A Study of Reconfigurable Accelerators for Cloud Computing

Mohammedali, Noor Abdalkarem; Agyeman, Michael Opoku

doi:10.1145/3284557.3284563

Cited by 8 publications

(3 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Catapult is difficult to be adapted to different applications. e target application accelerates the Bing web search engine with achieving an improvement of 95% in [10] ✓ ✓ Lee et al [11] ✓ ✓ Mohammedali et al [12] ✓ ✓ Vaishnav [13] ✓ Le et al [14] ✓ Vipin et al [15] ✓ Mondol et al [16] ✓ Will et al [17] ✓ is survey ✓ ✓ ✓ throughput improvement while consuming only 10% more power per CPU-FPGA server. ese tightly coupled servers enable an acceleration of local applications to meet performance demands.…”

Section: Fpga For Cloudmentioning

confidence: 99%

“…Our aim is to provide a better understanding of the current research issues in this emerging field. Several states of the art present cloud computing applications that have implemented FPGA as common hardware in data centers to accelerate remote applications [10][11][12]. Some surveys focus on the virtualization of FPGA in cloud [13][14][15] to provide an abstraction of the used FPGA hardware.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

From FPGA to Support Cloud to Cloud of FPGA: State of the Art

Skhiri

Fresse

Jamont

et al. 2019

International Journal of Reconfigurable Computing

View full text Add to dashboard Cite

Field Programmable Gate Array (FPGA) draws a significant attention from both industry and academia by accelerating computationally expensive applications and achieving low power consumption. FPGAs are interesting due to the flexibility and reconfigurabiltiy of their device. Cloud computing becomes a major trend towards infrastructure and computing resources dematerialization. It provides “unlimited” storage capacities and a large number of data and applications that make collaboration easier between multiple (not domain specific) designers. Many papers in the literature have surveyed Cloud and FPGA separately and, more precisely, their services and challenges. The acceleration of applications by FPGA and the unlimited capacities of the cloud are expected to be more and more pervasive. As more and more FPGA are being deployed in traditional cloud, it is appropriate to clarify what is the cloud FPGA and which drawbacks of using FPGA in local are resolved. We present a survey of the cloud FPGA works that have been proposed to exploit the advantages of using FPGA in the cloud. We classify these studies in three services to highlight their benefits and limitations. This survey aims at motivating further researches in cloud FPGA.

show abstract

Section: Fpga For Cloudmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

From FPGA to Support Cloud to Cloud of FPGA: State of the Art

Skhiri

Fresse

Jamont

et al. 2019

International Journal of Reconfigurable Computing

View full text Add to dashboard Cite

show abstract

“…Their strengths may derive from more efficient forms of parallelism, local/optimized memories and ad-hoc datapaths, reduced fetch and decode overhead, and support for ad-hoc data types [1], [7]. In the last years, accelerators have been commonly integrated into Systems on a Chip (SoCs) -mainly for mobile applications [8], [9], desktop workstations [10], [11], data-centers [10], [12], [13], and High-Performance Computing (HPC) systems [14].…”

Section: Introductionmentioning

confidence: 99%

A General Framework for Accelerator Management Based on ISA Extension

et al. 2022

View full text Add to dashboard Cite

Thanks to the promised improvements in performance and energy efficiency, hardware accelerators are taking momentum in many computing contexts, both in terms of variety and relative weight in the silicon area of many chips. Commonly, the way an application interacts with these hardware modules has many accelerator-specific traits and requires ad-hoc drivers that usually rely on potentially expensive system calls to manage accelerator resources and access orchestration. As a consequence, driverbased interfacing is far from uniform and can expose high latency, limiting the set of tasks suitable for acceleration. In this paper, we propose a uniform and low-latency interface based on Instruction Set Architecture (ISA) extension. All the previous studies that proposed extensions, were deeply tailored to address a single accelerator. One of the biggest disadvantages of those methods is their inability to scale. Adding more of these accelerators to one System-on-Chip (SoC) would result in ISA bloat, increasing power consumption and complexifying the decoding phase proportionally. Our proposed framework consists of a six-instruction ISA extension and the corresponding architectural support that implements the interface abstraction and the reservation logic at the hardware level. Our proposal allows controlling a broad class of integrated accelerators directly from the CPU. The proposed framework is ISA-independent, which means that it is applicable to all the existing ISAs. We implement it on the gem5 simulator by extending the RISC-V ISA. We evaluate it by simulating three compute-intensive accelerators and comparing our interfacing with a conventional driver-based one. The benchmarks highlight the performance benefits brought by our framework, with up to 10.38x speed up, as well as the ability to seamlessly support different accelerators with the same interface. The speed up advantage of our technique diminishes as the granularity of the workloads increases and the overhead for driver-based accelerators becomes less important. We also show that the impact of its hardware components on chip area and power consumption is limited.INDEX TERMS Accelerators, domain-specific architectures, heterogeneous systems, ISA extension, RISC-V.

show abstract

WITHDRAWN: Remote cloud classroom based on FPGA-based high-performance algorithms

Yang¹,

Zhu²

2020

Microprocessors and Microsystems

View full text Add to dashboard Cite

A Study of Reconfigurable Accelerators for Cloud Computing

Cited by 8 publications

References 17 publications

From FPGA to Support Cloud to Cloud of FPGA: State of the Art

From FPGA to Support Cloud to Cloud of FPGA: State of the Art

A General Framework for Accelerator Management Based on ISA Extension

WITHDRAWN: Remote cloud classroom based on FPGA-based high-performance algorithms

Contact Info

Product

Resources

About