Practical, Real-time Centralized Control for CDN-based Live Video Delivery
|
|
- Jasper Reeves
- 5 years ago
- Views:
Transcription
1 Practical, Real-time Centralized Control for CDN-based Live Video Delivery Matt Mukerjee, David Naylor, Junchen Jiang, Dongsu Han, Srini Seshan, Hui Zhang
2 Combating Latency in Wide Area Control Planes Centralization can provide major benefits e.g., better performance, reliability, policy management, Scalability is hard on WAN due to latency
3 Control Planes in the 4D* Model CENTRAL CONTROLLER DECISION DISCOVERY DISSEMINATION DATA HTTP Server HTTP Server ROUTERS, etc. *Yan, Hong, et al. "Tesseract: A 4D Network Control Plane." NSDI. Vol
4 WAN Control Plane Latency CENTRAL CONTROLLER DECISION DISCOVERY DISSEMINATION DATA HTTP Server HTTP Server ROUTERS, etc.
5 WAN Control Plane Latency CENTRAL CONTROLLER DECISION DISCOVERY DISSEMINATION DATA HTTP Server HTTP Server ROUTERS, etc.
6 WAN Control Plane Latency CENTRAL CONTROLLER DECISION DISCOVERY DISSEMINATION DATA HTTP Server HTTP Server ROUTERS, etc.
7 WAN Control Plane Latency CENTRAL CONTROLLER DECISION DISCOVERY DISSEMINATION DATA HTTP Server HTTP Server ROUTERS, etc.
8 WAN Problems and Decision Planes Low Latency Decision Plane Traffic Engineering Solve with LP Live Video Delivery Solve with ILP? High Latency Decision Plane!
9 Attacking Decision Plane Latency Central Optimization Distributed Control Hybrid Control Quality and cost management Responsiveness to joins and failures
10 Outline Centralized Control Problems with Live Video Today Hybrid Control Putting it all Together Distributed Control
11 CDN Live Video Delivery Background Video Sources Reflector C A D B Data Control HTTP RESPONSE HTTP GET Legend Requests: Video 2 Responses: Video 2 Edge E F G Video Requests
12 CDN Live Video Delivery Background Video Sources Reflector Edge K C A D B K 2K Data 300 E F G Control DNS Link Cost Link Capacity Clients H I J
13 CDN Live Video Delivery Background Video Sources Edge 3K A B Data Link Cost Reflector Maximize C service D quality Link Capacity DNS E F G Control Objective: 2K K 300 & 0 Minimize 1 10 delivery 1 cost Clients H I J
14 Outline Centralized Control Problems with Live Video Today Hybrid Control Putting it all Together Distributed Control
15 Motivating Centralized Optimization Video Sources 3K A K B Data Control Reflector C D K 300 DNS Link Capacity Edge E F G Clients H I J
16 Motivating Centralized Optimization Video Sources 3K A K B Data Control Reflector Edge C D K 300 E F G Congestion! DNS Link Capacity Clients 300 H I J
17 Motivating Centralized Optimization Video Sources 3K A K B Data Control Reflector C D K 300 DNS Link Capacity Edge E F G Clients H I J
18 Motivating Centralized Optimization Video Sources 3K A K B Data Control Needs global view to coordinate Reflector Link Capacity videos C and network D resources DNS K 300 Edge Clients E F G H I J
19 Motivating Centralized Optimization Video Sources 3K A K B Data Control Reflector C D K 300 DNS Link Capacity Edge E F G Clients H I J
20 Motivating Centralized Optimization Video Sources 3K A K B Data Control Reflector C D K 300 Central Controller Link Capacity Edge E F G Clients H I J
21 Solving Centralized Optimization MAXIMIZE MINIMIZE SUBJECT TO SERVICE QUALITY DELIVERY COST DON T EXCEED LINK CAPACITY SENDER MUST HAVE RECEIVED VIDEO
22 Solving Centralized Optimization w s Pl SERVICE QUALITY max 2L AS,o2O Priority o Request l,o Serves l,o w c Pl P 2 2L,o2O Cost(l) Bitrate(o) Serves l,o 2 subject to: P DELIVERY, 2 COST 2, subject to: P 8l8 2 L, o 2 O : Serves l,o 2{0, 1} 2{ } 8 2 8l L : P DON T EXCEED LINK CAPACITY apple o Bitrate(o) Serves l,o apple Capacity(l) 8l 2 L, o 2 O : P l 0 2InLinks(l) Serves l 0,o Serves l,o SENDER MUST HAVE RECEIVED VIDEO
23 Centralized Optimization Avg. Bitrate (Kbps) Service Quality Optimal CDN # of Videos (Thousands) Simulation using Conviva traces, modeling user-generated content Delivery Cost (per request) CDN 2.0x OPTIMAL 1.0x Simulation using Conviva traces, modeling large sports events
24 Effects of Latency in Decision Plane Join Time (Seconds) Light Load Med. Load Hvy. Load Fully Centralized Slow join times! # of videos Experiments on EC2 nodes with a centralized controller at CMU across the Internet
25 Problems with Centralization Video Sources A B Data Control Legend Data Requests: HIGH LATENCY Reflector C D Central Controller Control Traffic: HIGH LATENCY Edge E F G The Internet Clients H I J
26 Outline Slow join times Centralized Control Problems with Live Video Today Hybrid Control Putting it all Together Distributed Control
27 Alternate Approach: Distributed Video Sources 3K A 2K B Data Control Legend Data Requests: Responses: Reflector Edge C D 200 2K? E F G Central Controller Link Capacity Clients H I J
28 Alternate Approach: Distributed Video Sources 3K A 2K B Data Control Legend Data Requests: Responses: Reflector Build distance-to-video tables Link Capacity C D at each cluster, for each Central video 200 2K 2K Controller Edge? E F G Clients H I J
29 Alternate Approach: Distributed Video Sources Reflector Edge 3K C A 2K D B 200 2K Data E F G Control Central Controller Legend Data Requests: Responses: 1; (B, 2K) Link Capacity? DISTANCE AT CLUSTER F VIDEO 1: Clients H I J
30 Alternate Approach: Distributed Video Sources Reflector Edge 3K C A 2K D B 200 2K Data E F G Control Central Controller Legend Data Requests: Responses: 1; (B, 2K) Link Capacity? DISTANCE AT CLUSTER F VIDEO 1: VIA C: 2; (B, ) Clients H I J # OF HOPS TO VIDEO PATH BOTTLENECK
31 Alternate Approach: Distributed Video Sources Reflector Edge 3K C A 2K D B 200 2K Data E F G Control Central Controller Legend Data Requests: Responses: 1; (B, 2K) Link Capacity? DISTANCE AT CLUSTER F VIDEO 1: VIA C: 2; (B, ) VIA D: 1; (D, ) Clients H I J
32 Alternate Approach: Distributed Video Sources Reflector Edge 3K C A 2K D B 200 2K Data E F G Control Central Controller Legend Data Requests: Responses: 1; (B, 2K) Link Capacity? DISTANCE AT CLUSTER F VIDEO 1: VIA C: 2; (B, ) VIA D: 1; (D, ) Clients H I J PICK SHORTEST PATH WITH ENOUGH CAPACITY
33 Alternate Approach: Distributed Video Sources 3K A 2K B Data Control Legend Data Requests: Responses: Reflector Edge C D 200 2K E F G Central Controller Link Capacity DISTANCE AT CLUSTER F VIDEO 1: VIA C: 2; (B, ) VIA D: 1; (D, ) Clients H I J PICK SHORTEST PATH WITH ENOUGH CAPACITY
34 Alternate Approach: Distributed Video Sources 3K A 2K B Data Control Legend Data Requests: Responses: Reflector Distributed decisions fast (ms) Link Capacity C D Central 200 2K Controller Edge but sub-optimal E F G DISTANCE AT CLUSTER F VIDEO 1: VIA C: 2; (B, ) VIA D: 1; (D, ) Clients H I J PICK SHORTEST PATH WITH ENOUGH CAPACITY
35 Alternate Approach: Distributed Video Sources 3K A 2K B Data Control Legend Data Requests: Responses: Reflector Edge Combine approaches? C D 200 2K E F G Central Controller Hybrid Control Link Capacity DISTANCE AT CLUSTER F VIDEO 1: VIA C: 2; (B, ) VIA D: 1; (D, ) Clients H I J PICK SHORTEST PATH WITH ENOUGH CAPACITY
36 Outline Slow join times Centralized Control Problems with Live Video Today Low bitrate Distributed Control Hybrid Control Putting it all Together
37 Combining Approaches: Hybrid Video Sources 3K A 2K B Data Control Legend Data Requests: HIGH LATENCY Responses: Reflector Edge C D 200 2K? E F G Central Controller HIGH LATENCY Clients The Internet H I J
38 Combining Approaches: Hybrid Video Sources 3K A 2K B Data Control Legend Data Requests: HIGH LATENCY Responses: Reflector C D 200 2K Central Controller HIGH LATENCY Edge E F G Clients The Internet H I J
39 Combining Approaches: Hybrid Video Sources Reflector Edge 3K C A 2K B D 200 2K Data E F G Control Legend Data Requests: HIGH LATENCY Responses: Control Traffic: Central Controller HIGH LATENCY Clients H I J The Internet
40 Challenges of Hybrid Control Forwarding loops Always forward requests upwards TRIVIAL State transitions Versioning and shadow FIBS PRIOR WORK Avoid bad control loop interactions CHALLENGING
41 Challenges of Hybrid Control Avoid bad control loop interactions CHALLENGING 1. Centralized decision has priority 2. Distributed uses residual after centralized 3. Distributed has no impact on current/future centralized decisions 4. Distributed s changes don t propagate
42 Hybrid Control and Responsiveness Join Time (Seconds) Light Load Med. Load Hvy. Load Hybrid Control Fully Centralized Fully Distributed Slow join times! # of videos Experiments on EC2 nodes with a centralized controller at CMU across the Internet
43 Hybrid Control and Responsiveness Join Time (Seconds) Light Load Med. Load Hvy. Load Hybrid Control Fully Centralized Fully Distributed Slow join times! Not stable # of videos Experiments on EC2 nodes with a centralized controller at CMU across the Internet
44 Hybrid Control and Responsiveness Join Time (Seconds) Light Load Med. Load Hvy. Load Hybrid Control Fully Centralized Fully Distributed Slow join times! Not stable # of videos Experiments on EC2 nodes with a centralized controller at CMU across the Internet Great join times and more stable
45 Conclusion We present a possible solution for combating decision plane latency Central Optimization Distributed Control Hybrid Control Quality and cost management Responsiveness to joins and failures
46 Conclusion Traffic Engineering Live Video Delivery Solve with LP Solve with ILP? Solve with X? Solve with X?
47 Practical, Real-time Centralized Control for CDN-based Live Video Delivery Matt Mukerjee, David Naylor, Junchen Jiang, Dongsu Han, Srini Seshan, Hui Zhang
48 Backup slides
49 Problems with Traffic Engineering Video Sources 3K A B 1.5K 1.5K 2K Data Control Even Split () Reflector Edge C D K 300 E F G Link Capacity Clients H I J
50 Problems with Traffic Engineering Video Sources Reflector Edge 3K C A D B 1.5K 1.5K 2K K 300 Data E F G Control Uneven Split (1.5K / 500) Link Capacity Clients 1.5K H I J 200
51 Distributed: Example of Sub-optimal Video Sources 3K A 2K B Data Control Wasting bandwidth Legend Data Requests: Responses: Reflector C D 200 2K 2K Link Capacity Edge E F G Clients H I J
52 Distributed: Example of Sub-optimal Video Sources 3K A 2K B Data Control Legend Data Requests: Responses: Reflector C D 200 2K 2K Coordination difficult without centralization Link Capacity Edge E F G Clients H I J
53 Trace-Driven Eval 3 Traces Avg Day: raw trace of music video provider Large Event: synthesized basketball game Heavy Tail: synthesized twitch/ustream like workload 4 Systems Everything Everywhere: all vids to all servers Overlay Multicast: globally optimal; no coordination CDN: greedy distribution scheme w/ DNS VDN: our system
54 Trace-Driven Eval
55 Existing Solutions Traffic Engineering (SWAN, B4, ) Works on aggregates at coarse timescales Overlay Multicast (Overcast, Bullet, ) Not designed for coordinating across streams Modern CDNs Previous work shows a centralized system could greatly improve user experience but would be difficult to design over Internet
Practical, Real-time Centralized Control for CDN-based Live Video Delivery
Practical, Real-time Centralized Control for -based Live Video Delivery Matthew K. Mukerjee mukerjee@cs.cmu.edu Dongsu Han dongsuh@ee.kaist.ac.kr David Naylor dnaylor@cs.cmu.edu Srinivasan Seshan srini@cs.cmu.edu
More informationC3: INTERNET-SCALE CONTROL PLANE FOR VIDEO QUALITY OPTIMIZATION
C3: INTERNET-SCALE CONTROL PLANE FOR VIDEO QUALITY OPTIMIZATION Aditya Ganjam, Jibin Zhan, Xi Liu, Faisal Siddiqi, Conviva Junchen Jiang, Vyas Sekar, Carnegie Mellon University Ion Stoica, University of
More informationCSEP 561 Routing. David Wetherall
CSEP 561 Routing David Wetherall djw@cs.washington.edu Routing Focus: How to find and set up paths through a network Distance-vector and link-state Application Shortest path routing Transport Key properties
More informationIt s Not the Cost, It s the Quality! Ion Stoica Conviva Networks and UC Berkeley
It s Not the Cost, It s the Quality! Ion Stoica Conviva Networks and UC Berkeley 1 A Brief History! Fall, 2006: Started Conviva with Hui Zhang (CMU)! Initial goal: use p2p technologies to reduce distribution
More informationEfficient Point to Multipoint Transfers Across Datacenters
Efficient Point to Multipoint Transfers cross Datacenters Mohammad Noormohammadpour 1, Cauligi S. Raghavendra 1, Sriram Rao, Srikanth Kandula 1 University of Southern California, Microsoft Source: https://azure.microsoft.com/en-us/overview/datacenters/how-to-choose/
More informationArvind Krishnamurthy Fall 2003
Overlay Networks Arvind Krishnamurthy Fall 003 Internet Routing Internet routing is inefficient: Does not always pick the lowest latency paths Does not always pick paths with low drop rates Experimental
More informationCSE 461 Routing. Routing. Focus: Distance-vector and link-state Shortest path routing Key properties of schemes
CSE 46 Routing Routing Focus: How to find and set up paths through a network Distance-vector and link-state Shortest path routing Key properties of schemes Application Transport Network Link Physical Forwarding
More informationTraffic Engineering with Forward Fault Correction
Traffic Engineering with Forward Fault Correction Harry Liu Microsoft Research 06/02/2016 Joint work with Ratul Mahajan, Srikanth Kandula, Ming Zhang and David Gelernter 1 Cloud services require large
More informationCSCE 463/612 Networks and Distributed Processing Spring 2018
CSCE 463/612 Networks and Distributed Processing Spring 2018 Network Layer V Dmitri Loguinov Texas A&M University April 17, 2018 Original slides copyright 1996-2004 J.F Kurose and K.W. Ross Chapter 4:
More informationSome Foundational Problems in Interdomain Routing
Some Foundational Problems in Interdomain Routing Nick Feamster, Hari Balakrishnan M.I.T. Computer Science and Artificial Intelligence Laboratory Jennifer Rexford AT&T Labs -- Research The state of interdomain
More informationApplication of SDN: Load Balancing & Traffic Engineering
Application of SDN: Load Balancing & Traffic Engineering Outline 1 OpenFlow-Based Server Load Balancing Gone Wild Introduction OpenFlow Solution Partitioning the Client Traffic Transitioning With Connection
More informationMulticast EECS 122: Lecture 16
Multicast EECS 1: Lecture 16 Department of Electrical Engineering and Computer Sciences University of California Berkeley Broadcasting to Groups Many applications are not one-one Broadcast Group collaboration
More informationContents. Overview Multicast = Send to a group of hosts. Overview. Overview. Implementation Issues. Motivation: ISPs charge by bandwidth
EECS Contents Motivation Overview Implementation Issues Ethernet Multicast IGMP Routing Approaches Reliability Application Layer Multicast Summary Motivation: ISPs charge by bandwidth Broadcast Center
More informationDistributed Systems. 21. Content Delivery Networks (CDN) Paul Krzyzanowski. Rutgers University. Fall 2018
Distributed Systems 21. Content Delivery Networks (CDN) Paul Krzyzanowski Rutgers University Fall 2018 1 2 Motivation Serving web content from one location presents problems Scalability Reliability Performance
More informationCS November 2018
Distributed Systems 21. Delivery Networks (CDN) Paul Krzyzanowski Rutgers University Fall 2018 1 2 Motivation Serving web content from one location presents problems Scalability Reliability Performance
More informationEE122: Multicast. Kevin Lai October 7, 2002
EE122: Multicast Kevin Lai October 7, 2002 Internet Radio www.digitallyimported.com (techno station) - sends out 128Kb/s MP3 music streams - peak usage ~9000 simultaneous streams only 5 unique streams
More informationEE122: Multicast. Internet Radio. Multicast Service Model 1. Motivation
Internet Radio EE122: Multicast Kevin Lai October 7, 2002 wwwdigitallyimportedcom (techno station) - sends out 128Kb/s MP music streams - peak usage ~9000 simultaneous streams only 5 unique streams (trance,
More informationCache Management for TelcoCDNs. Daphné Tuncer Department of Electronic & Electrical Engineering University College London (UK)
Cache Management for TelcoCDNs Daphné Tuncer Department of Electronic & Electrical Engineering University College London (UK) d.tuncer@ee.ucl.ac.uk 06/01/2017 Agenda 1. Internet traffic: trends and evolution
More informationNetwork Layer (Routing)
Network Layer (Routing) Topics Network service models Datagrams (packets), virtual circuits IP (Internet Protocol) Internetworking Forwarding (Longest Matching Prefix) Helpers: ARP and DHCP Fragmentation
More informationOverlay Networks for Multimedia Contents Distribution
Overlay Networks for Multimedia Contents Distribution Vittorio Palmisano vpalmisano@gmail.com 26 gennaio 2007 Outline 1 Mesh-based Multicast Networks 2 Tree-based Multicast Networks Overcast (Cisco, 2000)
More informationEECS 122: Introduction to Computer Networks Overlay Networks and P2P Networks. Overlay Networks: Motivations
EECS 122: Introduction to Computer Networks Overlay Networks and P2P Networks Ion Stoica Computer Science Division Department of Electrical Engineering and Computer Sciences University of California, Berkeley
More information! Naive n-way unicast does not scale. ! IP multicast to the rescue. ! Extends IP architecture for efficient multi-point delivery. !
Bayeux: An Architecture for Scalable and Fault-tolerant Wide-area Data Dissemination ACM NOSSDAV 001 Shelley Q. Zhuang, Ben Y. Zhao, Anthony D. Joseph, Randy H. Katz, John D. Kubiatowicz {shelleyz, ravenben,
More informationPeer-to-Peer Streaming Systems. Behzad Akbari
Peer-to-Peer Streaming Systems Behzad Akbari 1 Outline Introduction Scaleable Streaming Approaches Application Layer Multicast Content Distribution Networks Peer-to-Peer Streaming Metrics Current Issues
More informationRPT: Re-architecting Loss Protection for Content-Aware Networks
RPT: Re-architecting Loss Protection for Content-Aware Networks Dongsu Han, Ashok Anand ǂ, Aditya Akella ǂ, and Srinivasan Seshan Carnegie Mellon University ǂ University of Wisconsin-Madison Motivation:
More informationEfficient Point to Multipoint Transfers Across Datacenters
Efficient Point to Multipoint Transfers cross atacenters Mohammad Noormohammadpour 1, Cauligi S. Raghavendra 1, Sriram Rao, Srikanth Kandula 1 University of Southern California, Microsoft Source: https://azure.microsoft.com/en-us/overview/datacenters/how-to-choose/
More informationLecture 4: Intradomain Routing. CS 598: Advanced Internetworking Matthew Caesar February 1, 2011
Lecture 4: Intradomain Routing CS 598: Advanced Internetworking Matthew Caesar February 1, 011 1 Robert. How can routers find paths? Robert s local DNS server 10.1.8.7 A 10.1.0.0/16 10.1.0.1 Routing Table
More informationCOMP/ELEC 429/556 Introduction to Computer Networks
COMP/ELEC 429/556 Introduction to Computer Networks Weighted Fair Queuing Some slides used with permissions from Edward W. Knightly, T. S. Eugene Ng, Ion Stoica, Hui Zhang T. S. Eugene Ng eugeneng at cs.rice.edu
More informationPage 1. Outline / Computer Networking : 1 st Generation Commercial PC/Packet Video Technologies
Outline 15-441/15-641 Computer Networking Lecture 18 Internet Video Delivery Peter Steenkiste Slides by Professor Hui Zhang Background Technologies: - HTTP download - Real-time streaming - HTTP streaming
More informationNeural Adaptive Content-aware Internet Video Delivery. Hyunho Yeo, Youngmok Jung, Jaehong Kim, Jinwoo Shin, Dongsu Han
Neural Adaptive Content-aware Internet Video Delivery Hyunho Yeo, Youngmok Jung, Jaehong Kim, Jinwoo Shin, Dongsu Han Observation on Current Video Ecosystem 2 Adaptive streaming has been widely deployed
More informationDynamic Load-Balanced Multicast for Data-Intensive Applications on Clouds 1
Dynamic Load-Balanced Multicast for Data-Intensive Applications on Clouds 1 Contents: Introduction Multicast on parallel distributed systems Multicast on P2P systems Multicast on clouds High performance
More informationOctoshape. Commercial hosting not cable to home, founded 2003
Octoshape Commercial hosting not cable to home, founded 2003 Broadcasting fee is paid by broadcasters Free for consumers Audio and Video, 32kbps to 800kbps Mesh based, bit-torrent like, Content Server
More informationCS 356: Computer Network Architectures. Lecture 24: IP Multicast and QoS [PD] Chapter 4.2, 6.5. Xiaowei Yang
CS 356: Computer Network Architectures Lecture 24: IP Multicast and QoS [PD] Chapter 4.2, 6.5 Xiaowei Yang xwy@cs.duke.edu Overview Two historic important topics in networking Multicast QoS Limited Deployment
More informationIntroduction. Routing & Addressing: Multihoming 10/25/04. The two SIGCOMM papers. A Comparison of Overlay Routing and Multihoming Route Control
Introduction Routing & Addressing: Multihoming 10/25/04 Hong Tat Tong Two attempts to control/ensure best performance over the Internet Multihoming from the endpoints Overlay with some help from middle-boxes
More informationOverlay Networks: Motivations. EECS 122: Introduction to Computer Networks Overlay Networks and P2P Networks. Motivations (cont d) Goals.
Overlay Networks: Motivations CS : Introduction to Computer Networks Overlay Networks and PP Networks Ion Stoica Computer Science Division Department of lectrical ngineering and Computer Sciences University
More informationBitTorrent and CoolStreaming
BitTorrent and CoolStreaming Jukka K. Nurminen Data Communications Software (DCS) Lab, Department of Computer Science and Engineering, Aalto University Jukka K. Nurminen Aalto University P2P Networks BitTorrent
More informationASM. Engineering Workshops
1 ASM 2 ASM Allows SPTs and RPTs RP: Matches senders with receivers Provides network source discovery Typically uses RPT to bootstrap SPT RPs can be learned via: Static configuration recommended Anycast-RP
More informationMohammad Hossein Manshaei 1393
Mohammad Hossein Manshaei manshaei@gmail.com 1393 Voice and Video over IP Slides derived from those available on the Web site of the book Computer Networking, by Kurose and Ross, PEARSON 2 Multimedia networking:
More informationCQNCR: Optimal VM Migration Planning in Cloud Data Centers
CQNCR: Optimal VM Migration Planning in Cloud Data Centers Presented By Md. Faizul Bari PhD Candidate David R. Cheriton School of Computer science University of Waterloo Joint work with Mohamed Faten Zhani,
More informationMany-to-Many Communications in HyperCast
Many-to-Many Communications in HyperCast Jorg Liebeherr University of Virginia Jörg Liebeherr, 2001 HyperCast Project HyperCast is a set of protocols for large-scale overlay multicasting and peer-to-peer
More informationChapter 7 CONCLUSION
97 Chapter 7 CONCLUSION 7.1. Introduction A Mobile Ad-hoc Network (MANET) could be considered as network of mobile nodes which communicate with each other without any fixed infrastructure. The nodes in
More informationCongestion and its control: Broadband access networks. David Clark MIT CFP October 2008
Congestion and its control: Broadband access networks David Clark MIT CFP October 2008 Outline Quick review/definition of congestion. Quick review of historical context. Quick review of access architecture.
More informationJunchen Jiang. 513 N. Neville St. Apt C3, Pittsburgh, PA, URL: https://www.cs.cmu.edu/ junchenj/
Junchen Jiang 513 N. Neville St. Apt C3, Pittsburgh, PA, 15213 Email: junchenj@cs.cmu.edu URL: https://www.cs.cmu.edu/ junchenj/ RESEARCH INTERESTS Computer networks, distributed systems, big data, and
More informationVirtualizing The Network For Fun and Profit. Building a Next-Generation Network Infrastructure using EVPN/VXLAN
Virtualizing The Network For Fun and Profit Building a Next-Generation Network Infrastructure using EVPN/VXLAN By Richard A Steenbergen A BRIEF HISTORY OF LAYER 2 NETWORKING Historically,
More informationUDP, TCP, IP multicast
UDP, TCP, IP multicast Dan Williams In this lecture UDP (user datagram protocol) Unreliable, packet-based TCP (transmission control protocol) Reliable, connection oriented, stream-based IP multicast Process-to-Process
More informationCollaborative Multi-Source Scheme for Multimedia Content Distribution
Collaborative Multi-Source Scheme for Multimedia Content Distribution Universidad Autónoma Metropolitana-Cuajimalpa, Departament of Information Technology, Mexico City, Mexico flopez@correo.cua.uam.mx
More informationCore-Stateless Fair Queueing: Achieving Approximately Fair Bandwidth Allocations in High Speed Networks. Congestion Control in Today s Internet
Core-Stateless Fair Queueing: Achieving Approximately Fair Bandwidth Allocations in High Speed Networks Ion Stoica CMU Scott Shenker Xerox PARC Hui Zhang CMU Congestion Control in Today s Internet Rely
More informationCS4700/CS5700 Fundamentals of Computer Networks
CS4700/CS5700 Fundamentals of Computer Networks Lecture 22: Overlay networks Slides used with permissions from Edward W. Knightly, T. S. Eugene Ng, Ion Stoica, Hui Zhang Alan Mislove amislove at ccs.neu.edu
More informationRouting in Delay Tolerant Networks (2)
Routing in Delay Tolerant Networks (2) Primary Reference: E. P. C. Jones, L. Li and P. A. S. Ward, Practical Routing in Delay-Tolerant Networks, SIGCOMM 05, Workshop on DTN, August 22-26, 2005, Philadelphia,
More informationMohammad Hossein Manshaei 1393
Mohammad Hossein Manshaei manshaei@gmail.com 1393 Voice and Video over IP Slides derived from those available on the Web site of the book Computer Networking, by Kurose and Ross, PEARSON 2 multimedia applications:
More informationOn low-latency-capable topologies, and their impact on the design of intra-domain routing
On low-latency-capable topologies, and their impact on the design of intra-domain routing Nikola Gvozdiev, Stefano Vissicchio, Brad Karp, Mark Handley University College London (UCL) We want low latency!
More informationInternet Video Delivery. Professor Hui Zhang
18-345 Internet Video Delivery Professor Hui Zhang 1 1990 2004: 1 st Generation Commercial PC/Packet Video Technologies Simple video playback, no support for rich app Not well integrated with Web browser
More informationSWAN: Software-driven wide area network. Ratul Mahajan
SWAN: Software-driven wide area network Ratul Mahajan Partners in crime Vijay Gill Chi-Yao Hong Srikanth Kandula Ratul Mahajan Mohan Nanduri Ming Zhang Roger Wattenhofer Rohan Gandhi Xin Jin Harry Liu
More informationConfused, Timid, and Unstable: Picking a Video Streaming Rate is Hard
Confused, Timid, and Unstable: Picking a Video Streaming Rate is Hard Five students from Stanford Published in 2012 ACM s Internet Measurement Conference (IMC) 23 citations Ahmad Tahir 1/26 o Problem o
More informationNext Steps Spring 2011 Lecture #18. Multi-hop Networks. Network Reliability. Have: digital point-to-point. Want: many interconnected points
Next Steps Have: digital point-to-point We ve worked on link signaling, reliability, sharing Want: many interconnected points 6.02 Spring 2011 Lecture #18 multi-hop networks: design criteria network topologies
More informationExpeditus: Congestion-Aware Load Balancing in Clos Data Center Networks
Expeditus: Congestion-Aware Load Balancing in Clos Data Center Networks Peng Wang, Hong Xu, Zhixiong Niu, Dongsu Han, Yongqiang Xiong ACM SoCC 2016, Oct 5-7, Santa Clara Motivation Datacenter networks
More informationIntroduction. Network Architecture Requirements of Data Centers in the Cloud Computing Era
Massimiliano Sbaraglia Network Engineer Introduction In the cloud computing era, distributed architecture is used to handle operations of mass data, such as the storage, mining, querying, and searching
More informationLecture 13: Traffic Engineering
Lecture 13: Traffic Engineering CSE 222A: Computer Communication Networks Alex C. Snoeren Thanks: Mike Freedman, Nick Feamster Lecture 13 Overview Evolution of routing in the ARPAnet Today s TE: Adjusting
More informationXIA: An Architecture for a Trustworthy and Evolvable Internet
XIA: An Architecture for a Trustworthy and Evolvable Internet Peter Steenkiste Dave Andersen, David Eckhardt, Sara Kiesler, Jon Peha, Adrian Perrig, Srini Seshan, Marvin Sirbu, Hui Zhang Carnegie Mellon
More informationNET ID. CS519, Prelim (March 17, 2004) NAME: You have 50 minutes to complete the test. 1/17
CS519, Prelim (March 17, 2004) NAME: You have 50 minutes to complete the test. 1/17 Q1. 2 points Write your NET ID at the top of every page of this test. Q2. X points Name 3 advantages of a circuit network
More informationLecture 6: Multicast
Lecture 6: Multicast Challenge: how do we efficiently send messages to a group of machines? Need to revisit all aspects of networking Last time outing This time eliable delivery Ordered delivery Congestion
More informationSYSC 5801 Protection and Restoration
SYSC 5801 Protection and Restoration Introduction Fact: Networks fail. Types of failures: Link failures Node failures Results: packet losses, waste of resources, and higher delay. What IGP does in the
More informationImproving Search In Peer-to-Peer Systems
Improving Search In Peer-to-Peer Systems Presented By Jon Hess cs294-4 Fall 2003 Goals Present alternative searching methods for systems with loose integrity constraints Probabilistically optimize over
More informationWhat is the difference between unicast and multicast? (P# 114)
1 FINAL TERM FALL2011 (eagle_eye) CS610 current final term subjective all solved data by eagle_eye MY paper of CS610 COPUTER NETWORKS There were 30 MCQs Question no. 31 (Marks2) Find the class in 00000001.001011.1001.111
More informationStaying FIT: Efficient Load Shedding Techniques for Distributed Stream Processing
Staying FIT: Efficient Load Shedding Techniques for Distributed Stream Processing Nesime Tatbul Uğur Çetintemel Stan Zdonik Talk Outline Problem Introduction Approach Overview Advance Planning with an
More informationMultipath Transport, Resource Pooling, and implications for Routing
Multipath Transport, Resource Pooling, and implications for Routing Mark Handley, UCL and XORP, Inc Also: Damon Wischik, UCL Marcelo Bagnulo Braun, UC3M The members of Trilogy project: www.trilogy-project.org
More informationA Routing Infrastructure for XIA
A Routing Infrastructure for XIA Aditya Akella and Peter Steenkiste Dave Andersen, John Byers, David Eckhardt, Sara Kiesler, Jon Peha, Adrian Perrig, Srini Seshan, Marvin Sirbu, Hui Zhang FIA PI Meeting,
More informationOverview. TCP & router queuing Computer Networking. TCP details. Workloads. TCP Performance. TCP Performance. Lecture 10 TCP & Routers
Overview 15-441 Computer Networking TCP & router queuing Lecture 10 TCP & Routers TCP details Workloads Lecture 10: 09-30-2002 2 TCP Performance TCP Performance Can TCP saturate a link? Congestion control
More informationJunchen Jiang. 200 West 16th Street, New York, NY, URL: https://www.cs.cmu.
Junchen Jiang 200 West 16th Street, New York, NY, 10011 Email: junchenj@uchicago.edu, junchenj@cs.cmu.edu URL: https://www.cs.cmu.edu/ junchenj/ RESEARCH INTERESTS Computer networks, large-scale data analytics
More informationASPERA HIGH-SPEED TRANSFER. Moving the world s data at maximum speed
ASPERA HIGH-SPEED TRANSFER Moving the world s data at maximum speed ASPERA HIGH-SPEED FILE TRANSFER Aspera FASP Data Transfer at 80 Gbps Elimina8ng tradi8onal bo
More informationBuilding an Internet-Scale Publish/Subscribe System
Building an Internet-Scale Publish/Subscribe System Ian Rose Mema Roussopoulos Peter Pietzuch Rohan Murty Matt Welsh Jonathan Ledlie Imperial College London Peter R. Pietzuch prp@doc.ic.ac.uk Harvard University
More information1-800-OVERLAYS: Using Overlay Networks to Improve VoIP Quality. Yair Amir, Claudiu Danilov, Stuart Goose, David Hedqvist, Andreas Terzis
1-800-OVERLAYS: Using Overlay Networks to Improve VoIP Quality Yair Amir, Claudiu Danilov, Stuart Goose, David Hedqvist, Andreas Terzis Voice over IP To call or not to call The Internet started as an overlay
More informationGoals. EECS 122: Introduction to Computer Networks Overlay Networks and P2P Networks. Solution. Overlay Networks: Motivations.
Goals CS : Introduction to Computer Networks Overlay Networks and PP Networks Ion Stoica Computer Science Division Department of lectrical ngineering and Computer Sciences University of California, Berkeley
More informationP2P content distribution Jukka K. Nurminen
P2P content distribution Jukka K. Nurminen 1 V1-Filename.ppt / yyyy-mm-dd / Initials BitTorrent content downloading Efficient content distribution Bram Cohen, 2001 File divided into pieces Each recipient
More informationProtecting Network Quality of Service Against Denial of Service Attacks
Protecting Network Quality of Service Against Denial of Service Attacks Douglas S. Reeves Peter Wurman NC State University S. Felix Wu U.C. Davis Dan Stevenson Xiaoyong Wu MCNC DARPA FTN PI Meeting January
More informationIMPROVING LIVE PERFORMANCE IN HTTP ADAPTIVE STREAMING SYSTEMS
IMPROVING LIVE PERFORMANCE IN HTTP ADAPTIVE STREAMING SYSTEMS Kevin Streeter Adobe Systems, USA ABSTRACT While HTTP adaptive streaming (HAS) technology has been very successful, it also generally introduces
More informationAn Implementation of the Homa Transport Protocol in RAMCloud. Yilong Li, Behnam Montazeri, John Ousterhout
An Implementation of the Homa Transport Protocol in RAMCloud Yilong Li, Behnam Montazeri, John Ousterhout Introduction Homa: receiver-driven low-latency transport protocol using network priorities HomaTransport
More informationOverview Computer Networking Lecture 16: Delivering Content: Peer to Peer and CDNs Peter Steenkiste
Overview 5-44 5-44 Computer Networking 5-64 Lecture 6: Delivering Content: Peer to Peer and CDNs Peter Steenkiste Web Consistent hashing Peer-to-peer Motivation Architectures Discussion CDN Video Fall
More informationLeveraging Software-Defined Storage to Meet Today and Tomorrow s Infrastructure Demands
Leveraging Software-Defined Storage to Meet Today and Tomorrow s Infrastructure Demands Unleash Your Data Center s Hidden Power September 16, 2014 Molly Rector CMO, EVP Product Management & WW Marketing
More informationPage 1. Quality of Service. CS 268: Lecture 13. QoS: DiffServ and IntServ. Three Relevant Factors. Providing Better Service.
Quality of Service CS 268: Lecture 3 QoS: DiffServ and IntServ Ion Stoica Computer Science Division Department of Electrical Engineering and Computer Sciences University of California, Berkeley Berkeley,
More informationimproving the performance and robustness of P2P live streaming with Contracts
MICHAEL PIATEK AND ARVIND KRISHNAMURTHY improving the performance and robustness of P2P live streaming with Contracts Michael Piatek is a graduate student at the University of Washington. After spending
More informationAutoTune: Game-based Adaptive Bitrate Streaming in P2P-Assisted Cloud-Based VoD Systems
AutoTune: Game-based Adaptive Bitrate Streaming in P2P-Assisted Cloud-Based VoD Systems Yuhua Lin and Haiying Shen Dept. of Electrical and Computer Engineering Clemson University, SC, USA Outline Introduction
More informationP2P content distribution
P2P content distribution T-110.7100 Applications and Services in Internet, Fall 2010 Jukka K. Nurminen 1 V1-Filename.ppt / yyyy-mm-dd / Initials Steps of content sharing Share content Find content Transfer
More informationCarnegie Mellon Computer Science Department Spring 2015 Midterm Exam
Carnegie Mellon Computer Science Department. 15-744 Spring 2015 Midterm Exam Name: Andrew ID: INSTRUCTIONS: There are 7 pages (numbered at the bottom). Make sure you have all of them. Please write your
More informationOutline. EE 122: Networks Performance & Modeling. Outline. Motivations. Definitions. Timing Diagrams. Ion Stoica TAs: Junda Liu, DK Moon, David Zats
EE 122: Networks Performance & Modeling Ion Stoica As: Junda Liu, DK Moon, David Zats http://inst.eecs.berkeley.edu/~ee122/fa09 (Materials with thanks to Vern Paxson, Jennifer Rexford, and colleagues at
More informationNSF Future Internet Architecture. Outline. Predicting the Future is Hard! The expressive Internet Architecture: from Architecture to Network
The expressive Internet Architecture: from Architecture to Network Peter Steenkiste Dave Andersen, David Eckhardt, Sara Kiesler, Jon Peha, Adrian Perrig, Srini Seshan, Marvin Sirbu, Hui Zhang Carnegie
More informationCS November 2017
Distributed Systems 21. Delivery Networks () Paul Krzyzanowski Rutgers University Fall 2017 1 2 Motivation Serving web content from one location presents problems Scalability Reliability Performance Flash
More informationInterdomain Routing Design for MobilityFirst
Interdomain Routing Design for MobilityFirst October 6, 2011 Z. Morley Mao, University of Michigan In collaboration with Mike Reiter s group 1 Interdomain routing design requirements Mobility support Network
More informationOutline 9.2. TCP for 2.5G/3G wireless
Transport layer 9.1 Outline Motivation, TCP-mechanisms Classical approaches (Indirect TCP, Snooping TCP, Mobile TCP) PEPs in general Additional optimizations (Fast retransmit/recovery, Transmission freezing,
More information1 Case Studies - AppEx CDN Acceleration. Case Studies. Introduction. Case Study # 1 CDN (Content Delivery Network)
Introduction AppEx was founded in 2007 and has focused on the CDN market for many years. We have had great success Internationally helping CDNs to improve user experiences and be competitive in the constantly
More informationDCRoute: Speeding up Inter-Datacenter Traffic Allocation while Guaranteeing Deadlines
DCRoute: Speeding up Inter-Datacenter Traffic Allocation while Guaranteeing Deadlines Mohammad Noormohammadpour, Cauligi S. Raghavendra Ming Hsieh Department of Electrical Engineering University of Southern
More informationPerformance Modeling
Performance Modeling EECS 489 Computer Networks http://www.eecs.umich.edu/~zmao/eecs489 Z. Morley Mao Tuesday Sept 14, 2004 Acknowledgement: Some slides taken from Kurose&Ross and Katz&Stoica 1 Administrivia
More informationConfiguring BGP community 43 Configuring a BGP route reflector 44 Configuring a BGP confederation 44 Configuring BGP GR 45 Enabling Guard route
Contents Configuring BGP 1 Overview 1 BGP speaker and BGP peer 1 BGP message types 1 BGP path attributes 2 BGP route selection 6 BGP route advertisement rules 6 BGP load balancing 6 Settlements for problems
More informationProject Proposal: Deep Routing Simulation
Project Proposal: Deep Routing Simulation Principal Investigator: Mr. A. Herbert February 2012 1 Statement of the Problem Currently the large scale network simulators lacks the ability to correctly simulate
More informationWarpTCP WHITE PAPER. Technology Overview. networks. -Improving the way the world connects -
WarpTCP WHITE PAPER Technology Overview -Improving the way the world connects - WarpTCP - Attacking the Root Cause TCP throughput reduction is often the bottleneck that causes data to move at slow speed.
More informationRCD: Rapid Close to Deadline Scheduling for Datacenter Networks
RCD: Rapid Close to Deadline Scheduling for Datacenter Networks Mohammad Noormohammadpour 1, Cauligi S. Raghavendra 1, Sriram Rao 2, Asad M. Madni 3 1 Ming Hsieh Department of Electrical Engineering, University
More informationBayeux: An Architecture for Scalable and Fault Tolerant Wide area Data Dissemination
Bayeux: An Architecture for Scalable and Fault Tolerant Wide area Data Dissemination By Shelley Zhuang,Ben Zhao,Anthony Joseph, Randy Katz,John Kubiatowicz Introduction Multimedia Streaming typically involves
More informationContent distribution networks
Content distribution networks v challenge: how to stream content (selected from millions of videos) to hundreds of thousands of simultaneous users? v option 2: store/serve multiple copies of videos at
More informationICS 451: Today's plan. Network Layer Protocols: virtual circuits Static Routing Distance-Vector Routing
ICS 451: Today's plan Network Layer Protocols: virtual circuits Static Routing Distance-Vector Routing Virtual Circuits: Motivation Implementing the routing table in hardware can be expensive to make it
More informationCOMP6218: Content Caches. Prof Leslie Carr
COMP6218: Content Caches Prof Leslie Carr 1 Slashdot.org The Slashdot effect, also known as slashdotting, occurs when a popular website links to a smaller site, causing a massive increase in traffic 2
More informationBGP Case Studies. ISP Workshops
BGP Case Studies ISP Workshops These materials are licensed under the Creative Commons Attribution-NonCommercial 4.0 International license (http://creativecommons.org/licenses/by-nc/4.0/) Last updated
More information