Snap

Arashloo, Mina Tahmasbi; Koral, Yaron; Greenberg, Michael; Rexford, Jennifer; Walker, David

doi:10.1145/2934872.2934892

Cited by 126 publications

(6 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our Work TCAM, BRAM, Register 80 Gbps Openstate [7] TCAM, RAM Unkown FlowBlaze [10] TCAM, BRAM, Register 14.8 Mpps SNAP [15] CAM, Register Unkown Opp [16] TCAM, RAM, Register 10-80 Mpps SDPA [17] TCAM, RAM 0.5-10 Gbps…”

Section: Hardware Storage Throughputmentioning

confidence: 96%

“…There is a need for a hardware-based method to schedule the packets processing sequence, allowing real-time detection of RAW hazards and utilizing hardware mechanisms to avoid RAW hazards while minimizing hardware pipeline stalls. However, the current methods [9,[11][12][13][14][15] can not meet these requirements simultaneously. OPP proposed a hardware implementable state data plane architecture, but it did not consider the consistency of the flow state, which has potential risks.…”

Section: Limitations Of Prior Artmentioning

confidence: 99%

See 1 more Smart Citation

An FPGA-Based High-Performance Stateful Packet Processing Method

Lu,

Guo

2023

Micromachines

View full text Add to dashboard Cite

Compared to a stateless data plane, a stateful data plane offloads part of state information and control logic from a controller to a data plane to reduce communication overhead and improve packet processing efficiency. However, existing methods for implementing stateful data planes face challenges, particularly maintaining state consistency during packet processing and improving throughput performance. This paper presents a high-performance, FPGA (Field Programmable Gate Array)-based stateful packet processing approach, which addresses these challenges utilizing the PHV (Packet Header Vector) dynamic scheduling technique to ensure flow state consistency. Our experiments demonstrate that the proposed method could operate at 200 MHz while adding 3–12 microseconds latency. The method we proposed also provides a considerable degree of programmability.

show abstract

Section: Hardware Storage Throughputmentioning

confidence: 96%

Section: Limitations Of Prior Artmentioning

confidence: 99%

An FPGA-Based High-Performance Stateful Packet Processing Method

Lu,

Guo

2023

Micromachines

View full text Add to dashboard Cite

show abstract

“…Wang j et al [105] present a highlevel language for programming distributed collections of network switches and evaluate its performance in several industrial scenarios. Arashloo M T et al [48] propose a centralized programming model SNAP, which modifies a big switch. The role of the big switch was responsible for forwarding packets to multi-switches, so the modification on the big switch was convenient to broadcast quickly.…”

Section: Interoperability Between the Sdn And Cdnmentioning

confidence: 99%

A Review on Software Defined Content Delivery Network: A Novel Combination of CDN and SDN

Yang

Pan

2023

IEEE Access

View full text Add to dashboard Cite

With the rapid growth of multimedia services, the burden of infrastructure is becoming heavier, which may not guarantee a high-quality experience for users. Whether the multimedia category is video on demand (VOD), live broadcast, or social media is a hot issue. Traditionally, content delivery network (CDN) service providers deploy servers as close to clients as possible to reduce latency. CDN is an overlay network mainly responsible for routing requests, distribution, delivery, and audit. As a supplement to computing and storage capacity, CDN service providers have begun migrating some of their services to the cloud to focus on the delivery process. At the same time, the collaboration between CDN service providers promotes scalability to meet the growing number of requests. However, the underlying physical layer needs to be improved. The cost of the traditional underlying network is very high and cannot be easily expanded. The concept of software definition network (SDN) is proposed to solve these problems. SDN functions are divided into control layer and data layer. In the SDN, the global view is achieved to receive information by centralization, and then the deployment is dynamically adjusted to control the network. In this paper, we summarize the advantage structure and problems of a newly developed system SDCDN (software defined content distributed network). We introduce similar research progress. Then we summarize the current situation of SDN and CDN, compare and analyze their advantages and disadvantages, and finally put forward the directions of open issues and the future research.

show abstract

“…SNAP [1] is a network-centric, high-level language for network programming, extending stateless packet processing with primitive stateful operations. It offers a stateful network-wise abstraction for packet processing by enabling access to a persistent global array in control programs, while making the distribution of that state in the data plane transparent to the programmer.…”

Section: Related Workmentioning

confidence: 99%

P4cep

Kohler

Mayer

Dürr

et al. 2018

Proceedings of the 2018 Morning Workshop on in-Network Computing

View full text Add to dashboard Cite

In-network computing using programmable networking hardware is a strong trend in networking that promises to reduce latency and consumption of server resources through offloading to network elements (programmable switches and smart NICs). In particular, the data plane programming language P4 together with powerful P4 networking hardware has spawned projects offloading services into the network, e.g., consensus services or caching services. In this paper, we present a novel case for in-network computing, namely, Complex Event Processing (CEP). CEP processes streams of basic events, e.g., stemming from networked sensors, into meaningful complex events. Traditionally, CEP processing has been performed on servers or overlay networks. However, we argue in this paper that CEP is a good candidate for in-network computing along the communication path avoiding detouring streams to distant servers to minimize communication latency while also exploiting processing capabilities of novel networking hardware. We show that it is feasible to express CEP operations in P4 and also present a tool to compile CEP operations, formulated in our P4CEP rule specification language, to P4 code. Moreover, we identify challenges and problems that we have encountered to show future research directions for implementing full-fledged in-network CEP systems. CCS CONCEPTS• Networks → Programmable networks; In-network processing; • Software and its engineering → Publish-subscribe / event-based architectures; Message oriented middleware;

show abstract

Snap

Cited by 126 publications

References 34 publications

An FPGA-Based High-Performance Stateful Packet Processing Method

An FPGA-Based High-Performance Stateful Packet Processing Method

A Review on Software Defined Content Delivery Network: A Novel Combination of CDN and SDN

P4cep

Contact Info

Product

Resources

About