2017 IEEE 18th International Conference on High Performance Switching and Routing (HPSR) 2017
DOI: 10.1109/hpsr.2017.7968684
|View full text |Cite
|
Sign up to set email alerts
|

MACSAD: High performance dataplane applications on the move

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
5
0

Year Published

2018
2018
2023
2023

Publication Types

Select...
6
2

Relationship

1
7

Authors

Journals

citations
Cited by 8 publications
(5 citation statements)
references
References 12 publications
0
5
0
Order By: Relevance
“…Our needs on a single packet generator tool to meet the packet trace requirements of our research use cases to evaluate the performance and scalability of programmable data planes [Patra et al 2016, Patra et al 2017] and the identified limitations of existing tools motivated our work to develop a packet generator prioritizing two essential characteristics: (i) simple to use, (ii) wide protocol support and rich customizability. To that end, as our baseline, we opted for the well known Scapy python library [Biondi 2008], which is easy to extend and manipulate.…”
Section: Motivation and Goalsmentioning
confidence: 99%
See 1 more Smart Citation
“…Our needs on a single packet generator tool to meet the packet trace requirements of our research use cases to evaluate the performance and scalability of programmable data planes [Patra et al 2016, Patra et al 2017] and the identified limitations of existing tools motivated our work to develop a packet generator prioritizing two essential characteristics: (i) simple to use, (ii) wide protocol support and rich customizability. To that end, as our baseline, we opted for the well known Scapy python library [Biondi 2008], which is easy to extend and manipulate.…”
Section: Motivation and Goalsmentioning
confidence: 99%
“…Multi-Architecture Compiler System for Abstract Dataplanes (MACSAD) [Patra et al 2016, Patra et al 2017 tries to converge Programming Protocol-Independent Packet Processors (P4) [Bosshart et al 2014] and OpenDataPlane (ODP) [OpenDataPlane 2013] through a common compilation process delivering portability of dataplane applications without compromising target performance improvements while translating P4-defined dataplane abstractions into high-level ODP Application Programming Interfaces (APIs).…”
Section: Use Casementioning
confidence: 99%
“…Multi-Architecture Compiler System for Abstract Dataplanes (MACSAD) [Patra et al 2016, Patra et al 2017 is an approach to converge P4 and ODP through a common compilation process delivering portability of dataplane applications without compromising target performance. During it's development we observed the absence of IPv6 lookup mechanism in ODP (limited to IPv4 and no short-term plans related to this implementation by ODP team) to support more levels of addressing hierarchy, and a greater number of addressable nodes, and proposed an ODP Helper library with a lookup mechanism for IPv6 as a contribution to ODP open-source community and also extended the effort to add the support to MACSAD too.…”
Section: Introductionmentioning
confidence: 99%
“…With P4 and ODP working together, it is possible to determine and program dataplanes beyond multiple targets with a common compiler system. MACSAD (PATRA et al, 2016;PATRA et al, 2017) is an approach to accommodate P4 and ODP through a common compilation process delivering portability of dataplane applications without compromising target performance improvements, translating P4-deĄned dataplanes into high-level ODP APIs. MACSAD has a designed compiler module that generates an Intermediate Representation (IR) for P4 applications.…”
Section: Introductionmentioning
confidence: 99%
“…Our need for a single packet generator tool to meet the packet trace requirements of our research use cases to evaluate the performance and scalability of programmable dataplanes (PATRA et al, 2016;PATRA et al, 2017;PATRA et al, 2018) and identify the limitations of existing tools, motivated our work to develop a packet generator prioritizing two essential characteristics: (i) simple to use, (ii) wide protocol support and rich customizability. To that end, as our baseline, we opted for the well known Scapy python library (BIONDI, 2008) for easier extension and packet manipulation.…”
Section: Introductionmentioning
confidence: 99%