-design process. However, chip designers often need to choose between richly featured but expensive emulation platforms or faster, cheaper, but less debuggable FPGA prototyping solutions. FireSim, an open-source, FPGA-accelerated hardware emulation platform hosted in the public cloud, attempts to accessibly offer the best of both worlds. This article highlights two new FireSim capabilities that help realize this goal: multicycle resource optimizations, which can enable an eight-fold increase emulated core count, and FPGA-agnostic support for multiclock systems. These supplement existing FireSim features which provide a foundation for productive emulation, including a cloud manager to automatically scale out experiments and a rich debug toolkit.