SlideShare a Scribd company logo
PCルータ研究歴史
小原 泰弘
NTTコミュニケーションズ
yasu@nttv6.jp
目次
• Click (452Kpps/333Kpps) (ACM SOSP ’99)
• RouteBricks (18.96Mpps/12(?)Mpps) (ACM SOSP
‘09)
• PacketShader (58.4Mpps/50+(?)Mpps) (ACM
SIGCOMM ’11)
• NetMap (USENIX ATC ’12)
• DIR-24-8-BASIC
• Waldvogel
• E. Kohler, R. Morris, B. Chen, J. Jannotti, and M. F. Kaashoek., “The Click modular
router”, ACM Transactions on Computer Systems 18(3), August 2000, pages 263-
297.
• SOSP ‘99, TOCS ’00
• Still active
– Used in SEATTLE(SIGCOMM’08)
– Used in RouteBricks(SOSP’09)
– Last release: 2011/09/24
– Last commit: 2014/04/06
• Interrupt v.s. Polling (i.e., spin-wait, busy-wait)
• Pentium III 700 MHz, 100Mbps Ether DEC tulip PCI, …
• Linux 84kpps, Polling Linux 284kpps
• Routing 333kpps, Forwarding 452kpps
• Time taken: 2,798 ns/packet
Click Modular Router (1)
Click Modular Router (2)
RouteBricks (1)
• M. Dobrescu, N. Egi, K. Argyraki, B.-G. Chun, K. Fall, Gianluca Iannaccone, Allan Knies, Maziar Manesh,
Sylvia Ratnasamy, “RouteBricks: Exploiting Parallelism To Scale Software Routers, ” 22nd ACM Symposium
on Operating Systems Principles (SOSP), October 2009
• SOSP ’09
• Click-base
• Leverages:
– multi-cores.
– multi-queue NICs.
– batch processing.
• NUMA-aware data placement didn’t make differences (in contrast
to PacketShader).
• Achieved 18.96Mpps Forwarding, 12(?)Mpps Routing
• (Routing: 256K routes with memory expansion (DIR-24-8-BASIC))
RouteBricks (2)
PacketShader (1)
• S. Han, K. Jang, K. Park and S. Moon. “PacketShader: a GPU-accelerated
Software Router.” In proceedings of ACM SIGCOMM 2010, Delhi, India.
September 2010.
• Speeds routing lookups (not just forwarding)
• Forwarding: 41.1 Gbps or 58.4 Mpps
• IPv4 Routing (DIR-24-8-BASIC, 282K routes):
39Gbps.
• IPv6 Routing (Waldvogel, 200K routes): 38Gbps.
• Up to 130us/200us (IPv4/IPv6) routing delay
(260us/400us for roundtrip)
PacketShader (2)
• Not a pps-limitation ?
PacketShader (3)
(a) Cores contend for queues.
Leads to expensive cache bouncing
and lock contention.
(b) Better coupling of queues and
cores.
pipeline batch pipeline
• Buffer management
• Pipelining
• Batch processing
• Decoupling among queues and cores
NetMap (1)
• Luigi Rizzo, “netmap: a novel framework for fast packet I/O,” Usenix ATC'12,
Boston, June 2012
• Forwarding 14.88 Mpps (1 core 900 MHz)
• Objective: to move packets quickly between
the apps and the network cards.
• Reduced copy, modified buffer management.
• Protection. User-land v.s. Kernel.
• 408 ns / batch.
• Implementation on Linux/FreeBSD/Click.
NetMap (2)
Techniques
• Polling (i.e., spin-wait or busy-wait)
• Parallelization
– i.e., multi-queues, multi-cores, GPUs
• Pipelining
• Reduce overhead
– Batch processing (amortized over a batch)
– Remove copy, improve buffer managements
Routing tables
DIR-24-8-BASIC
• P. Gupta, S. Lin, and N. McKeown. “Routing Lookups in Hardware at
Memory Access Speeds.” In Proceedings of the IEEE INFOCOM Conference,
San Francisco, CA, USA, March 1998.
• INFOCOM ‘98
Waldvogel
• M. Waldvogel, G. Varghese, J. Turner, B. Plattner. “Scalable High-Speed IP
Routing Lookups.” Proc. ACM SIGCOMM 1997 , pp. 25-36, Cannes, France.

More Related Content

What's hot

Achieving congestion diversity in multi hop wireless mesh networks
Achieving congestion diversity in multi hop wireless mesh networksAchieving congestion diversity in multi hop wireless mesh networks
Achieving congestion diversity in multi hop wireless mesh networks
ieeeprojectschennai
 
Network on Chip Architecture and Routing Techniques: A survey
Network on Chip Architecture and Routing Techniques: A surveyNetwork on Chip Architecture and Routing Techniques: A survey
Network on Chip Architecture and Routing Techniques: A survey
IJRES Journal
 
Coscup2021 open source network os for datacenter
Coscup2021  open source network os for datacenterCoscup2021  open source network os for datacenter
Coscup2021 open source network os for datacenter
Dung-Ru Tsai
 
Network on chip by shoeb
Network on chip by shoebNetwork on chip by shoeb
Network on chip by shoeb
Shoeb Ahmed
 
TRACK B: Multicores & Network On Chip Architectures/ Oren Hollander
TRACK B: Multicores & Network On Chip Architectures/ Oren HollanderTRACK B: Multicores & Network On Chip Architectures/ Oren Hollander
TRACK B: Multicores & Network On Chip Architectures/ Oren Hollander
chiportal
 
ODSA PoC: Network Flow Processor Overview
ODSA PoC: Network Flow Processor OverviewODSA PoC: Network Flow Processor Overview
ODSA PoC: Network Flow Processor Overview
jennimenni
 
Univa and SUSE at SC17: Scaling Machine Learning for SUSE Linux Containers, S...
Univa and SUSE at SC17: Scaling Machine Learning for SUSE Linux Containers, S...Univa and SUSE at SC17: Scaling Machine Learning for SUSE Linux Containers, S...
Univa and SUSE at SC17: Scaling Machine Learning for SUSE Linux Containers, S...
Ian Lumb
 
Managing Containerized HPC and AI Workloads on TSUBAME3.0
Managing Containerized HPC and AI Workloads on TSUBAME3.0Managing Containerized HPC and AI Workloads on TSUBAME3.0
Managing Containerized HPC and AI Workloads on TSUBAME3.0
Ian Lumb
 
ZigBee
ZigBee ZigBee
Presentation reliable NoC
Presentation reliable NoCPresentation reliable NoC
Presentation reliable NoC
shahanianmol
 
Deccan RubyConf 2016 - Lighning Talk - SpiceRub
Deccan RubyConf 2016 - Lighning Talk - SpiceRubDeccan RubyConf 2016 - Lighning Talk - SpiceRub
Deccan RubyConf 2016 - Lighning Talk - SpiceRub
Gaurav Tamba
 
Devils Logic PDR presentation
Devils Logic PDR presentationDevils Logic PDR presentation
Devils Logic PDR presentation
Shota Ichikawa
 
Tsfpga
TsfpgaTsfpga
Emerging Technologies in On-Chip and Off-Chip Interconnection Networks
Emerging Technologies in On-Chip and Off-Chip Interconnection NetworksEmerging Technologies in On-Chip and Off-Chip Interconnection Networks
Emerging Technologies in On-Chip and Off-Chip Interconnection Networks
Ashif Sikder
 
ICIECA 2014 Paper 10
ICIECA 2014 Paper 10ICIECA 2014 Paper 10
Mobile Ad-hoc and Sensor Networks
Mobile Ad-hoc and Sensor NetworksMobile Ad-hoc and Sensor Networks
Mobile Ad-hoc and Sensor Networks
Sushant Kushwaha
 
Arteris network on chip: The growing cost of wires
Arteris network on chip:  The growing cost of wiresArteris network on chip:  The growing cost of wires
Arteris network on chip: The growing cost of wires
Arteris
 
Cassandra 1.2 by Eddie Satterly
Cassandra 1.2 by Eddie SatterlyCassandra 1.2 by Eddie Satterly
Cassandra 1.2 by Eddie Satterly
DataStax Academy
 
Lustre, RoCE, and MAN
Lustre, RoCE, and MANLustre, RoCE, and MAN
Lustre, RoCE, and MAN
inside-BigData.com
 
Alexander Ignatyev "MapReduce infrastructure"
Alexander Ignatyev "MapReduce infrastructure"Alexander Ignatyev "MapReduce infrastructure"
Alexander Ignatyev "MapReduce infrastructure"
Yandex
 

What's hot (20)

Achieving congestion diversity in multi hop wireless mesh networks
Achieving congestion diversity in multi hop wireless mesh networksAchieving congestion diversity in multi hop wireless mesh networks
Achieving congestion diversity in multi hop wireless mesh networks
 
Network on Chip Architecture and Routing Techniques: A survey
Network on Chip Architecture and Routing Techniques: A surveyNetwork on Chip Architecture and Routing Techniques: A survey
Network on Chip Architecture and Routing Techniques: A survey
 
Coscup2021 open source network os for datacenter
Coscup2021  open source network os for datacenterCoscup2021  open source network os for datacenter
Coscup2021 open source network os for datacenter
 
Network on chip by shoeb
Network on chip by shoebNetwork on chip by shoeb
Network on chip by shoeb
 
TRACK B: Multicores & Network On Chip Architectures/ Oren Hollander
TRACK B: Multicores & Network On Chip Architectures/ Oren HollanderTRACK B: Multicores & Network On Chip Architectures/ Oren Hollander
TRACK B: Multicores & Network On Chip Architectures/ Oren Hollander
 
ODSA PoC: Network Flow Processor Overview
ODSA PoC: Network Flow Processor OverviewODSA PoC: Network Flow Processor Overview
ODSA PoC: Network Flow Processor Overview
 
Univa and SUSE at SC17: Scaling Machine Learning for SUSE Linux Containers, S...
Univa and SUSE at SC17: Scaling Machine Learning for SUSE Linux Containers, S...Univa and SUSE at SC17: Scaling Machine Learning for SUSE Linux Containers, S...
Univa and SUSE at SC17: Scaling Machine Learning for SUSE Linux Containers, S...
 
Managing Containerized HPC and AI Workloads on TSUBAME3.0
Managing Containerized HPC and AI Workloads on TSUBAME3.0Managing Containerized HPC and AI Workloads on TSUBAME3.0
Managing Containerized HPC and AI Workloads on TSUBAME3.0
 
ZigBee
ZigBee ZigBee
ZigBee
 
Presentation reliable NoC
Presentation reliable NoCPresentation reliable NoC
Presentation reliable NoC
 
Deccan RubyConf 2016 - Lighning Talk - SpiceRub
Deccan RubyConf 2016 - Lighning Talk - SpiceRubDeccan RubyConf 2016 - Lighning Talk - SpiceRub
Deccan RubyConf 2016 - Lighning Talk - SpiceRub
 
Devils Logic PDR presentation
Devils Logic PDR presentationDevils Logic PDR presentation
Devils Logic PDR presentation
 
Tsfpga
TsfpgaTsfpga
Tsfpga
 
Emerging Technologies in On-Chip and Off-Chip Interconnection Networks
Emerging Technologies in On-Chip and Off-Chip Interconnection NetworksEmerging Technologies in On-Chip and Off-Chip Interconnection Networks
Emerging Technologies in On-Chip and Off-Chip Interconnection Networks
 
ICIECA 2014 Paper 10
ICIECA 2014 Paper 10ICIECA 2014 Paper 10
ICIECA 2014 Paper 10
 
Mobile Ad-hoc and Sensor Networks
Mobile Ad-hoc and Sensor NetworksMobile Ad-hoc and Sensor Networks
Mobile Ad-hoc and Sensor Networks
 
Arteris network on chip: The growing cost of wires
Arteris network on chip:  The growing cost of wiresArteris network on chip:  The growing cost of wires
Arteris network on chip: The growing cost of wires
 
Cassandra 1.2 by Eddie Satterly
Cassandra 1.2 by Eddie SatterlyCassandra 1.2 by Eddie Satterly
Cassandra 1.2 by Eddie Satterly
 
Lustre, RoCE, and MAN
Lustre, RoCE, and MANLustre, RoCE, and MAN
Lustre, RoCE, and MAN
 
Alexander Ignatyev "MapReduce infrastructure"
Alexander Ignatyev "MapReduce infrastructure"Alexander Ignatyev "MapReduce infrastructure"
Alexander Ignatyev "MapReduce infrastructure"
 

Similar to past-research-on-pc-router

Cisco crs1
Cisco crs1Cisco crs1
Cisco crs1
wjunjmt
 
NUMA-aware Scalable Graph Traversal on SGI UV Systems
NUMA-aware Scalable Graph Traversal on SGI UV SystemsNUMA-aware Scalable Graph Traversal on SGI UV Systems
NUMA-aware Scalable Graph Traversal on SGI UV Systems
Yuichiro Yasui
 
Brkdct 3101
Brkdct 3101Brkdct 3101
Brkdct 3101
Nguyen Van Linh
 
Cisco CCNA Data Center Networking Fundamentals
Cisco CCNA Data Center Networking FundamentalsCisco CCNA Data Center Networking Fundamentals
Cisco CCNA Data Center Networking Fundamentals
E.S.G. JR. Consulting, Inc.
 
SRv6 On-Path Delay Measurement with Anomaly Detection OPSAWG WG
SRv6 On-Path Delay Measurement with Anomaly Detection OPSAWG WGSRv6 On-Path Delay Measurement with Anomaly Detection OPSAWG WG
SRv6 On-Path Delay Measurement with Anomaly Detection OPSAWG WG
ThomasGraf42
 
Barcelona Supercomputing Center, Generador de Riqueza
Barcelona Supercomputing Center, Generador de RiquezaBarcelona Supercomputing Center, Generador de Riqueza
Barcelona Supercomputing Center, Generador de Riqueza
Facultad de Informática UCM
 
Recent advance in netmap/VALE(mSwitch)
Recent advance in netmap/VALE(mSwitch)Recent advance in netmap/VALE(mSwitch)
Recent advance in netmap/VALE(mSwitch)
micchie
 
End nodes in the Multigigabit era
End nodes in the Multigigabit eraEnd nodes in the Multigigabit era
End nodes in the Multigigabit era
rinnocente
 
Lte overview titus
Lte overview titusLte overview titus
Lte overview titus
Nabil Al_Mutawakel
 
Network-aware Data Management for Large Scale Distributed Applications, IBM R...
Network-aware Data Management for Large Scale Distributed Applications, IBM R...Network-aware Data Management for Large Scale Distributed Applications, IBM R...
Network-aware Data Management for Large Scale Distributed Applications, IBM R...
balmanme
 
High performace network of Cloud Native Taiwan User Group
High performace network of Cloud Native Taiwan User GroupHigh performace network of Cloud Native Taiwan User Group
High performace network of Cloud Native Taiwan User Group
HungWei Chiu
 
cisco-n3k-c3064pq-10gx-datasheet.pdf
cisco-n3k-c3064pq-10gx-datasheet.pdfcisco-n3k-c3064pq-10gx-datasheet.pdf
cisco-n3k-c3064pq-10gx-datasheet.pdf
Hi-Network.com
 
Distance Vector Multicast Routing Protocol (DVMRP) : Combined Presentation
Distance Vector Multicast Routing Protocol (DVMRP) : Combined PresentationDistance Vector Multicast Routing Protocol (DVMRP) : Combined Presentation
Distance Vector Multicast Routing Protocol (DVMRP) : Combined Presentation
Subhajit Sahu
 
NST Product Catalog
NST Product CatalogNST Product Catalog
NST Product Catalog
moonhyo
 
GPRS/EDGE Basics / knowledge sharing
GPRS/EDGE Basics /  knowledge sharingGPRS/EDGE Basics /  knowledge sharing
GPRS/EDGE Basics / knowledge sharing
Mustafa Golam
 
802.11.ppt
802.11.ppt802.11.ppt
802.11.ppt
SatishRajuGorkhode
 
Performance Analysis of Mesh-based NoC’s on Routing Algorithms
Performance Analysis of Mesh-based NoC’s on Routing Algorithms Performance Analysis of Mesh-based NoC’s on Routing Algorithms
Performance Analysis of Mesh-based NoC’s on Routing Algorithms
IJECEIAES
 
100 M pps on PC.
100 M pps on PC.100 M pps on PC.
100 M pps on PC.
Redge Technologies
 
Dynamic Routing All Algorithms, Working And Basics
Dynamic Routing All Algorithms, Working And BasicsDynamic Routing All Algorithms, Working And Basics
Dynamic Routing All Algorithms, Working And Basics
Harsh Mehta
 
An FPGA for high end Open Networking
An FPGA for high end Open NetworkingAn FPGA for high end Open Networking
An FPGA for high end Open Networking
rinnocente
 

Similar to past-research-on-pc-router (20)

Cisco crs1
Cisco crs1Cisco crs1
Cisco crs1
 
NUMA-aware Scalable Graph Traversal on SGI UV Systems
NUMA-aware Scalable Graph Traversal on SGI UV SystemsNUMA-aware Scalable Graph Traversal on SGI UV Systems
NUMA-aware Scalable Graph Traversal on SGI UV Systems
 
Brkdct 3101
Brkdct 3101Brkdct 3101
Brkdct 3101
 
Cisco CCNA Data Center Networking Fundamentals
Cisco CCNA Data Center Networking FundamentalsCisco CCNA Data Center Networking Fundamentals
Cisco CCNA Data Center Networking Fundamentals
 
SRv6 On-Path Delay Measurement with Anomaly Detection OPSAWG WG
SRv6 On-Path Delay Measurement with Anomaly Detection OPSAWG WGSRv6 On-Path Delay Measurement with Anomaly Detection OPSAWG WG
SRv6 On-Path Delay Measurement with Anomaly Detection OPSAWG WG
 
Barcelona Supercomputing Center, Generador de Riqueza
Barcelona Supercomputing Center, Generador de RiquezaBarcelona Supercomputing Center, Generador de Riqueza
Barcelona Supercomputing Center, Generador de Riqueza
 
Recent advance in netmap/VALE(mSwitch)
Recent advance in netmap/VALE(mSwitch)Recent advance in netmap/VALE(mSwitch)
Recent advance in netmap/VALE(mSwitch)
 
End nodes in the Multigigabit era
End nodes in the Multigigabit eraEnd nodes in the Multigigabit era
End nodes in the Multigigabit era
 
Lte overview titus
Lte overview titusLte overview titus
Lte overview titus
 
Network-aware Data Management for Large Scale Distributed Applications, IBM R...
Network-aware Data Management for Large Scale Distributed Applications, IBM R...Network-aware Data Management for Large Scale Distributed Applications, IBM R...
Network-aware Data Management for Large Scale Distributed Applications, IBM R...
 
High performace network of Cloud Native Taiwan User Group
High performace network of Cloud Native Taiwan User GroupHigh performace network of Cloud Native Taiwan User Group
High performace network of Cloud Native Taiwan User Group
 
cisco-n3k-c3064pq-10gx-datasheet.pdf
cisco-n3k-c3064pq-10gx-datasheet.pdfcisco-n3k-c3064pq-10gx-datasheet.pdf
cisco-n3k-c3064pq-10gx-datasheet.pdf
 
Distance Vector Multicast Routing Protocol (DVMRP) : Combined Presentation
Distance Vector Multicast Routing Protocol (DVMRP) : Combined PresentationDistance Vector Multicast Routing Protocol (DVMRP) : Combined Presentation
Distance Vector Multicast Routing Protocol (DVMRP) : Combined Presentation
 
NST Product Catalog
NST Product CatalogNST Product Catalog
NST Product Catalog
 
GPRS/EDGE Basics / knowledge sharing
GPRS/EDGE Basics /  knowledge sharingGPRS/EDGE Basics /  knowledge sharing
GPRS/EDGE Basics / knowledge sharing
 
802.11.ppt
802.11.ppt802.11.ppt
802.11.ppt
 
Performance Analysis of Mesh-based NoC’s on Routing Algorithms
Performance Analysis of Mesh-based NoC’s on Routing Algorithms Performance Analysis of Mesh-based NoC’s on Routing Algorithms
Performance Analysis of Mesh-based NoC’s on Routing Algorithms
 
100 M pps on PC.
100 M pps on PC.100 M pps on PC.
100 M pps on PC.
 
Dynamic Routing All Algorithms, Working And Basics
Dynamic Routing All Algorithms, Working And BasicsDynamic Routing All Algorithms, Working And Basics
Dynamic Routing All Algorithms, Working And Basics
 
An FPGA for high end Open Networking
An FPGA for high end Open NetworkingAn FPGA for high end Open Networking
An FPGA for high end Open Networking
 

Recently uploaded

Y-Combinator seed pitch deck template PP
Y-Combinator seed pitch deck template PPY-Combinator seed pitch deck template PP
Y-Combinator seed pitch deck template PP
c5vrf27qcz
 
Apps Break Data
Apps Break DataApps Break Data
Apps Break Data
Ivo Velitchkov
 
JavaLand 2024: Application Development Green Masterplan
JavaLand 2024: Application Development Green MasterplanJavaLand 2024: Application Development Green Masterplan
JavaLand 2024: Application Development Green Masterplan
Miro Wengner
 
Overcoming the PLG Trap: Lessons from Canva's Head of Sales & Head of EMEA Da...
Overcoming the PLG Trap: Lessons from Canva's Head of Sales & Head of EMEA Da...Overcoming the PLG Trap: Lessons from Canva's Head of Sales & Head of EMEA Da...
Overcoming the PLG Trap: Lessons from Canva's Head of Sales & Head of EMEA Da...
saastr
 
Session 1 - Intro to Robotic Process Automation.pdf
Session 1 - Intro to Robotic Process Automation.pdfSession 1 - Intro to Robotic Process Automation.pdf
Session 1 - Intro to Robotic Process Automation.pdf
UiPathCommunity
 
LF Energy Webinar: Carbon Data Specifications: Mechanisms to Improve Data Acc...
LF Energy Webinar: Carbon Data Specifications: Mechanisms to Improve Data Acc...LF Energy Webinar: Carbon Data Specifications: Mechanisms to Improve Data Acc...
LF Energy Webinar: Carbon Data Specifications: Mechanisms to Improve Data Acc...
DanBrown980551
 
Columbus Data & Analytics Wednesdays - June 2024
Columbus Data & Analytics Wednesdays - June 2024Columbus Data & Analytics Wednesdays - June 2024
Columbus Data & Analytics Wednesdays - June 2024
Jason Packer
 
Main news related to the CCS TSI 2023 (2023/1695)
Main news related to the CCS TSI 2023 (2023/1695)Main news related to the CCS TSI 2023 (2023/1695)
Main news related to the CCS TSI 2023 (2023/1695)
Jakub Marek
 
AppSec PNW: Android and iOS Application Security with MobSF
AppSec PNW: Android and iOS Application Security with MobSFAppSec PNW: Android and iOS Application Security with MobSF
AppSec PNW: Android and iOS Application Security with MobSF
Ajin Abraham
 
Connector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectors
Connector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectorsConnector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectors
Connector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectors
DianaGray10
 
Dandelion Hashtable: beyond billion requests per second on a commodity server
Dandelion Hashtable: beyond billion requests per second on a commodity serverDandelion Hashtable: beyond billion requests per second on a commodity server
Dandelion Hashtable: beyond billion requests per second on a commodity server
Antonios Katsarakis
 
Taking AI to the Next Level in Manufacturing.pdf
Taking AI to the Next Level in Manufacturing.pdfTaking AI to the Next Level in Manufacturing.pdf
Taking AI to the Next Level in Manufacturing.pdf
ssuserfac0301
 
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
Edge AI and Vision Alliance
 
"Choosing proper type of scaling", Olena Syrota
"Choosing proper type of scaling", Olena Syrota"Choosing proper type of scaling", Olena Syrota
"Choosing proper type of scaling", Olena Syrota
Fwdays
 
Nordic Marketo Engage User Group_June 13_ 2024.pptx
Nordic Marketo Engage User Group_June 13_ 2024.pptxNordic Marketo Engage User Group_June 13_ 2024.pptx
Nordic Marketo Engage User Group_June 13_ 2024.pptx
MichaelKnudsen27
 
Astute Business Solutions | Oracle Cloud Partner |
Astute Business Solutions | Oracle Cloud Partner |Astute Business Solutions | Oracle Cloud Partner |
Astute Business Solutions | Oracle Cloud Partner |
AstuteBusiness
 
inQuba Webinar Mastering Customer Journey Management with Dr Graham Hill
inQuba Webinar Mastering Customer Journey Management with Dr Graham HillinQuba Webinar Mastering Customer Journey Management with Dr Graham Hill
inQuba Webinar Mastering Customer Journey Management with Dr Graham Hill
LizaNolte
 
A Deep Dive into ScyllaDB's Architecture
A Deep Dive into ScyllaDB's ArchitectureA Deep Dive into ScyllaDB's Architecture
A Deep Dive into ScyllaDB's Architecture
ScyllaDB
 
Skybuffer SAM4U tool for SAP license adoption
Skybuffer SAM4U tool for SAP license adoptionSkybuffer SAM4U tool for SAP license adoption
Skybuffer SAM4U tool for SAP license adoption
Tatiana Kojar
 
Must Know Postgres Extension for DBA and Developer during Migration
Must Know Postgres Extension for DBA and Developer during MigrationMust Know Postgres Extension for DBA and Developer during Migration
Must Know Postgres Extension for DBA and Developer during Migration
Mydbops
 

Recently uploaded (20)

Y-Combinator seed pitch deck template PP
Y-Combinator seed pitch deck template PPY-Combinator seed pitch deck template PP
Y-Combinator seed pitch deck template PP
 
Apps Break Data
Apps Break DataApps Break Data
Apps Break Data
 
JavaLand 2024: Application Development Green Masterplan
JavaLand 2024: Application Development Green MasterplanJavaLand 2024: Application Development Green Masterplan
JavaLand 2024: Application Development Green Masterplan
 
Overcoming the PLG Trap: Lessons from Canva's Head of Sales & Head of EMEA Da...
Overcoming the PLG Trap: Lessons from Canva's Head of Sales & Head of EMEA Da...Overcoming the PLG Trap: Lessons from Canva's Head of Sales & Head of EMEA Da...
Overcoming the PLG Trap: Lessons from Canva's Head of Sales & Head of EMEA Da...
 
Session 1 - Intro to Robotic Process Automation.pdf
Session 1 - Intro to Robotic Process Automation.pdfSession 1 - Intro to Robotic Process Automation.pdf
Session 1 - Intro to Robotic Process Automation.pdf
 
LF Energy Webinar: Carbon Data Specifications: Mechanisms to Improve Data Acc...
LF Energy Webinar: Carbon Data Specifications: Mechanisms to Improve Data Acc...LF Energy Webinar: Carbon Data Specifications: Mechanisms to Improve Data Acc...
LF Energy Webinar: Carbon Data Specifications: Mechanisms to Improve Data Acc...
 
Columbus Data & Analytics Wednesdays - June 2024
Columbus Data & Analytics Wednesdays - June 2024Columbus Data & Analytics Wednesdays - June 2024
Columbus Data & Analytics Wednesdays - June 2024
 
Main news related to the CCS TSI 2023 (2023/1695)
Main news related to the CCS TSI 2023 (2023/1695)Main news related to the CCS TSI 2023 (2023/1695)
Main news related to the CCS TSI 2023 (2023/1695)
 
AppSec PNW: Android and iOS Application Security with MobSF
AppSec PNW: Android and iOS Application Security with MobSFAppSec PNW: Android and iOS Application Security with MobSF
AppSec PNW: Android and iOS Application Security with MobSF
 
Connector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectors
Connector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectorsConnector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectors
Connector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectors
 
Dandelion Hashtable: beyond billion requests per second on a commodity server
Dandelion Hashtable: beyond billion requests per second on a commodity serverDandelion Hashtable: beyond billion requests per second on a commodity server
Dandelion Hashtable: beyond billion requests per second on a commodity server
 
Taking AI to the Next Level in Manufacturing.pdf
Taking AI to the Next Level in Manufacturing.pdfTaking AI to the Next Level in Manufacturing.pdf
Taking AI to the Next Level in Manufacturing.pdf
 
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
 
"Choosing proper type of scaling", Olena Syrota
"Choosing proper type of scaling", Olena Syrota"Choosing proper type of scaling", Olena Syrota
"Choosing proper type of scaling", Olena Syrota
 
Nordic Marketo Engage User Group_June 13_ 2024.pptx
Nordic Marketo Engage User Group_June 13_ 2024.pptxNordic Marketo Engage User Group_June 13_ 2024.pptx
Nordic Marketo Engage User Group_June 13_ 2024.pptx
 
Astute Business Solutions | Oracle Cloud Partner |
Astute Business Solutions | Oracle Cloud Partner |Astute Business Solutions | Oracle Cloud Partner |
Astute Business Solutions | Oracle Cloud Partner |
 
inQuba Webinar Mastering Customer Journey Management with Dr Graham Hill
inQuba Webinar Mastering Customer Journey Management with Dr Graham HillinQuba Webinar Mastering Customer Journey Management with Dr Graham Hill
inQuba Webinar Mastering Customer Journey Management with Dr Graham Hill
 
A Deep Dive into ScyllaDB's Architecture
A Deep Dive into ScyllaDB's ArchitectureA Deep Dive into ScyllaDB's Architecture
A Deep Dive into ScyllaDB's Architecture
 
Skybuffer SAM4U tool for SAP license adoption
Skybuffer SAM4U tool for SAP license adoptionSkybuffer SAM4U tool for SAP license adoption
Skybuffer SAM4U tool for SAP license adoption
 
Must Know Postgres Extension for DBA and Developer during Migration
Must Know Postgres Extension for DBA and Developer during MigrationMust Know Postgres Extension for DBA and Developer during Migration
Must Know Postgres Extension for DBA and Developer during Migration
 

past-research-on-pc-router

  • 2. 目次 • Click (452Kpps/333Kpps) (ACM SOSP ’99) • RouteBricks (18.96Mpps/12(?)Mpps) (ACM SOSP ‘09) • PacketShader (58.4Mpps/50+(?)Mpps) (ACM SIGCOMM ’11) • NetMap (USENIX ATC ’12) • DIR-24-8-BASIC • Waldvogel
  • 3. • E. Kohler, R. Morris, B. Chen, J. Jannotti, and M. F. Kaashoek., “The Click modular router”, ACM Transactions on Computer Systems 18(3), August 2000, pages 263- 297. • SOSP ‘99, TOCS ’00 • Still active – Used in SEATTLE(SIGCOMM’08) – Used in RouteBricks(SOSP’09) – Last release: 2011/09/24 – Last commit: 2014/04/06 • Interrupt v.s. Polling (i.e., spin-wait, busy-wait) • Pentium III 700 MHz, 100Mbps Ether DEC tulip PCI, … • Linux 84kpps, Polling Linux 284kpps • Routing 333kpps, Forwarding 452kpps • Time taken: 2,798 ns/packet Click Modular Router (1)
  • 5. RouteBricks (1) • M. Dobrescu, N. Egi, K. Argyraki, B.-G. Chun, K. Fall, Gianluca Iannaccone, Allan Knies, Maziar Manesh, Sylvia Ratnasamy, “RouteBricks: Exploiting Parallelism To Scale Software Routers, ” 22nd ACM Symposium on Operating Systems Principles (SOSP), October 2009 • SOSP ’09 • Click-base • Leverages: – multi-cores. – multi-queue NICs. – batch processing. • NUMA-aware data placement didn’t make differences (in contrast to PacketShader). • Achieved 18.96Mpps Forwarding, 12(?)Mpps Routing • (Routing: 256K routes with memory expansion (DIR-24-8-BASIC))
  • 7. PacketShader (1) • S. Han, K. Jang, K. Park and S. Moon. “PacketShader: a GPU-accelerated Software Router.” In proceedings of ACM SIGCOMM 2010, Delhi, India. September 2010. • Speeds routing lookups (not just forwarding) • Forwarding: 41.1 Gbps or 58.4 Mpps • IPv4 Routing (DIR-24-8-BASIC, 282K routes): 39Gbps. • IPv6 Routing (Waldvogel, 200K routes): 38Gbps. • Up to 130us/200us (IPv4/IPv6) routing delay (260us/400us for roundtrip)
  • 8. PacketShader (2) • Not a pps-limitation ?
  • 9. PacketShader (3) (a) Cores contend for queues. Leads to expensive cache bouncing and lock contention. (b) Better coupling of queues and cores. pipeline batch pipeline • Buffer management • Pipelining • Batch processing • Decoupling among queues and cores
  • 10. NetMap (1) • Luigi Rizzo, “netmap: a novel framework for fast packet I/O,” Usenix ATC'12, Boston, June 2012 • Forwarding 14.88 Mpps (1 core 900 MHz) • Objective: to move packets quickly between the apps and the network cards. • Reduced copy, modified buffer management. • Protection. User-land v.s. Kernel. • 408 ns / batch. • Implementation on Linux/FreeBSD/Click.
  • 12. Techniques • Polling (i.e., spin-wait or busy-wait) • Parallelization – i.e., multi-queues, multi-cores, GPUs • Pipelining • Reduce overhead – Batch processing (amortized over a batch) – Remove copy, improve buffer managements
  • 14. DIR-24-8-BASIC • P. Gupta, S. Lin, and N. McKeown. “Routing Lookups in Hardware at Memory Access Speeds.” In Proceedings of the IEEE INFOCOM Conference, San Francisco, CA, USA, March 1998. • INFOCOM ‘98
  • 15. Waldvogel • M. Waldvogel, G. Varghese, J. Turner, B. Plattner. “Scalable High-Speed IP Routing Lookups.” Proc. ACM SIGCOMM 1997 , pp. 25-36, Cannes, France.