libvnf: building VNFs made easy
|
|
- Merryl Price
- 5 years ago
- Views:
Transcription
1 libvnf: building VNFs made easy Priyanka Naik, Akash Kanase, Trishal Patel, Mythili Vutukuru Dept. of Computer Science and Engineering Indian Institute of Technology, Bombay SoCC th October, 2018
2 NFV ecosystem Orchestrator Firewall VNF VNF VNF VNF Load balancer VM VM VM VM Router Hypervisor Network address translator NFV: Network Function Virtualization VNF: Virtual Network Function 2
3 NFV ecosystem Will they give good performance? Is it easy to build them? Orchestrator Firewall VNF VNF VNF VNF Load balancer VM VM VM VM Router Hypervisor Network address translator NFV: Network Function Virtualization VNF: Virtual Network Function 3
4 How to build VNF? VNF code developed by VNF developer 4
5 How to build VNF? VNF code developed by VNF developer 38% EPC code read/write packets CORD Intel EPC: 4
6 How to build VNF? VNF Processing logic VNF code developed by VNF developer VNF Framework 38% EPC code read/write packets CORD Intel EPC: 4
7 How to build VNF? VNF Processing logic netbricks VNF Framework StatelessNF OpenNF VPP 4
8 How to build VNF? VNF Processing logic netbricks VNF Framework StatelessNF OpenNF VPP What is missing in these frameworks? 4
9 What is required from VNF frameworks? " Requirement 1: Support for both L3 and Transport VNF " Requirement 2: Flexibility of network stack " Requirement 3: Support for distributed state management 5
10 What is required from VNF frameworks? " Requirement 1: Support for both L3 and Transport VNF " Requirement 2: Flexibility of network stack " Requirement 3: Support for distributed state management 6
11 Support for L3 and transport VNFs Layer 3 VNFs Network address translator Layer 3 Load balancer 7
12 Support for L3 and transport VNFs Layer 3 VNFs Network address translator Layer 3 Load balancer Header manipulations N/W layer Data link layer 7
13 Support for L3 and transport VNFs Layer 3 VNFs Network address translator Layer 3 Load balancer Header manipulations N/W layer Data link layer Frameworks: netbricks, YANFF 7
14 Support for L3 and transport VNFs Layer 3 VNFs Transport Layer VNFs vepc internet Network address translator Layer 3 Load balancer Header manipulations N/W layer Data link layer Frameworks: netbricks, YANFF 7
15 Support for L3 and transport VNFs Layer 3 VNFs Transport Layer VNFs vepc internet Network address translator Layer 3 Load balancer Header manipulations N/W layer Connection termination Request processing Transport Layer N/W layer Connection initiation Data link layer Data link layer Frameworks: netbricks, YANFF 7
16 Support for L3 and transport VNFs Layer 3 VNFs Transport Layer VNFs vepc internet Network address translator Layer 3 Load balancer Header manipulations N/W layer Connection termination Request processing Transport Layer N/W layer Connection initiation Data link layer Data link layer Frameworks: netbricks, YANFF Frameworks: mtcp, TLDK Netbricks: Taking the v out of nfv. In Proc. of OSDI 16 YANFF: mtcp: A highly scalable user-level tcp stack for multicore systems. In Proc. of NSDI 14 TLDK: 7
17 Support for L3 and transport VNFs Layer 3 VNFs Transport Layer VNFs vepc internet Network address translator Layer 3 Load balancer Header manipulations N/W layer Connection termination Request processing Transport Layer N/W layer Connection initiation Data link layer Data link layer Frameworks: netbricks, YANFF Frameworks: mtcp, TLDK Netbricks: Taking the v out of nfv. In Proc. of OSDI 16 YANFF: mtcp: A highly scalable user-level tcp stack for multicore systems. In Proc. of NSDI 14 TLDK: Are these frameworks enough? 7
18 Event driven I/O Existing transport-layer frameworks are event-driven 8
19 Event driven I/O Pros: Existing transport-layer frameworks are event-driven 8
20 Event driven I/O Existing transport-layer frameworks are event-driven Pros: " Efficient for multi-core scalability 8
21 Event driven I/O Existing transport-layer frameworks are event-driven Pros: " Efficient for multi-core scalability 8
22 Event driven I/O Existing transport-layer frameworks are event-driven Pros: " Efficient for multi-core scalability Cons: 8
23 Event driven I/O Existing transport-layer frameworks are event-driven Pros: " Efficient for multi-core scalability Cons: " Needs explicit request state storage 8
24 Need to maintain request state 1 2 A B C 4 3 9
25 Need to maintain request state 1 2 A B C 4 3 State at B to process A s request A s request C s reply Connection identifiers 9
26 Need to maintain request state VNF processing layer (abstraction?) 1 2 A B C 4 3 State at B to process A s request network stack (mtcp) (connection) A s request C s reply Connection identifiers DPDK and netmap layer (packet) 9
27 Need to maintain request state VNF processing layer (abstraction?) 1 2 A B C 4 3 State at B to process A s request network stack (mtcp) (connection) A s request C s reply Connection identifiers DPDK and netmap layer (packet) Existing frameworks do not provide this support 9
28 What is required from VNF frameworks? " Requirement 1: Support for both Layer 3 and Transport VNF " Requirement 2: Flexibility of network stack " Requirement 3: Support for distributed state management 10
29 Flexibility of network stack Kernel Stack Application VNF Kernel network stack vnic 11
30 Flexibility of network stack Kernel Stack Kernel Bypass Stack Application VNF Kernel network stack vnic Application VNF + userspace stack DPDK/netmap vnic 11
31 Flexibility of network stack Kernel Stack Kernel Bypass Stack Application VNF Kernel network stack vnic Application VNF + userspace stack DPDK/netmap vnic Easy switch between stacks 11
32 What is required from VNF frameworks? " Requirement 1: Support for both L3 and Transport VNF " Requirement 2: Flexibility of network stack " Requirement 3: Support for distributed state management 12
33 Support for distributed state management Orchestrator VNF 1 VNF 2 VNF 2 VNF 3 VM VM VM VM Hypervisor 13
34 Support for distributed state management VNF 2 VNF 2 State Synchronization 13
35 Support for distributed state management VNF 2 VNF 2 State Synchronization State Migration opennf, split/merge 13
36 Support for distributed state management VNF 2 Data Store VNF 2 State Synchronization Remote store statelessnf State Migration opennf, split/merge Stateless network functions: Breaking the tight coupling of state and processing. In Proc. of NSDI 17 Split/merge: System support for elastic execution in virtual middleboxes. In Proc. of NSDI 13 Opennf: Enabling innovation in network function control. In Proc. of SIGCOMM 14 13
37 Support for distributed state management VNF 2 Data Store VNF 2 State Synchronization Remote store statelessnf State Migration opennf, split/merge None of above support transport layer VNFs Stateless network functions: Breaking the tight coupling of state and processing. In Proc. of NSDI 17 Split/merge: System support for elastic execution in virtual middleboxes. In Proc. of NSDI 13 Opennf: Enabling innovation in network function control. In Proc. of SIGCOMM 14 13
38 Summary of VNF Frameworks Requirement/ Framework netbricks Flick StatelessNF Split-Merge/ OpenNF libvnf Layer 3 + Applayer support Flexibility of network stack Distributed State Management no yes no no yes no no no no yes no no yes yes yes Netbricks: Taking the v out of nfv. In Proc. of OSDI 16 Flick: Developing and running application-specific network services. In Proc. of USENIX ATC 16 Stateless network functions: Breaking the tight coupling of state and processing. In Proc. of NSDI 17 Split/merge: System support for elastic execution in virtual middleboxes. In Proc. of NSDI 13 Opennf: Enabling innovation in network function control. In Proc. of SIGCOMM 14 14
39 libvnf Design Goals VNF processing logic Handled by VNF developer R3 R2 Distributed State Management Support for network and transport layer VNF Handled by libvnf R1 Flexibility of network stack R: Requirement 15
40 libvnf overview VNF code API Calls libvnf API 16
41 libvnf overview VNF code API Calls libvnf API mtcp+ netmap/dpdk initialization Kernel stack initialization Stack initialization 16
42 libvnf overview VNF code API Calls libvnf API mtcp+ netmap/dpdk initialization Kernel stack initialization Stack initialization Per-core threads 16
43 libvnf overview VNF code API Calls libvnf API mtcp+ netmap/dpdk initialization Kernel stack initialization Stack initialization Per-core threads Per-core data structures Lock-free Cache optimized 16
44 libvnf API libvnf API Communication Request state State Management 17
45 libvnf API libvnf API Request state State Management 18
46 libvnf API libvnf API Communication Request state State Management 18
47 Communication API VNF code libvnf API Per-core data structures Per-core packet pool Pre-allocated memory pools (Per-core packet pools) 19
48 Communication API VNF code registercallback(socket, fn) Store mapping libvnf API Per-core data structures Per-core packet pool Pre-allocated memory pools (Per-core packet pools) 19
49 Communication API VNF code registercallback(socket, fn) Store mapping libvnf API Per-core data structures fn(packet) Packet arrives on socket Per-core packet pool Pre-allocated memory pools (Per-core packet pools) 19
50 Communication API VNF code getpktbuf libvnf API Per-core data structures Per-core packet pool Pre-allocated memory pools (Per-core packet pools) 19
51 Communication API VNF code getpktbuf Buffer to write packet libvnf API Per-core data structures Per-core packet pool Pre-allocated memory pools (Per-core packet pools) 19
52 libvnf API VNF Design Requirements Communication State Management 20
53 libvnf API VNF Design Requirements Communication Request state State Management 20
54 Need for request state VNF (abstraction processing?) layer (abstraction?) 1 2 A B C 4 3 State at B to process A s request A s request C s reply Connection identifiers network stack (mtcp) (connection) DPDK and netmap layer (packet) 21
55 Need for request state VNF (abstraction processing?) layer REQUEST OBJECT 1 2 A B C 4 3 State at B to process A s request A s request C s reply Connection identifiers network stack (mtcp) (connection) DPDK and netmap layer (packet) Request object 22
56 Request Object API 1 2 A B C 4 3 libvnf API Per-core request pool 23
57 Request Object API 1 2 A B C 4 3 allocreqobj(a connection_id) Allocate request object block for A s request libvnf API Per-core request pool A s request C s reply Connection identifiers 23
58 Request Object API 1 2 A B C 4 3 allocreqobj(a connection_id) Allocate request object block for A s request libvnf API Per-core request pool A s request Per-core packet pool C s reply Connection identifiers 23
59 Request Object API 1 2 A B C 4 3 linkreqobj(c connection_id) Link to the existing A request object libvnf API Per-core request pool A s request Per-core packet pool C s reply Connection identifiers 23
60 libvnf API VNF Design Requirements Communication Request state 24
61 libvnf API VNF Design Requirements Communication Request state State Management 24
62 State Management API VNF code libvnf API 25
63 State Management API VNF code libvnf API Local data store pool 25
64 State Management API VNF code setdata( LOCAL ) libvnf API Store in local datastore Local data store pool 25
65 State Management API VNF code setdata( LOCAL ) libvnf API Store in local datastore Local data store pool libvnf data store wrapper Redis KV store Remote Data store 25
66 State Management API VNF code setdata( REMOTE ) libvnf API Cache locally Local data store pool Store in remote data store libvnf data store wrapper Redis KV store Remote Data store 25
67 Evaluation " Overhead of libvnf " Scalability with cores " Benefits of libvnf 26
68 Setup 1 2 A B C
69 Setup 1 2 A B C 4 3 VNF A VNF C S/W switch (on kernel) 27
70 Setup 1 2 A B C 4 3 VNF A VNF C VNF B S/W switch (on kernel) S/W switch (like netmap-vale) NIC Queue Physical NIC 27
71 Setup 1 2 A B C 4 3 VNF A VNF C VNF B S/W switch (on kernel) S/W switch (like netmap-vale) NIC Queue VNF A, C: 4 core, 4GB RAM VNF B: 4 GB RAM, cores varied Physical NIC 27
72 Evaluation " Overhead of libvnf " Scalability with cores " Benefits of libvnf 28
73 Overhead check 29
74 Overhead check <5% overhead of libvnf DPDK~ netmap performance 29
75 Evaluation " Overhead of libvnf " Scalability with cores " Benefits of libvnf 30
76 Core scalability 31
77 Core scalability scales linearly with cores 31
78 Evaluation " Overhead of libvnf " Scalability with cores " Benefits of libvnf 32
79 Building VNFs VNF Performance Overhead of libvnf LoC Saved IMS (IP Multimedia Subsystem) EPC (LTE-Evolved Packet Core ) 3.4% 42% 5.5% 38% Layer 3 Load Balancer 14% 52% 33
80 Building VNFs VNF Performance Overhead of libvnf LoC Saved IMS (IP Multimedia Subsystem) EPC (LTE-Evolved Packet Core ) 3.4% 42% 5.5% 38% Layer 3 Load Balancer 14% 52% Low overhead in app-layer VNF Higher overhead in L3 VNF 33
81 Summary " Library to ease building of VNFs " Expressive to build L3 and App-layer VNF " Supports multiple network stacks " Low performance overhead 34
82 Thank You 35
83 Setup 1 2 A B C 4 3 VNF A VNF C LB VNF VNF B VNF B Data store VM S/W switch (on kernel) S/W switch (vale on netmap) NIC Queue Physical NIC 36
84 Setup 1 2 A B C 4 3 VNF A VNF C LB VNF VNF B VNF B Data store VM S/W switch (on kernel) S/W switch (vale on netmap) VNF A, C: 4 core, 4GB RAM VNF B: 4 GB RAM, cores varied Data Store VM: 6 core, 16GB RAM LB: 1 core, 4GB RAM NIC Queue Physical NIC 36
Stateless Network Functions:
Stateless Network Functions: Breaking the Tight Coupling of State and Processing Murad Kablan, Azzam Alsudais, Eric Keller, Franck Le University of Colorado IBM Networks Need Network Functions Firewall
More informationPASTE: A Network Programming Interface for Non-Volatile Main Memory
PASTE: A Network Programming Interface for Non-Volatile Main Memory Michio Honda (NEC Laboratories Europe) Giuseppe Lettieri (Università di Pisa) Lars Eggert and Douglas Santry (NetApp) USENIX NSDI 2018
More informationSoftware Routers: NetMap
Software Routers: NetMap Hakim Weatherspoon Assistant Professor, Dept of Computer Science CS 5413: High Performance Systems and Networking October 8, 2014 Slides from the NetMap: A Novel Framework for
More informationIX: A Protected Dataplane Operating System for High Throughput and Low Latency
IX: A Protected Dataplane Operating System for High Throughput and Low Latency Adam Belay et al. Proc. of the 11th USENIX Symp. on OSDI, pp. 49-65, 2014. Presented by Han Zhang & Zaina Hamid Challenges
More informationEmpower Diverse Open Transport Layer Protocols in Cloud Networking GEORGE ZHAO DIRECTOR OSS & ECOSYSTEM, HUAWEI
Empower Diverse Open Transport Layer Protocols in Cloud Networking GEORGE ZHAO DIRECTOR OSS & ECOSYSTEM, HUAWEI Agenda FD.io Introduction Challenges in Container & Cloud Native Apps Proposed Solutions
More informationDesign and Implementation of Virtual TAP for Software-Defined Networks
Design and Implementation of Virtual TAP for Software-Defined Networks - Master Thesis Defense - Seyeon Jeong Supervisor: Prof. James Won-Ki Hong Dept. of CSE, DPNM Lab., POSTECH, Korea jsy0906@postech.ac.kr
More informationEnabling Fast, Dynamic Network Processing with ClickOS
Enabling Fast, Dynamic Network Processing with ClickOS Joao Martins*, Mohamed Ahmed*, Costin Raiciu, Roberto Bifulco*, Vladimir Olteanu, Michio Honda*, Felipe Huici* * NEC Labs Europe, Heidelberg, Germany
More informationProgrammable NICs. Lecture 14, Computer Networks (198:552)
Programmable NICs Lecture 14, Computer Networks (198:552) Network Interface Cards (NICs) The physical interface between a machine and the wire Life of a transmitted packet Userspace application NIC Transport
More informationAgenda. Introduction Network functions virtualization (NFV) promise and mission cloud native approach Where do we want to go with NFV?
August, 2018 Agenda Introduction Network functions virtualization (NFV) promise and mission cloud native approach Where do we want to go with NFV? 2 Miroslaw Walukiewicz I m from Gdansk, Poland. 25 years
More informationSpeeding up Linux TCP/IP with a Fast Packet I/O Framework
Speeding up Linux TCP/IP with a Fast Packet I/O Framework Michio Honda Advanced Technology Group, NetApp michio@netapp.com With acknowledge to Kenichi Yasukata, Douglas Santry and Lars Eggert 1 Motivation
More informationPerformance Considerations of Network Functions Virtualization using Containers
Performance Considerations of Network Functions Virtualization using Containers Jason Anderson, et al. (Clemson University) 2016 International Conference on Computing, Networking and Communications, Internet
More informationIsoStack Highly Efficient Network Processing on Dedicated Cores
IsoStack Highly Efficient Network Processing on Dedicated Cores Leah Shalev Eran Borovik, Julian Satran, Muli Ben-Yehuda Outline Motivation IsoStack architecture Prototype TCP/IP over 10GE on a single
More informationPDP : A Flexible and Programmable Data Plane. Massimo Gallo et al.
PDP : A Flexible and Programmable Data Plane Massimo Gallo et al. Introduction Network Function evolution L7 Load Balancer TLS/SSL Server Proxy Server Firewall Introduction Network Function evolution Can
More informationDPDK Summit China 2017
DPDK Summit China 2017 2 Practice of Network Monitoring and Security Technologies in Cloud Data Center Kai, Wang YunShan Networks Data center is evolving to be cloud based and software defined The monitoring
More informationAchieve Low Latency NFV with Openstack*
Achieve Low Latency NFV with Openstack* Yunhong Jiang Yunhong.Jiang@intel.com *Other names and brands may be claimed as the property of others. Agenda NFV and network latency Why network latency on NFV
More informationData Path acceleration techniques in a NFV world
Data Path acceleration techniques in a NFV world Mohanraj Venkatachalam, Purnendu Ghosh Abstract NFV is a revolutionary approach offering greater flexibility and scalability in the deployment of virtual
More informationLITE Kernel RDMA. Support for Datacenter Applications. Shin-Yeh Tsai, Yiying Zhang
LITE Kernel RDMA Support for Datacenter Applications Shin-Yeh Tsai, Yiying Zhang Time 2 Berkeley Socket Userspace Kernel Hardware Time 1983 2 Berkeley Socket TCP Offload engine Arrakis & mtcp IX Userspace
More informationAccelerating 4G Network Performance
WHITE PAPER Accelerating 4G Network Performance OFFLOADING VIRTUALIZED EPC TRAFFIC ON AN OVS-ENABLED NETRONOME SMARTNIC NETRONOME AGILIO SMARTNICS PROVIDE A 5X INCREASE IN vepc BANDWIDTH ON THE SAME NUMBER
More informationAgenda How DPDK can be used for your Application DPDK Ecosystem boosting your Development Meet the Community Challenges
SPEED MATTERS. All rights reserved. All brand names, trademarks and copyright information cited in this presentation shall remain the property of its registered owners. Agenda How DPDK can be used for
More informationPerformance Comparison of State Synchronization Techniques in a Distributed LTE EPC
Performance Comparison of State Synchronization Techniques in a Distributed LTE EPC Pratik Satapathy, Jash Dave, Priyanka Naik, Mythili Vutukuru Department of Computer Science and Engineering, Indian Institute
More informationBuilding a Platform Optimized for the Network Edge
Building a Platform Optimized for the Network Edge MPLS + SDN + NFV WORLD 2018 Nicolas Bouthors, Enea Innovation Agenda Software Virtualization - Key Requirements Leveraging DPDK Multi-Function VNFs at
More informationAccelerating Telco NFV Deployments with DPDK and SmartNICs
x Accelerating Telco NFV Deployments with and SmartNICs Kalimani Venkatesan G, Aricent Kalimani.Venkatesan@aricent.com Barak Perlman, Ethernity Networks Barak@Ethernitynet.com Summit North America 2018
More informationContainer Adoption for NFV Challenges & Opportunities. Sriram Natarajan, T-Labs Silicon Valley Innovation Center
Container Adoption for NFV Challenges & Opportunities Sriram Natarajan, T-Labs Silicon Valley Innovation Center Virtual Machine vs. Container Stack KVM Container-stack Libraries Guest-OS Hypervisor Libraries
More informationSoftRDMA: Rekindling High Performance Software RDMA over Commodity Ethernet
SoftRDMA: Rekindling High Performance Software RDMA over Commodity Ethernet Mao Miao, Fengyuan Ren, Xiaohui Luo, Jing Xie, Qingkai Meng, Wenxue Cheng Dept. of Computer Science and Technology, Tsinghua
More informationLight & NOS. Dan Li Tsinghua University
Light & NOS Dan Li Tsinghua University Performance gain The Power of DPDK As claimed: 80 CPU cycles per packet Significant gain compared with Kernel! What we care more How to leverage the performance gain
More informationIX: A Protected Dataplane Operating System for High Throughput and Low Latency
IX: A Protected Dataplane Operating System for High Throughput and Low Latency Belay, A. et al. Proc. of the 11th USENIX Symp. on OSDI, pp. 49-65, 2014. Reviewed by Chun-Yu and Xinghao Li Summary In this
More informationA Look at Intel s Dataplane Development Kit
A Look at Intel s Dataplane Development Kit Dominik Scholz Chair for Network Architectures and Services Department for Computer Science Technische Universität München June 13, 2014 Dominik Scholz: A Look
More informationNEC Virtualized Evolved Packet Core vepc
TE-524262 NEC Virtualized Evolved Packet Core vepc Design Concepts and Benefits INDEX Leading the transformation into Mobile Packet Core Virtualization P.3 vepc System Architecture Overview P.4 Elastic
More informationSurvey of ETSI NFV standardization documents BY ABHISHEK GUPTA FRIDAY GROUP MEETING FEBRUARY 26, 2016
Survey of ETSI NFV standardization documents BY ABHISHEK GUPTA FRIDAY GROUP MEETING FEBRUARY 26, 2016 VNFaaS (Virtual Network Function as a Service) In our present work, we consider the VNFaaS use-case
More informationTLDK Overview. Transport Layer Development Kit Ray Kinsella February ray.kinsella [at] intel.com IRC: mortderire
TLDK Overview Transport Layer Development Kit Ray Kinsella February 2017 Email : ray.kinsella [at] intel.com IRC: mortderire Contributions from Keith Wiles & Konstantin Ananyev Legal Disclaimer General
More informationIntroduction of the SONATA project / Distributed VNF State Management
Introduction of the SONATA project / Distributed VNF State Management Manuel Peuster, Sevil Mehraghdam, Holger Karl 27.10.2015 Computer Networks Group Universität Paderborn Agenda Marketing: The 5G-PPP
More informationFLICK: Developing and Running Application-Specific Network Services
FLICK: Developing and Running Application-Specific Network Services Presenter: Richard G. Clegg, Imperial College Imperial College: Abdul Alim, Luo Mai, Lukas Rupprecht, Eric Seckler, Paolo Costa, Peter
More informationAdvanced Computer Networks. End Host Optimization
Oriana Riva, Department of Computer Science ETH Zürich 263 3501 00 End Host Optimization Patrick Stuedi Spring Semester 2017 1 Today End-host optimizations: NUMA-aware networking Kernel-bypass Remote Direct
More informationNFV Infrastructure for Media Data Center Applications
NFV Infrastructure for Media Data Center Applications Today s Presenters Roger Sherwood Global Strategy & Business Development, Cisco Systems Damion Desai Account Manager for Datacenter, SDN, NFV and Mobility,
More informationC/U Separated and Modular Architecture for NFV. Dr. Mo li Chief Architect of CTO Group, ZTE Corporation
C/U Separated and Modular Architecture for NFV Dr. Mo li Chief Architect of CTO Group, ZTE Corporation C/U Separated Requirement for Multi-tier DC Scenario enterprise Different processing character Different
More informationA High Performance Packet Core for Next Generation Cellular Networks
A High Performance Packet Core for Next Generation Cellular Networks Zafar Qazi + Melvin Walls, Aurojit Panda +, Vyas Sekar, Sylvia Ratnasamy +, Scott Shenker + + 1 Explosive Cellular Growth Many Diverse
More informationArrakis: The Operating System is the Control Plane
Arrakis: The Operating System is the Control Plane Simon Peter, Jialin Li, Irene Zhang, Dan Ports, Doug Woos, Arvind Krishnamurthy, Tom Anderson University of Washington Timothy Roscoe ETH Zurich Building
More informationVNF Chain Allocation and Management at Data Center Scale
VNF Chain Allocation and Management at Data Center Scale Internet Cloud Provider Tenants Nodir Kodirov, Sam Bayless, Fabian Ruffy, Ivan Beschastnikh, Holger Hoos, Alan Hu Network Functions (NF) are useful
More informationThe latency of user-to-user, kernel-to-kernel and interrupt-to-interrupt level communication
The latency of user-to-user, kernel-to-kernel and interrupt-to-interrupt level communication John Markus Bjørndalen, Otto J. Anshus, Brian Vinter, Tore Larsen Department of Computer Science University
More informationEnabling innovation in the Internet: Main Achievements of the CHANGE Project. Felipe Huici, NEC Europe
Enabling innovation in the Internet: Main Achievements of the CHANGE Project Felipe Huici, NEC Europe EU FP7 CHANGE Project Info: Start in Oct. 2010, will end in December 2013 Partner Eurescom (Prime Contractor)
More informationEvolution of the netmap architecture
L < > T H local Evolution of the netmap architecture Evolution of the netmap architecture -- Page 1/21 Evolution of the netmap architecture Luigi Rizzo, Università di Pisa http://info.iet.unipi.it/~luigi/vale/
More informationMaking Network Functions Software-Defined
Making Network Functions Software-Defined Yotam Harchol VMware Research / The Hebrew University of Jerusalem Joint work with Anat Bremler-Barr and David Hay Appeared in ACM SIGCOMM 2016 THE HEBREW UNIVERSITY
More informationElastic Scaling of Stateful Network Functions
NSDI 2018 Elastic Scaling of Stateful Network Functions Shinae Woo *+, Justine Sherry *, Sangjin Han *, Sue Moon +, Sylvia Ratnasamy *, Scott Shenker * + KAIST, * UC Berkeley Elastic Scaling of NFs NFV
More informationNetBricks: Taking the V out of NFV. Aurojit Panda, Sangjin Han, Keon Jang, Melvin Walls, Sylvia Ratnasamy, Scott Shenker UC Berkeley, Google, ICSI
NetBricks: Taking the V out of NFV Aurojit Panda, Sangjin Han, Keon Jang, Melvin Walls, Sylvia Ratnasamy, Scott Shenker UC Berkeley, Google, ICSI What the heck is NFV? A Short Introduction to NFV A Short
More informationDisaggregation and Virtualization within the Juniper Networks Mobile Cloud Architecture. White Paper
Disaggregation and Virtualization within the Juniper Networks Mobile Cloud Architecture White Paper June 2017 Juniper Networks, Inc. 1133 Innovation Way Sunnyvale, California 94089 USA 408-745-2000 www.juniper.net
More informationIntroduction of ASTRI s Network Functions Virtualization (NFV) Technologies. ASTRI Proprietary
Introduction of ASTRI s Network Functions Virtualization (NFV) Technologies ASTRI LTE NFV Technology Demo on Intel ONP Virtual Network Functions Provides LTE Core Network Functions (VNF) SGW MME PGW LTE
More informationFlexNIC: Rethinking Network DMA
FlexNIC: Rethinking Network DMA Antoine Kaufmann Simon Peter Tom Anderson Arvind Krishnamurthy University of Washington HotOS 2015 Networks: Fast and Growing Faster 1 T 400 GbE Ethernet Bandwidth [bits/s]
More informationKernel Bypass. Sujay Jayakar (dsj36) 11/17/2016
Kernel Bypass Sujay Jayakar (dsj36) 11/17/2016 Kernel Bypass Background Why networking? Status quo: Linux Papers Arrakis: The Operating System is the Control Plane. Simon Peter, Jialin Li, Irene Zhang,
More informationSmartNIC Programming Models
SmartNIC Programming Models Johann Tönsing 206--09 206 Open-NFP Agenda SmartNIC hardware Pre-programmed vs. custom (C and/or P4) firmware Programming models / offload models Switching on NIC, with SR-IOV
More informationSmartNIC Programming Models
SmartNIC Programming Models Johann Tönsing 207-06-07 207 Open-NFP Agenda SmartNIC hardware Pre-programmed vs. custom (C and/or P4) firmware Programming models / offload models Switching on NIC, with SR-IOV
More informationLight: A Scalable, High-performance and Fully-compatible User-level TCP Stack. Dan Li ( 李丹 ) Tsinghua University
Light: A Scalable, High-performance and Fully-compatible User-level TCP Stack Dan Li ( 李丹 ) Tsinghua University Data Center Network Performance Hardware Capability of Modern Servers Multi-core CPU Kernel
More informationDemystifying Network Cards
Demystifying Network Cards Paul Emmerich December 27, 2017 Chair of Network Architectures and Services About me PhD student at Researching performance of software packet processing systems Mostly working
More informationResearch on DPDK Based High-Speed Network Traffic Analysis. Zihao Wang Network & Information Center Shanghai Jiao Tong University
Research on DPDK Based High-Speed Network Traffic Analysis Zihao Wang Network & Information Center Shanghai Jiao Tong University Outline 1 Background 2 Overview 3 DPDK Based Traffic Analysis 4 Experiment
More informationG-NET: Effective GPU Sharing In NFV Systems
G-NET: Effective Sharing In NFV Systems Kai Zhang*, Bingsheng He^, Jiayu Hu #, Zeke Wang^, Bei Hua #, Jiayi Meng #, Lishan Yang # *Fudan University ^National University of Singapore #University of Science
More informationNFVnice: Dynamic Backpressure and Scheduling for NFV Service Chains
NFVnice: Dynamic Backpressure and Scheduling for NFV Service Chains Sameer G Kulkarni 1, Wei Zhang 2, Jinho Hwang 3, Shriram Rajagopalan 3, K.K. Ramakrishnan 4, Timothy Wood 2, Mayutan Arumaithurai 1 &
More informationTLDK Overview. Transport Layer Development Kit Keith Wiles April Contributions from Ray Kinsella & Konstantin Ananyev
TLDK Overview Transport Layer Development Kit Keith Wiles April 2017 Contributions from Ray Kinsella & Konstantin Ananyev Notices and Disclaimers Intel technologies features and benefits depend on system
More informationMuch Faster Networking
Much Faster Networking David Riddoch driddoch@solarflare.com Copyright 2016 Solarflare Communications, Inc. All rights reserved. What is kernel bypass? The standard receive path The standard receive path
More informationHow DPI enables effective deployment of CloudNFV. David Le Goff / Director, Strategic & Product Marketing March 2014
How DPI enables effective deployment of CloudNFV David Le Goff / Director, Strategic & Product Marketing March 2014 Key messages of this presentation 1. DPI (Deep Packet Inspection) is critical for effective
More informationWindows Server System Center Azure Pack
Windows Server System Center Azure Pack Tenant Deployment Multi-Tier LOB Application Tenant Deployment Multi-Tier LOB Application Inbox feature for integrated management of IP addresses, domain
More informationGot Loss? Get zovn! Daniel Crisan, Robert Birke, Gilles Cressier, Cyriel Minkenberg, and Mitch Gusat. ACM SIGCOMM 2013, August, Hong Kong, China
Got Loss? Get zovn! Daniel Crisan, Robert Birke, Gilles Cressier, Cyriel Minkenberg, and Mitch Gusat ACM SIGCOMM 2013, 12-16 August, Hong Kong, China Virtualized Server 1 Application Performance in Virtualized
More informationDataplane Networking journey in Containers
Dataplane Networking journey in Gary Loughnane gary.loughnane@intel.com Kuralamudhan Ramakrishnan kuralamudhan.ramakrishnan@intel.com DPDK Summit Userspace - Dublin- 2017 Discussion topics Container Deployment
More informationExploring mtcp based Single-Threaded and Multi-Threaded Web Server Design
Exploring mtcp based Single-Threaded and Multi-Threaded Web Server Design A Thesis Submitted in partial fulfillment of the requirements for the degree of Master of Technology by Pijush Chakraborty (153050015)
More informationIntroduction. Delivering Management as Agile as the Cloud: Enabling New Architectures with CA Technologies Virtual Network Assurance Solution
Delivering Management as Agile as the Cloud: Enabling New Architectures with CA Technologies Virtual Network Assurance Solution Introduction Service providers and IT departments of every type are seeking
More informationNetwork Function Virtualization over Open DC/OS Yung-Han Chen
Network Function Virtualization over Open DC/OS Yung-Han Chen 2016.05.18 1 Outlines Network Function Virtualization (NFV) Framework Container-based Open Source Solutions for NFV Use Cases 2 NFV Architectural
More informationStateless Network Functions: Breaking the Tight Coupling of State and Processing
Stateless Network Functions: Breaking the Tight Coupling of State and Processing Murad Kablan, Azzam Alsudais, and Eric Keller, University of Colorado Boulder; Franck Le, IBM Research https://www.usenix.org/conference/nsdi17/technical-sessions/presentation/kablan
More informationPASTE: Fast End System Networking with netmap
PASTE: Fast End System Networking with netmap Michio Honda, Giuseppe Lettieri, Lars Eggert and Douglas Santry BSDCan 2018 Contact: @michioh, micchie@sfc.wide.ad.jp Code: https://github.com/micchie/netmap/tree/stack
More informationBuilding high performance network functions in VPP. Ole Trøan, VPP contributor FOSDEM 2018
Building high performance network functions in VPP Ole Trøan, ot@cisco.com, VPP contributor FOSDEM 2018 1 2 This talk? Goal: Make you into VPP developers Agenda: VPP architecture An example decomposed
More informationAusgewählte Betriebssysteme - Mark Russinovich & David Solomon (used with permission of authors)
Outline Windows 2000 - The I/O Structure Ausgewählte Betriebssysteme Institut Betriebssysteme Fakultät Informatik Components of I/O System Plug n Play Management Power Management I/O Data Structures File
More informationVirtualization, Xen and Denali
Virtualization, Xen and Denali Susmit Shannigrahi November 9, 2011 Susmit Shannigrahi () Virtualization, Xen and Denali November 9, 2011 1 / 70 Introduction Virtualization is the technology to allow two
More informationSafeBricks: Shielding Network Functions in the Cloud
SafeBricks: Shielding Network Functions in the Cloud Rishabh Poddar, Chang Lan, Raluca Ada Popa, Sylvia Ratnasamy UC Berkeley Network Functions (NFs) in the cloud Clients 2 Enterprise Destination Network
More informationDPDK Summit 2016 OpenContrail vrouter / DPDK Architecture. Raja Sivaramakrishnan, Distinguished Engineer Aniket Daptari, Sr.
DPDK Summit 2016 OpenContrail vrouter / DPDK Architecture Raja Sivaramakrishnan, Distinguished Engineer Aniket Daptari, Sr. Product Manager CONTRAIL (MULTI-VENDOR) ARCHITECTURE ORCHESTRATOR Interoperates
More informationCisco Virtualized Infrastructure Manager
DEVNET-2570 Virtualized Infrastructure Manager Suhail Syed, Product Manager Vamsi Krihsna Kuppur, Product Manager Spark How Questions? Use Spark to communicate with the speaker after the session 1. Find
More informationNokia Virtualized Mobility Manager
Release 9.1 The (VMM) performs the 3GPP mobility and session management control functions in the evolved packet core (EPC) network. The VMM delivers the deployment flexibility, reliability, high performance
More informationNetworking at the Speed of Light
Networking at the Speed of Light Dror Goldenberg VP Software Architecture MaRS Workshop April 2017 Cloud The Software Defined Data Center Resource virtualization Efficient services VM, Containers uservices
More informationImprove VNF safety with Vhost-User/DPDK IOMMU support
Improve VNF safety with Vhost-User/DPDK IOMMU support No UIO anymore! Maxime Coquelin Software Engineer KVM Forum 2017 AGENDA Background Vhost-user device IOTLB implementation Benchmarks Future improvements
More informationDeveloping Stateful Middleboxes with the mos API KYOUNGSOO PARK & YOUNGGYOUN MOON
Developing Stateful Middleboxes with the mos API KYOUNGSOO PARK & YOUNGGYOUN MOON ASIM JAMSHED, DONGHWI KIM, & DONGSU HAN SCHOOL OF ELECTRICAL ENGINEERING, KAIST Network Middlebox Networking devices that
More informationPASTE: A Networking API for Non-Volatile Main Memory
PASTE: A Networking API for Non-Volatile Main Memory Michio Honda (NEC Laboratories Europe) Lars Eggert (NetApp) Douglas Santry (NetApp) TSVAREA@IETF 99, Prague May 22th 2017 More details at our HotNets
More informationEnd to End SLA for Enterprise Multi-Tenant Applications
End to End SLA for Enterprise Multi-Tenant Applications Girish Moodalbail, Principal Engineer, Oracle Inc. Venugopal Iyer, Principal Engineer, Oracle Inc. The following is intended to outline our general
More informationImprove Performance of Kube-proxy and GTP-U using VPP
Improve Performance of Kube-proxy and GTP-U using VPP Hongjun Ni (hongjun.ni@intel.com) Danny Zhou (danny.zhou@intel.com) Johnson Li (johnson.li@intel.com) Network Platform Group, DCG, Intel Acknowledgement:
More informationRDMA and Hardware Support
RDMA and Hardware Support SIGCOMM Topic Preview 2018 Yibo Zhu Microsoft Research 1 The (Traditional) Journey of Data How app developers see the network Under the hood This architecture had been working
More informationNetwork Virtualization in Multi-tenant Datacenters
Network Virtualization in Multi-tenant Datacenters Teemu Koponen, Keith Amidon, Peter Balland, Martín Casado, Anupam Chanda, Bryan Fulton, Igor Ganichev, Jesse Gross, Natasha Gude, Paul Ingram, Ethan Jackson,
More informationThe Power of Batching in the Click Modular Router
The Power of Batching in the Click Modular Router Joongi Kim, Seonggu Huh, Keon Jang, * KyoungSoo Park, Sue Moon Computer Science Dept., KAIST Microsoft Research Cambridge, UK * Electrical Engineering
More informationData Centers and Cloud Computing
Data Centers and Cloud Computing CS677 Guest Lecture Tim Wood 1 Data Centers Large server and storage farms 1000s of servers Many TBs or PBs of data Used by Enterprises for server applications Internet
More informationData Centers and Cloud Computing. Slides courtesy of Tim Wood
Data Centers and Cloud Computing Slides courtesy of Tim Wood 1 Data Centers Large server and storage farms 1000s of servers Many TBs or PBs of data Used by Enterprises for server applications Internet
More informationPocket: Elastic Ephemeral Storage for Serverless Analytics
Pocket: Elastic Ephemeral Storage for Serverless Analytics Ana Klimovic*, Yawen Wang*, Patrick Stuedi +, Animesh Trivedi +, Jonas Pfefferle +, Christos Kozyrakis* *Stanford University, + IBM Research 1
More informationEnabling Efficient and Scalable Zero-Trust Security
WHITE PAPER Enabling Efficient and Scalable Zero-Trust Security FOR CLOUD DATA CENTERS WITH AGILIO SMARTNICS THE NEED FOR ZERO-TRUST SECURITY The rapid evolution of cloud-based data centers to support
More informationData Centers and Cloud Computing. Data Centers
Data Centers and Cloud Computing Slides courtesy of Tim Wood 1 Data Centers Large server and storage farms 1000s of servers Many TBs or PBs of data Used by Enterprises for server applications Internet
More informationETSI FUTURE Network SDN and NFV for Carriers MP Odini HP CMS CT Office April 2013
ETSI FUTURE Network SDN and NFV for Carriers MP Odini HP CMS CT Office April 2013 Challenges and Opportunities Reduce Capex/Opex Challenges Innovation at devices and OTT side Number of devices explode
More informationWIND RIVER TITANIUM CLOUD FOR TELECOMMUNICATIONS
WIND RIVER TITANIUM CLOUD FOR TELECOMMUNICATIONS Carrier networks are undergoing their biggest transformation since the beginning of the Internet. The ability to get to market quickly and to respond to
More informationExecutive Summary. Introduction. Test Highlights
Executive Summary Today, LTE mobile operators typically deploy All-IP and flat network architectures. This elegant and flexible solution requires deployment of an adequate security infrastructure. One
More informationDPDK Summit China 2017
Summit China 2017 Embedded Network Architecture Optimization Based on Lin Hao T1 Networks Agenda Our History What is an embedded network device Challenge to us Requirements for device today Our solution
More informationKVM as The NFV Hypervisor
KVM as The NFV Hypervisor Jun Nakajima Contributors: Mesut Ergin, Yunhong Jiang, Krishna Murthy, James Tsai, Wei Wang, Huawei Xie, Yang Zhang 1 Legal Disclaimer INFORMATION IN THIS DOCUMENT IS PROVIDED
More informationEnterprise Network Compute System (ENCS)
Enterprise Network Compute System (ENCS) Cisco vbranch Architecture Per Jensen, per@cisco.com Sept 2017 Agenda: Tech Update september-2017 1. ENCS update + demo v/per 2. Viptela update + demo v/dr Søren
More informationLatest Developments with NVMe/TCP Sagi Grimberg Lightbits Labs
Latest Developments with NVMe/TCP Sagi Grimberg Lightbits Labs 2018 Storage Developer Conference. Insert Your Company Name. All Rights Reserved. 1 NVMe-oF - Short Recap Early 2014: Initial NVMe/RDMA pre-standard
More informationFast packet processing in the cloud. Dániel Géhberger Ericsson Research
Fast packet processing in the cloud Dániel Géhberger Ericsson Research Outline Motivation Service chains Hardware related topics, acceleration Virtualization basics Software performance and acceleration
More informationProgrammable Software Switches. Lecture 11, Computer Networks (198:552)
Programmable Software Switches Lecture 11, Computer Networks (198:552) Software-Defined Network (SDN) Centralized control plane Data plane Data plane Data plane Data plane Why software switching? Early
More informationOverview of the Juniper Mobile Cloud Architecture Laying the Foundation for a Next-gen Secure Distributed Telco Cloud. Mobile World Congress 2017
Overview of the Juniper Mobile Cloud Laying the Foundation for a Next-gen Secure Telco Cloud Mobile World Congress 2017 Summary Mobile Challenges and Trends through 2020 Business Drivers Enable new revenue
More informationTOWARDS FAST IP FORWARDING
TOWARDS FAST IP FORWARDING IP FORWARDING PERFORMANCE IMPROVEMENT AND MEASUREMENT IN FREEBSD Nanako Momiyama Keio University 25th September 2016 EuroBSDcon 2016 OUTLINE Motivation Design and implementation
More informationBe Fast, Cheap and in Control with SwitchKV. Xiaozhou Li
Be Fast, Cheap and in Control with SwitchKV Xiaozhou Li Goal: fast and cost-efficient key-value store Store, retrieve, manage key-value objects Get(key)/Put(key,value)/Delete(key) Target: cluster-level
More informationFarewell to Servers: Hardware, Software, and Network Approaches towards Datacenter Resource Disaggregation
Farewell to Servers: Hardware, Software, and Network Approaches towards Datacenter Resource Disaggregation Yiying Zhang Datacenter 3 Monolithic Computer OS / Hypervisor 4 Can monolithic Application Hardware
More information