Alberto Scolari
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Alberto Scolari has a total of 29 co-authors. Here are the top co-authors by number of shared publications:
- Marco Domenico Santambrogio
Dipartimento di Elettronica e Informazione, Politecnico di Milano, Milan, Italy
6 shared publications - D. Sciuto
Politecnico di Milano, Milano, Italy
3 shared publications - Davide Basilio Bartolini
Politecnico di Milano
2 shared publications Filippo Sironi
2 shared publicationsA. Brokalakis
1 shared publication- Albert-Jan N. Yzelman
Computing Systems Lab, Huawei Zurich Research Center, Zurich, Switzerland
1 shared publication
Latest Publications
- Open Access
Effective implementation of the High Performance Conjugate Gradient benchmark on GraphBLAS
Applications in High-Performance Computing (HPC) environments face challenges due to increasing complexity. Among them, the increasing usage of sparse data pushes the limits of data structures and programming models and hampers the efficient usage of existing, highly parallel hardware. The GraphBLAS specification tackles these challenges by proposing a set of data containers and primitives, coupled with a semantics based on abstract algebraic concepts: this allows multiple applications on sparse data to be described with a small set of primitives and benefit from the many optimizations of a compile-time-known algebraic specification. Among HPC applications, the High Performance Conjugate Gradient (HPCG) benchmark is an important representative of a large body of sparse workloads, and its structure poses several programmability and performance challenges. This work tackles them by proposing and evaluating an implementation on GraphBLAS of HPCG, highlighting the main changes to its kernels. The results for shared memory systems outperforms the reference, while results in distributed systems highlight fundamental limitations of GraphBLAS-compliant implementations, which suggests several future directions.
Open AccessEffective implementation of the High Performance Conjugate Gradient benchmark on GraphBLAS
Applications in High-Performance Computing (HPC) environments face challenges due to increasing complexity. Among them, the increasing usage of sparse data pushes the limits of data structures and programming models and hampers the efficient usage of existing, highly parallel hardware. The GraphBLAS specification tackles these challenges by proposing a set of data containers and primitives, coupled with a semantics based on abstract algebraic concepts: this allows multiple applications on sparse data to be described with a small set of primitives and benefit from the many optimizations of a compile-time-known algebraic specification. Among HPC applications, the High Performance Conjugate Gradient (HPCG) benchmark is an important representative of a large body of sparse workloads, and its structure poses several programmability and performance challenges. This work tackles them by proposing and evaluating an implementation on GraphBLAS of HPCG, highlighting the main changes to its kernels. The results for shared memory systems outperforms the reference, while results in distributed systems highlight fundamental limitations of GraphBLAS-compliant implementations, which suggests several future directions.
- Open Access
- 1 May 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 107-114
Pareto Optimal Design Space Exploration for Accelerated CNN on FPGA
Convolutional Neural Networks (CNNs) are at the base of many applications, both in embedded and in server-class contexts. While Graphics Processing Units (GPUs) are predominantly used for training, solutions for inference often rely on Field Programmable Gate Arrays (FPGAs) since they are more flexible and cost-efficient in many scenarios. However, existing approaches fall short to accomplish several conflicting goals, like efficiently using resources on multiple platforms while retaining deep configurability and allowing a quick Design Space Exploration (DSE) towards the best solution. This paper proposes a solution composed of highly configurable kernels designed for resources time-sharing with an analytical model of their resource/performance characteristics. Building on such models, we propose an Integer Linear Programming (ILP)-based approach to effectively identify pareto optimal kernel configurations in terms of throughput and resource consumption. We evaluate our DSE on two state-of-the-art CNNs, showing how it identifies hundreds of pareto optimal solutions in less than a minute. Guided from the DSE configurations of the AlexNet network, we quickly identified a candidate design for a Xilinx Virtex-7 XC7VX485T FPGA and achieved a peak throughput of 4.05 ms per image, while we measured a maximum estimation error of 6.69% with respect to the proposed analytical models.
- 1 May 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 107-114
Open AccessPareto Optimal Design Space Exploration for Accelerated CNN on FPGA
Convolutional Neural Networks (CNNs) are at the base of many applications, both in embedded and in server-class contexts. While Graphics Processing Units (GPUs) are predominantly used for training, solutions for inference often rely on Field Programmable Gate Arrays (FPGAs) since they are more flexible and cost-efficient in many scenarios. However, existing approaches fall short to accomplish several conflicting goals, like efficiently using resources on multiple platforms while retaining deep configurability and allowing a quick Design Space Exploration (DSE) towards the best solution. This paper proposes a solution composed of highly configurable kernels designed for resources time-sharing with an analytical model of their resource/performance characteristics. Building on such models, we propose an Integer Linear Programming (ILP)-based approach to effectively identify pareto optimal kernel configurations in terms of throughput and resource consumption. We evaluate our DSE on two state-of-the-art CNNs, showing how it identifies hundreds of pareto optimal solutions in less than a minute. Guided from the DSE configurations of the AlexNet network, we quickly identified a candidate design for a Xilinx Virtex-7 XC7VX485T FPGA and achieved a peak throughput of 4.05 ms per image, while we measured a maximum estimation error of 6.69% with respect to the proposed analytical models.
- 1 May 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 131-137
TiReX: Tiled Regular eXpression Matching Architecture
Regular Expressions (RE) are widely used to find patterns among data, like in genomic markers research for DNA analysis, deep packet inspection or signature-based detection for network intrusion detection system. This paper proposes a novel and efficient RE matching architecture for FPGAs, based on the concept of matching core. RE can be software-compiled into sequences of basic matching instructions that a matching core runs on input data, and can be replaced to change the RE to be matched. This architecture can easily scale up with the available resources and is customizable to multiple usage scenarios. We ran several experiments and compared the obtained results with a software solution, reaching speedups over 100×, while running at 130MHz, over a Flex-based matching application running on an Intel i7 CPU at 2.8GHz.
- 1 May 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 131-137
TiReX: Tiled Regular eXpression Matching Architecture
Regular Expressions (RE) are widely used to find patterns among data, like in genomic markers research for DNA analysis, deep packet inspection or signature-based detection for network intrusion detection system. This paper proposes a novel and efficient RE matching architecture for FPGAs, based on the concept of matching core. RE can be software-compiled into sequences of basic matching instructions that a matching core runs on input data, and can be replaced to change the RE to be matched. This architecture can easily scale up with the available resources and is customizable to multiple usage scenarios. We ran several experiments and compared the obtained results with a software solution, reaching speedups over 100×, while running at 130MHz, over a Flex-based matching application running on an Intel i7 CPU at 2.8GHz.
- 1 November 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 10636404,p. 431-434
Towards Accelerating Generic Machine Learning Prediction Pipelines
Machine Learning models are often composed by sequences of transformations. While this design makes easy to decompose and accelerate single model components at training time, predictions requires low latency and high performance predictability whereby end-to-end runtime optimizations and acceleration is needed to meet such goals. This paper shed some light on the problem by using a production-like model, and showing how by redesigning model pipelines for efficient execution over CPUs and FPGAs performance improvements of several folds can be achieved.
- 1 November 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 10636404,p. 431-434
Towards Accelerating Generic Machine Learning Prediction Pipelines
Machine Learning models are often composed by sequences of transformations. While this design makes easy to decompose and accelerate single model components at training time, predictions requires low latency and high performance predictability whereby end-to-end runtime optimizations and acceleration is needed to meet such goals. This paper shed some light on the problem by using a production-like model, and showing how by redesigning model pipelines for efficient execution over CPUs and FPGAs performance improvements of several folds can be achieved.
- 1 March 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 33, 410-415
Heterogeneous exascale supercomputing: The role of CAD in the exaFPGA project
- 1 March 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 33, 410-415
Heterogeneous exascale supercomputing: The role of CAD in the exaFPGA project
- Open Access
- 16 December 2016
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Architecture and Code Optimization
- Vol. 13 (4) , 1-24
A Software Cache Partitioning System for Hash-Based Caches
Contention on the shared Last-Level Cache (LLC) can have a fundamental negative impact on the performance of applications executed on modern multicores. An interesting software approach to address LLC contention issues is based on page coloring , which is a software technique that attempts to achieve performance isolation by partitioning a shared cache through careful memory management. The key assumption of traditional page coloring is that the cache is physically addressed. However, recent multicore architectures (e.g., Intel Sandy Bridge and later) switched from a physical addressing scheme to a more complex scheme that involves a hash function. Traditional page coloring is ineffective on these recent architectures. In this article, we extend page coloring to work on these recent architectures by proposing a mechanism able to handle their hash-based LLC addressing scheme. Just as for traditional page coloring, the goal of this new mechanism is to deliver performance isolation by avoiding contention on the LLC, thus enabling predictable performance. We implement this mechanism in the Linux kernel, and evaluate it using several benchmarks from the SPEC CPU2006 and PARSEC 3.0 suites. Our results show that our solution is able to deliver performance isolation to concurrently running applications by enforcing partitioning of a Sandy Bridge LLC, which traditional page coloring techniques are not able to handle.
- 16 December 2016
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Architecture and Code Optimization
- Vol. 13 (4) , 1-24
Open AccessA Software Cache Partitioning System for Hash-Based Caches
Contention on the shared Last-Level Cache (LLC) can have a fundamental negative impact on the performance of applications executed on modern multicores. An interesting software approach to address LLC contention issues is based on page coloring , which is a software technique that attempts to achieve performance isolation by partitioning a shared cache through careful memory management. The key assumption of traditional page coloring is that the cache is physically addressed. However, recent multicore architectures (e.g., Intel Sandy Bridge and later) switched from a physical addressing scheme to a more complex scheme that involves a hash function. Traditional page coloring is ineffective on these recent architectures. In this article, we extend page coloring to work on these recent architectures by proposing a mechanism able to handle their hash-based LLC addressing scheme. Just as for traditional page coloring, the goal of this new mechanism is to deliver performance isolation by avoiding contention on the LLC, thus enabling predictable performance. We implement this mechanism in the Linux kernel, and evaluate it using several benchmarks from the SPEC CPU2006 and PARSEC 3.0 suites. Our results show that our solution is able to deliver performance isolation to concurrently running applications by enforcing partitioning of a Sandy Bridge LLC, which traditional page coloring techniques are not able to handle.
- 1 August 2014
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 21589178,p. 134-141
FPGA-Based Design Using the FASTER Toolchain: The Case of STM Spear Development Board
Even though FPGAs are becoming more and more popular as they are used in many different scenarios like communications and HPC, the steep learning curve needed to work with this technology is still the major limiting factor to their full success. Many works proposed to mitigate this problem by creating a companion of tools to support the designer during the development phase for this technology. The EU FASTER Project aims at realizing an integrated toolchain that assists the designer in the steps of the design flow that are necessary to port a given application onto an FPGA device. The novelty of the framework relies in the fact that the partial dynamic reconfiguration, which FPGA devices can exploit, is seen as a first class citizen throughout the whole design flow. This work reports a case study in which the FASTER toolchain has been used to port a raytracer application onto the STM Spear prototyping embedded platform. The paper discusses the steps done for the realization of the prototype and the results obtained on the target device. It finally reports some improvements that can be exploited to improve the performance of the hardware implementation that has been realized.
- 1 August 2014
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 21589178,p. 134-141
FPGA-Based Design Using the FASTER Toolchain: The Case of STM Spear Development Board
Even though FPGAs are becoming more and more popular as they are used in many different scenarios like communications and HPC, the steep learning curve needed to work with this technology is still the major limiting factor to their full success. Many works proposed to mitigate this problem by creating a companion of tools to support the designer during the development phase for this technology. The EU FASTER Project aims at realizing an integrated toolchain that assists the designer in the steps of the design flow that are necessary to port a given application onto an FPGA device. The novelty of the framework relies in the fact that the partial dynamic reconfiguration, which FPGA devices can exploit, is seen as a first class citizen throughout the whole design flow. This work reports a case study in which the FASTER toolchain has been used to port a raytracer application onto the STM Spear prototyping embedded platform. The paper discusses the steps done for the realization of the prototype and the results obtained on the target device. It finally reports some improvements that can be exploited to improve the performance of the hardware implementation that has been realized.
- 1 August 2014
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 242-247
A Survey on Recent Hardware and Software-Level Cache Management Techniques
Multi and many-core processors have emerged as the dominant solution for processing in the whole range of computer system, from small devices to large-scale installations. Chip multi-processors, which are homogeneous, multi and manycore processors, offer an unprecedented amount of on-chip, shared resources and brings a unique set of challenges. Given the importance of the Last-Level Cache management techniques to achieve near-perfect isolation, we survey the state of the art and propose research directions to address the most pressing issues in modern computer systems. To better understand the various research directions in the field, we propose a classification of the presented techniques. Finally, we discuss possible research directions.
- 1 August 2014
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 242-247
A Survey on Recent Hardware and Software-Level Cache Management Techniques
Multi and many-core processors have emerged as the dominant solution for processing in the whole range of computer system, from small devices to large-scale installations. Chip multi-processors, which are homogeneous, multi and manycore processors, offer an unprecedented amount of on-chip, shared resources and brings a unique set of challenges. Given the importance of the Last-Level Cache management techniques to achieve near-perfect isolation, we survey the state of the art and propose research directions to address the most pressing issues in modern computer systems. To better understand the various research directions in the field, we propose a classification of the presented techniques. Finally, we discuss possible research directions.
- Open Access
- 1 October 2013
- Published by Association for Computing Machinery (ACM)
- p. 1-2
Coloring the cloud for predictable performance
Motivation and Contribution The commodity multicores that power cloud infrastructures hide memory latency through deep memory hierarchies, with the last-level cache (LLC) usually shared among cores. While a shared LLC improves utilization of on-chip resources, it may also lead to unpredictable performance of colocated virtual machines (VMs) as a result of unanticipated contention. Past research showed that the operating system page allocator can favor performance predictability on a physically-addressed shared LLC through page coloring [4, 8, 9]: a software technique that can work on commodity multicores, unlike hardware approaches [2, 7]. The main drawback of page coloring is the high cost of modifying allocations (i.e., recoloring), making this technique almost impractical for applications with varying memory footprints [6].
- 1 October 2013
- Published by Association for Computing Machinery (ACM)
- p. 1-2
Open AccessColoring the cloud for predictable performance
Motivation and Contribution The commodity multicores that power cloud infrastructures hide memory latency through deep memory hierarchies, with the last-level cache (LLC) usually shared among cores. While a shared LLC improves utilization of on-chip resources, it may also lead to unpredictable performance of colocated virtual machines (VMs) as a result of unanticipated contention. Past research showed that the operating system page allocator can favor performance predictability on a physically-addressed shared LLC through page coloring [4, 8, 9]: a software technique that can work on commodity multicores, unlike hardware approaches [2, 7]. The main drawback of page coloring is the high cost of modifying allocations (i.e., recoloring), making this technique almost impractical for applications with varying memory footprints [6].