PacketShader

Han, Sangjin; Jang, Keon; Park, KyoungSoo; Moon, Sang Heup

doi:10.1145/1851182.1851207

Cited by 316 publications

(30 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Figure 11 illustrates the pipelined organization of the system. There are three pipeline stages in the system: -Input Stage (shown IP in Figure 11): One core in this stage receives packets from 10Gbps network interface cards (NIC) through an optimized Linux NIC driver [8], and then performs load balance by distributing the packets among multiple pipelines. -Application Stage (shown AP in Figure 11): Each core in this stage gets packets by polling a CLF queue that connects it to the IP core.…”

Section: Performance Evaluation On Real Testbedmentioning

confidence: 99%

See 1 more Smart Citation

B-Queue: Efficient and Practical Queuing for Fast Core-to-Core Communication

Wang

Zhang

Tang

et al. 2012

Int J Parallel Prog

View full text Add to dashboard Cite

Core-to-core communication is critical to the effective use of multi-core processors. A number of software based concurrent lock-free queues have been proposed to address this problem. Existing solutions, however, suffer from performance degradation in real testbeds, or rely on auxiliary hardware or software timers to handle the deadlock problem when batching is used, making those solutions good in theory but difficult to use in practice. This paper describes the pros and cons of existing concurrent lock-free queues in both dummy and real testbeds and proposes BQueue, an efficient and practical single-producer-single-consumer concurrent lockfree queue that solves the deadlock problem gracefully by introducing a self-adaptive backtracking mechanism. Experiments show that in real massively-parallel applications, B-Queue is faster than FastForward and MCRingBuffer, the two state-of-the-art concurrent lock-free queues, by up to 10x and 5x, respectively. Moreover, B-Queue outperforms FastForward and MCRingBuffer in terms of stability and scalability, making it a good candidate for fast core-to-core communication on multi-core architectures.

show abstract

Section: Performance Evaluation On Real Testbedmentioning

confidence: 99%

“…This paper studies CLF queues in building up a real multi-10Gbps network processing system where parallelism is widely used [6,7,18,35], and both the hardware and software capabilities are stressed [5,8]. The major contributions of this paper are as follows:…”

Section: Introductionmentioning

confidence: 99%

B-Queue: Efficient and Practical Queuing for Fast Core-to-Core Communication

Wang

Zhang

Tang

et al. 2012

Int J Parallel Prog

View full text Add to dashboard Cite

show abstract

“…353 . A recent paper on GPU-accelerated software packet router says 350 : "We believe that the increased power consumption is tolerable, considering the performance improvement from GPUs." But Finally it turns out that The GPU remains a specialized processor, so that we still need the traditional CPU after all.…”

Section: The Role Of Graphics Processorsmentioning

confidence: 99%

The Paramountcy of Reconfigurable Computing

Hartenstein¹

2012

Energy‐Efficient Distributed Computing Systems

View full text Add to dashboard Cite

“…The researchers of the "Networked & Distributed Computing Systems Lab" in Korea released papers about how to speed-up SSL [15] or how to accelerate a software router-also in connection with IPsec [16]with GPUs.…”

Section: Related Workmentioning

confidence: 99%

GPU-Assisted AES Encryption Using GCM

Schönberger

Fuß

2011

Communications and Multimedia Security

View full text Add to dashboard Cite

Abstract.We are presenting an implementation of the Galois/Counter Mode (GCM) for the Advanced Encryption Standard (AES) in IPsec in this paper. GCM is a so called "authenticated encryption" as it can ensure confidentiality, integrity and authentication. It uses the Counter Mode for encryption, therefore counters are encrypted for an exclusive-OR with the plaintext. We describe a technique where these encryptions are precomputed on a Graphic Processing Unit (GPU) and can later be used to encrypt the plaintext, whereupon only the exclusive-OR and authentication part of GCM are left to be computed. This technique should primarily not limit the performance to the speed of the AES implementation but allow Gigabit throughput and at the same time minimize the CPU load.

show abstract

PacketShader

Cited by 316 publications

References 29 publications

B-Queue: Efficient and Practical Queuing for Fast Core-to-Core Communication

B-Queue: Efficient and Practical Queuing for Fast Core-to-Core Communication

The Paramountcy of Reconfigurable Computing

GPU-Assisted AES Encryption Using GCM

Contact Info

Product

Resources

About