Coping with network performance

Size: px
Start display at page:

Download "Coping with network performance"

Transcription

1 Coping with network performance Ankit Singla ETH Zürich P. Brighten Godfrey UIUC

2 The end to end principle END-TO-END ARGUMENTS IN SYSTEM DESIGN J.H. Saltzer, D.P. Reed and D.D. Clark* M.I.T. Laboratory for Computer Science IEEE ICDCS, 1981

3 The end to end principle END-TO-END ARGUMENTS IN SYSTEM DESIGN J.H. Saltzer, D.P. Reed and D.D. Clark* M.I.T. Laboratory for Computer Science IEEE ICDCS, 1981 the end-to-end check of the file transfer application must still be implemented no matter how reliable the communication system becomes

4 A few examples in this lesson 1 Redundant requests to lower tail latency 2 Application layer handling of TCP incast 3 Rate adaptation for video streaming

5 Latency in the tail

6 Latency in the tail the median workflow in production at Bing has 15 stages and 10% of the stages process the query in parallel on 1000s of servers Speeding up Distributed Request-Response Workflows Jalaparti et al., ACM SIGCOMM 2013

7 Latency in the tail the median workflow in production at Bing has 15 stages and 10% of the stages process the query in parallel on 1000s of servers Speeding up Distributed Request-Response Workflows Jalaparti et al., ACM SIGCOMM 2013 loading one of our popular pages results in an average of 521 distinct items fetched from memcache Scaling Memcache at Facebook Nishtala et al., USENIX NSDI 2013

8 Latency in the tail

9 Latency in distributed data stores Rapid read protection allows the coordinator to monitor the outstanding requests and send redundant requests to other replicas when the original is slower than expected DataStax, Rapid read protection in Cassandra 2.0.2

10 Latency in distributed data stores Rapid read protection allows the coordinator to monitor the outstanding requests and send redundant requests to other replicas when the original is slower than expected DataStax, Rapid read protection in Cassandra Replica1 Replica 2 Client Coordinator Replica 3

11 Latency in distributed data stores Rapid read protection allows the coordinator to monitor the outstanding requests and send redundant requests to other replicas when the original is slower than expected DataStax, Rapid read protection in Cassandra Replica1 Replica 2 Client Coordinator Replica 3

12 Latency in distributed data stores Rapid read protection allows the coordinator to monitor the outstanding requests and send redundant requests to other replicas when the original is slower than expected DataStax, Rapid read protection in Cassandra Replica1 Replica 2 Client Coordinator Replica 3

13 Latency in distributed data stores Rapid read protection allows the coordinator to monitor the outstanding requests and send redundant requests to other replicas when the original is slower than expected DataStax, Rapid read protection in Cassandra Client Coordinator Replica1 x Replica 3 Replica 2

14 Latency in distributed data stores Rapid read protection allows the coordinator to monitor the outstanding requests and send redundant requests to other replicas when the original is slower than expected DataStax, Rapid read protection in Cassandra Client Coordinator Replica1 x Replica 3 Replica 2

15 Latency in distributed data stores Rapid read protection allows the coordinator to monitor the outstanding requests and send redundant requests to other replicas when the original is slower than expected DataStax, Rapid read protection in Cassandra Client Coordinator Replica1 x Replica 3 Replica 2

16 Latency in distributed data stores Rapid read protection allows the coordinator to monitor the outstanding requests and send redundant requests to other replicas when the original is slower than expected DataStax, Rapid read protection in Cassandra Achieving Rapid Response Times in Large Online Services, Jeff Google, 2012 Avg Std Dev 95%ile 99%ile 99.9%ile No backups 33 ms 1524 ms 24 ms 52 ms 994 ms Backup after 10 ms 14 ms 4 ms 20 ms 23 ms 50 ms

17 When and why does this work? Server response-time is 10ms for 99% of requests; 1s for 1%.

18 When and why does this work? Server response-time is 10ms for 99% of requests; 1s for 1%. P(request is slow) = 0.01

19 When and why does this work? Server response-time is 10ms for 99% of requests; 1s for 1%. P(request is slow) = 0.01 P(request is slow AND redundant request is slow)

20 When and why does this work? Server response-time is 10ms for 99% of requests; 1s for 1%. P(request is slow) = 0.01 P(request is slow AND redundant request is slow) Crucial assumption: independence across requests

21 When and why does this work? Server response-time is 10ms for 99% of requests; 1s for 1%. P(request is slow) = 0.01 P(request is slow AND redundant request is slow) = P(request is slow) * P(request is slow) = Crucial assumption: independence across requests

22 When and why does this work? Server response-time is 10ms for 99% of requests; 1s for 1%. My job is 100 requests. What is the likelihood it takes 1s? Crucial assumption: independence across requests

23 When and why does this work? Server response-time is 10ms for 99% of requests; 1s for 1%. My job is 100 requests. What is the likelihood it takes 1s? Bad event Probability 1 slow requests 0.63 Crucial assumption: independence across requests

24 When and why does this work? Server response-time is 10ms for 99% of requests; 1s for 1%. My job is 100 requests. What is the likelihood it takes 1s? Bad event Probability 1 slow requests slow requests 0.26 Crucial assumption: independence across requests

25 When and why does this work? Server response-time is 10ms for 99% of requests; 1s for 1%. My job is 100 requests. What is the likelihood it takes 1s? Bad event Probability 1 slow requests slow requests slow requests slow requests slow requests Crucial assumption: independence across requests

26 When and why does this work? Server response-time is 10ms for 99% of requests; 1s for 1%. My job is 100 requests. What is the likelihood it takes 1s? Bad event Probability 1 slow requests slow requests slow requests slow requests 0.02 What if we replicated requests outstanding after 10ms? 5 slow requests Crucial assumption: independence across requests

27 The costs of redundancy 1 Client-side cost of making redundant requests 2 Additional network traffic for requests 3 Higher server load from additional work

28 The costs of redundancy 1 Client-side cost of making redundant requests 2 Additional network traffic for requests 3 Higher server load from additional work The costs are often low, and worth incurring

29 Dealing with TCP Incast

30 Dealing with TCP Incast Limit the size of individual responses Ignore a late response from a worker Deliberately space out requests and responses

31 Spacing out requests / responses DCTCP: Efficient Packet Transport for the Commoditized Data Center Mohammad Alizadeh, Albert Greenberg, David A. Maltz, Jitu Padhye, Parveen Patel, Balaji Prabhakar, Sudipta Sengupta, Murari Sridharan TRACT ACM SIGCOMM, 2010 Latency (ms) Time

32 Spacing out requests / responses DCTCP: Efficient Packet Transport for the Commoditized Data Center Mohammad Alizadeh, Albert Greenberg, David A. Maltz, Jitu Padhye, Parveen Patel, Balaji Prabhakar, Sudipta Sengupta, Murari Sridharan TRACT ACM SIGCOMM, 2010 Latency (ms) Time After request jittering is turned off

33 Dealing with TCP Incast USENIX NSDI 2013 Scaling Memcache at Facebook Rajesh Nishtala, Hans Fugal, Steven Grimm, Marc Kwiatkowski, Herman Lee, Harry C. Li, Ryan McElroy, Mike Paleczny, Daniel Peek, Paul Saab, David Stafford, Tony Tung, Venkateshwaran Venkataramani {sgrimm, {herman, hcli, rm, mpal, dpeek, ps, dstaff, ttung, Facebook Inc. Outstanding requests All requests queued at a client

34 Dealing with TCP Incast USENIX NSDI 2013 Scaling Memcache at Facebook Rajesh Nishtala, Hans Fugal, Steven Grimm, Marc Kwiatkowski, Herman Lee, Harry C. Li, Ryan McElroy, Mike Paleczny, Daniel Peek, Paul Saab, David Stafford, Tony Tung, Venkateshwaran Venkataramani {sgrimm, {herman, hcli, rm, mpal, dpeek, ps, dstaff, ttung, Facebook Inc. All requests queued at a client

35 Dealing with TCP Incast USENIX NSDI 2013 Scaling Memcache at Facebook Rajesh Nishtala, Hans Fugal, Steven Grimm, Marc Kwiatkowski, Herman Lee, Harry C. Li, Ryan McElroy, Mike Paleczny, Daniel Peek, Paul Saab, David Stafford, Tony Tung, Venkateshwaran Venkataramani {sgrimm, {herman, hcli, rm, mpal, dpeek, ps, dstaff, ttung, Facebook Inc. All requests queued at a client

36 Dealing with TCP Incast USENIX NSDI 2013 Scaling Memcache at Facebook Rajesh Nishtala, Hans Fugal, Steven Grimm, Marc Kwiatkowski, Herman Lee, Harry C. Li, Ryan McElroy, Mike Paleczny, Daniel Peek, Paul Saab, David Stafford, Tony Tung, Venkateshwaran Venkataramani {sgrimm, {herman, hcli, rm, mpal, dpeek, ps, dstaff, ttung, Facebook Inc. All requests queued at a client

37 Dealing with TCP Incast USENIX NSDI 2013 Scaling Memcache at Facebook Rajesh Nishtala, Hans Fugal, Steven Grimm, Marc Kwiatkowski, Herman Lee, Harry C. Li, Ryan McElroy, Mike Paleczny, Daniel Peek, Paul Saab, David Stafford, Tony Tung, Venkateshwaran Venkataramani {sgrimm, {herman, hcli, rm, mpal, dpeek, ps, dstaff, ttung, Facebook Inc. All requests queued at a client

38 Dealing with TCP Incast USENIX NSDI 2013 Scaling Memcache at Facebook Rajesh Nishtala, Hans Fugal, Steven Grimm, Marc Kwiatkowski, Herman Lee, Harry C. Li, Ryan McElroy, Mike Paleczny, Daniel Peek, Paul Saab, David Stafford, Tony Tung, Venkateshwaran Venkataramani {sgrimm, {herman, hcli, rm, mpal, dpeek, ps, dstaff, ttung, Facebook Inc. All requests queued at a client TCP-like flow-control across requests

39 Dealing with TCP Incast USENIX NSDI 2013 Scaling Memcache at Facebook Rajesh Nishtala, Hans Fugal, Steven Grimm, Marc Kwiatkowski, Herman Lee, Harry C. Li, Ryan McElroy, Mike Paleczny, Daniel Peek, Paul Saab, David Stafford, Tony Tung, Venkateshwaran Venkataramani {sgrimm, {herman, hcli, rm, mpal, dpeek, ps, dstaff, ttung, Facebook Inc. Wait time before request is scheduled (ms) th Percentile Median Window size

40 Dealing with TCP Incast USENIX NSDI 2013 Scaling Memcache at Facebook Rajesh Nishtala, Hans Fugal, Steven Grimm, Marc Kwiatkowski, Herman Lee, Harry C. Li, Ryan McElroy, Mike Paleczny, Daniel Peek, Paul Saab, David Stafford, Tony Tung, Venkateshwaran Venkataramani {sgrimm, {herman, hcli, rm, mpal, dpeek, ps, dstaff, ttung, UDP for GET requests: Facebook Inc. 0.25% failure rate but lower latency and connection-less operation Wait time before request is scheduled (ms) th Percentile Median Window size

41 Rate adaptation in video streaming

42 We want the highest video quality (c) copyright 2008, Blender Foundation / CC-BY-3.0

43 Without seeing this

44 Common solution approach Encode video in multiple bit-rates Estimate connection s available bandwidth Pick a video rate available bandwidth

45 Capacity estimation Network Buffer

46 Capacity estimation Capacity (Mbps) Time Network Buffer

47 Capacity estimation Capacity (Mbps) Time Network 1s

48 Capacity estimation Capacity (Mbps) Time Network 1s chunks at different bit-rates 1s

49 Capacity estimation Capacity (Mbps) Time Network Playing out

50 Capacity estimation Capacity (Mbps) Time Network Downloading Playing out

51 Capacity estimation Capacity (Mbps) Time Network Capacity < current rate decrease rate

52 Issue Encode video in multiple bit-rates Estimate connection s available bandwidth Pick a video rate available bandwidth

53 Estimating available capacity ACM SIGCOMM 2014 A Buffer-Based Approach to Rate Adaptation: Evidence from a Large Video Streaming Service Te-Yuan Huang, Ramesh Johari, Nick McKeown, Matthew Trunnell, Mark Watson Stanford University, Netflix {huangty,rjohari,nickm}@stanford.edu, {mtrunnell,watsonm}@netflix.com Avg. throughput over chunk download (kbps) Time(s)

54 Estimating available capacity ACM SIGCOMM 2014 A Buffer-Based Approach to Rate Adaptation: Evidence from a Large Video Streaming Service Te-Yuan Huang, Ramesh Johari, Nick McKeown, Matthew Trunnell, Mark Watson Stanford University, Netflix {huangty,rjohari,nickm}@stanford.edu, {mtrunnell,watsonm}@netflix.com A random sample of 300,000 Netflix sessions shows that roughly 10% of sessions Avg. throughput over chunk download (kbps) experience a median throughput less than half of the 95th percentile throughput Time(s)

55 Estimating available capacity ACM SIGCOMM 2014 A Buffer-Based Approach to Rate Adaptation: Evidence from a Large Video Streaming Service Te-Yuan Huang, Ramesh Johari, Nick McKeown, Matthew Trunnell, Mark Watson Stanford University, Netflix {huangty,rjohari,nickm}@stanford.edu, {mtrunnell,watsonm}@netflix.com A random sample of 300,000 Netflix sessions shows that roughly 10% of sessions Avg. throughput over chunk download (kbps) experience a median throughput less than half of the 95th percentile throughput % of rebuffers are unnecessary Time(s)

56 Capacity estimation Capacity (Mbps) Time Network Capacity < current rate decrease rate

57 Capacity estimation Network Decide based on the buffer alone?

58 Buffer-based adaptation Network Nearly full buffer large rate

59 Buffer-based adaptation Network Nearly empty buffer small rate

60 Buffer-based adaptation ACM SIGCOMM 2014 A Buffer-Based Approach to Rate Adaptation: Evidence from a Large Video Streaming Service Te-Yuan Huang, Ramesh Johari, Nick McKeown, Matthew Trunnell, Mark Watson Stanford University, Netflix {huangty,rjohari,nickm}@stanford.edu, {mtrunnell,watsonm}@netflix.com Next&Chunk s&video&rate& Next chunk s rate R max& & R min& Risky'' Area' Safe'from'' Unnecessary'' rebuffering' Buffer occupancy Playout&Buffer&Occupancy& B max&

61 Buffer-based adaptation ACM SIGCOMM 2014 A Buffer-Based Approach to Rate Adaptation: Evidence from a Large Video Streaming Service Te-Yuan Huang, Ramesh Johari, Nick McKeown, Matthew Trunnell, Mark Watson Stanford University, Netflix {huangty,rjohari,nickm}@stanford.edu, {mtrunnell,watsonm}@netflix.com Next&Chunk s&video&rate& Next chunk s rate R max& & R min& Risky'' Area' Safe'from'' Unnecessary'' rebuffering' High buffer: use Rmax Buffer occupancy Playout&Buffer&Occupancy& B max&

62 Buffer-based adaptation ACM SIGCOMM 2014 A Buffer-Based Approach to Rate Adaptation: Evidence from a Large Video Streaming Service Te-Yuan Huang, Ramesh Johari, Nick McKeown, Matthew Trunnell, Mark Watson Stanford University, Netflix {huangty,rjohari,nickm}@stanford.edu, {mtrunnell,watsonm}@netflix.com Next&Chunk s&video&rate& Next chunk s rate Low buffer: R max& & R min& Risky'' Area' Safe'from'' Unnecessary'' rebuffering' High buffer: use Rmax use Rmin Buffer occupancy Playout&Buffer&Occupancy& B max&

63 Problem: startup phase? Pick a rate based on immediate past throughput

64 Buffer-based adaptation ACM SIGCOMM 2014 A Buffer-Based Approach to Rate Adaptation: Evidence from a Large Video Streaming Service Te-Yuan Huang, Ramesh Johari, Nick McKeown, Matthew Trunnell, Mark Watson Stanford University, Netflix {huangty,rjohari,nickm}@stanford.edu, {mtrunnell,watsonm}@netflix.com Peak Hours Control& Normalized number of rebuffers per hour (%) BBA& Lower&bound& Hours in GMT

65 Buffer-based adaptation ACM SIGCOMM 2014 A Buffer-Based Approach to Rate Adaptation: Evidence from a Large Video Streaming Service Te-Yuan Huang, Ramesh Johari, Nick McKeown, Matthew Trunnell, Mark Watson Stanford University, Netflix {huangty,rjohari,nickm}@stanford.edu, {mtrunnell,watsonm}@netflix.com Video rate difference (kbps) 100 Peak Hours 50 0 BBA Control algorithm Hours in GMT

Data Center TCP (DCTCP)

Data Center TCP (DCTCP) Data Center TCP (DCTCP) Mohammad Alizadeh, Albert Greenberg, David A. Maltz, Jitendra Padhye Parveen Patel, Balaji Prabhakar, Sudipta Sengupta, Murari Sridharan Microsoft Research Stanford University 1

More information

Data Center TCP (DCTCP)

Data Center TCP (DCTCP) Data Center Packet Transport Data Center TCP (DCTCP) Mohammad Alizadeh, Albert Greenberg, David A. Maltz, Jitendra Padhye Parveen Patel, Balaji Prabhakar, Sudipta Sengupta, Murari Sridharan Cloud computing

More information

Congestion Control in Datacenters. Ahmed Saeed

Congestion Control in Datacenters. Ahmed Saeed Congestion Control in Datacenters Ahmed Saeed What is a Datacenter? Tens of thousands of machines in the same building (or adjacent buildings) Hundreds of switches connecting all machines What is a Datacenter?

More information

Data Center TCP(DCTCP)

Data Center TCP(DCTCP) Data Center TCP(DCTCP) Mohammad Alizadeh * +, Albert Greenberg *, David A. Maltz *, Jitendra Padhye *, Parveen Patel *, Balaji Prabhakar +, Sudipta Sengupta *, Murari Sridharan * * + Microsoft Research

More information

Data Center TCP (DCTCP)

Data Center TCP (DCTCP) Data Center TCP (DCTCP) Mohammad Alizadeh, Albert Greenberg, David A. Maltz, Jitendra Padhye Parveen Patel, Balaji Prabhakar, Sudipta Sengupta, Murari Sridharan Stanford University MicrosoD Research Case

More information

Cloud e Datacenter Networking

Cloud e Datacenter Networking Cloud e Datacenter Networking Università degli Studi di Napoli Federico II Dipartimento di Ingegneria Elettrica e delle Tecnologie dell Informazione DIETI Laurea Magistrale in Ingegneria Informatica Prof.

More information

Lecture 15: Datacenter TCP"

Lecture 15: Datacenter TCP Lecture 15: Datacenter TCP" CSE 222A: Computer Communication Networks Alex C. Snoeren Thanks: Mohammad Alizadeh Lecture 15 Overview" Datacenter workload discussion DC-TCP Overview 2 Datacenter Review"

More information

SARA: Segment Aware Rate Adaptation for DASH Video Services

SARA: Segment Aware Rate Adaptation for DASH Video Services SARA: Segment Aware Rate Adaptation for DASH Video Services, Venkatesh Tamarapalli*, Deep Medhi University of Missouri Kansas City * Indian Institute of Technology-Guwahati, India Overview Introduction

More information

D 2 P: A Distributed Deadline Propagation Approach to Tolerate Long-Tail Latency in Datacenters

D 2 P: A Distributed Deadline Propagation Approach to Tolerate Long-Tail Latency in Datacenters D 2 P: A Distributed Deadline Propagation Approach to Tolerate Long-Tail Latency in Datacenters Rui Ren, Jiuyue Ma, Xiufeng Sui, Yungang Bao Institute of Computing Technology, Chinese Academy Sciences

More information

Per-Packet Load Balancing in Data Center Networks

Per-Packet Load Balancing in Data Center Networks Per-Packet Load Balancing in Data Center Networks Yagiz Kaymak and Roberto Rojas-Cessa Abstract In this paper, we evaluate the performance of perpacket load in data center networks (DCNs). Throughput and

More information

Implementation of PI 2 Queuing Discipline for Classic TCP Traffic in ns-3

Implementation of PI 2 Queuing Discipline for Classic TCP Traffic in ns-3 Implementation of PI Queuing Discipline for Classic TCP Traffic in ns-3 Rohit P. Tahiliani, Hitesh Tewari School of Computer Science & Statistics, Trinity College Dublin Email: tahiliar@tcd.ie, htewari@cs.tcd.ie

More information

Tales of the Tail Hardware, OS, and Application-level Sources of Tail Latency

Tales of the Tail Hardware, OS, and Application-level Sources of Tail Latency Tales of the Tail Hardware, OS, and Application-level Sources of Tail Latency Jialin Li, Naveen Kr. Sharma, Dan R. K. Ports and Steven D. Gribble February 2, 2015 1 Introduction What is Tail Latency? What

More information

A Comparative Case Study of HTTP Adaptive Streaming Algorithms in Mobile Networks

A Comparative Case Study of HTTP Adaptive Streaming Algorithms in Mobile Networks Theodoros Karagkioules, Cyril Concolato, Dimitrios Tsilimantos and Stefan Valentin LTCI Telecom ParisTech Universite Paris-Saclay Mathematical and Algorithmic Sciences Lab France Research Center Huawei

More information

Fastpass A Centralized Zero-Queue Datacenter Network

Fastpass A Centralized Zero-Queue Datacenter Network Fastpass A Centralized Zero-Queue Datacenter Network Jonathan Perry Amy Ousterhout Hari Balakrishnan Devavrat Shah Hans Fugal Ideal datacenter network properties No current design satisfies all these properties

More information

Research Statement. 1 Thesis research: Efficient networked systems for datacenters with RPCs. Anuj Kalia

Research Statement. 1 Thesis research: Efficient networked systems for datacenters with RPCs. Anuj Kalia Research Statement Anuj Kalia With the seeming end of Moore s law, software does not automatically get faster every year: per-core CPU performance has stagnated, and core count increases slowly. My goal

More information

A Method Based on Data Fragmentation to Increase the Performance of ICTCP During Incast Congestion in Networks

A Method Based on Data Fragmentation to Increase the Performance of ICTCP During Incast Congestion in Networks A Method Based on Data Fragmentation to Increase the Performance of ICTCP During Incast Congestion in Networks Sneha Sebastian P G Scholar, Dept. of Computer Science and Engg. Amal Jyothi College of Engg.

More information

Packet Scheduling in Data Centers. Lecture 17, Computer Networks (198:552)

Packet Scheduling in Data Centers. Lecture 17, Computer Networks (198:552) Packet Scheduling in Data Centers Lecture 17, Computer Networks (198:552) Datacenter transport Goal: Complete flows quickly / meet deadlines Short flows (e.g., query, coordination) Large flows (e.g., data

More information

Got 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, 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 information

COSC4377. Network milestones

COSC4377. Network milestones Lecture 13 1969 1974 Network milestones 81 83 1988 1991 1996 1999 2003 2006 ARPANet TCP/IP TCP Backbone speed: Cutover to TCP/IP Tahoe HTTP 50 56kbps, ARPANet T1 NSFNet T3, NSFNet Network is exploding

More information

Proactive Incast Congestion Control in a Datacenter Serving Web Applications

Proactive Incast Congestion Control in a Datacenter Serving Web Applications Proactive Incast Congestion Control in a Datacenter Serving Web Applications Haoyu Wang and Haiying Shen Department of Computer Science University of Virginia, Charlottesville, VA 2293, USA Email: {hw8c,

More information

DeTail Reducing the Tail of Flow Completion Times in Datacenter Networks. David Zats, Tathagata Das, Prashanth Mohan, Dhruba Borthakur, Randy Katz

DeTail Reducing the Tail of Flow Completion Times in Datacenter Networks. David Zats, Tathagata Das, Prashanth Mohan, Dhruba Borthakur, Randy Katz DeTail Reducing the Tail of Flow Completion Times in Datacenter Networks David Zats, Tathagata Das, Prashanth Mohan, Dhruba Borthakur, Randy Katz 1 A Typical Facebook Page Modern pages have many components

More information

RAMPS REAL-TIME APPLICATION MEASUREMENT and PLACEMENT SYSTEM

RAMPS REAL-TIME APPLICATION MEASUREMENT and PLACEMENT SYSTEM RAMPS REAL-TIME APPLICATION MEASUREMENT and PLACEMENT SYSTEM Pratheek Nagaraj, Manali Naik, Jenny Shen pnagaraj@mit.edu, mnaik@mit.edu, jenshen@mit.edu May 9, 2014 6.033 Design Project 2 Report 1 Introduction

More information

Master Course Computer Networks IN2097

Master Course Computer Networks IN2097 Chair for Network Architectures and Services Prof. Carle Department of Computer Science TU München Master Course Computer Networks IN2097 Prof. Dr.-Ing. Georg Carle Chair for Network Architectures and

More information

A Proxy-based Query Aggregation Method for Distributed Key-Value Stores

A Proxy-based Query Aggregation Method for Distributed Key-Value Stores A Proxy-based Query Aggregation Method for Distributed Key-Value Stores Daichi Kawanami, Masanari Kamoshita, Ryota Kawashima and Hiroshi Matsuo Nagoya Institute of Technology, in Nagoya, Aichi, 466-8555,

More information

On the Effectiveness of CoDel in Data Centers

On the Effectiveness of CoDel in Data Centers On the Effectiveness of in Data Centers Saad Naveed Ismail, Hasnain Ali Pirzada, Ihsan Ayyub Qazi Computer Science Department, LUMS Email: {14155,15161,ihsan.qazi}@lums.edu.pk Abstract Large-scale data

More information

CS 598: Advanced Internet

CS 598: Advanced Internet CS 598: Advanced Internet Lecture 3: TCP / IP Brighten Godfrey pbg@illinois.edu Fall 2009 1 Today Announcements A few more project ideas Cerf and Kahn: TCP / IP Clark: TCP / IP design philosophy 2 Announcements

More information

Presented by: Fabián E. Bustamante

Presented by: Fabián E. Bustamante Presented by: Fabián E. Bustamante A. Nikravesh, H. Yao, S. Xu, D. Choffnes*, Z. Morley Mao Mobisys 2015 *Based on the authors slides Mobile apps are increasingly popular Mobile platforms is the dominant

More information

Advanced Computer Networks. Datacenter TCP

Advanced Computer Networks. Datacenter TCP Advanced Computer Networks 263 3501 00 Datacenter TCP Spring Semester 2017 1 Oriana Riva, Department of Computer Science ETH Zürich Today Problems with TCP in the Data Center TCP Incast TPC timeouts Improvements

More information

Confused, Timid, and Unstable: Picking a Video Streaming Rate is Hard

Confused, 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 information

Confused, Timid, and Unstable: Picking a Video Streaming Rate is Hard

Confused, Timid, and Unstable: Picking a Video Streaming Rate is Hard Confused, Timid, and Unstable: Picking a Video Streaming Rate is Hard Araz Jangiaghdam Seminar Networks and Distributed Systems School of Engineering and Sciences Jacobs University Bremen Campus Ring 1,

More information

TCP. The TCP Protocol. TCP Header Format. TCP Flow Control. TCP Congestion Control Datacenter TCP

TCP. The TCP Protocol. TCP Header Format. TCP Flow Control. TCP Congestion Control Datacenter TCP The TCP Protocol TCP TCP Congestion Control Datacenter TCP Frame format Connection management Flow control TCP reliable data transfer Congestion control TCP Header Format TCP Flow Control HL 32 bits Source

More information

Multimedia-unfriendly TCP Congestion Control and Home Gateway Queue Management

Multimedia-unfriendly TCP Congestion Control and Home Gateway Queue Management Multimedia-unfriendly TCP Congestion Control and Home Gateway Queue Management Lawrence Stewart α, David Hayes α, Grenville Armitage α, Michael Welzl β, Andreas Petlund β α Centre for Advanced Internet

More information

DAQ: Deadline-Aware Queue Scheme for Scheduling Service Flows in Data Centers

DAQ: Deadline-Aware Queue Scheme for Scheduling Service Flows in Data Centers DAQ: Deadline-Aware Queue Scheme for Scheduling Service Flows in Data Centers Cong Ding and Roberto Rojas-Cessa Abstract We propose a scheme to schedule the transmission of data center traffic to guarantee

More information

Providing Multi-tenant Services with FPGAs: Case Study on a Key-Value Store

Providing Multi-tenant Services with FPGAs: Case Study on a Key-Value Store Zsolt István *, Gustavo Alonso, Ankit Singla Systems Group, Computer Science Dept., ETH Zürich * Now at IMDEA Software Institute, Madrid Providing Multi-tenant Services with FPGAs: Case Study on a Key-Value

More information

/ Cloud Computing. Recitation 7 October 10, 2017

/ Cloud Computing. Recitation 7 October 10, 2017 15-319 / 15-619 Cloud Computing Recitation 7 October 10, 2017 Overview Last week s reflection Project 3.1 OLI Unit 3 - Module 10, 11, 12 Quiz 5 This week s schedule OLI Unit 3 - Module 13 Quiz 6 Project

More information

TCP Incast problem Existing proposals

TCP Incast problem Existing proposals TCP Incast problem & Existing proposals Outline The TCP Incast problem Existing proposals to TCP Incast deadline-agnostic Deadline-Aware Datacenter TCP deadline-aware Picasso Art is TLA 1. Deadline = 250ms

More information

arxiv: v1 [cs.ni] 16 Jun 2013

arxiv: v1 [cs.ni] 16 Jun 2013 Low Latency via Redundancy Ashish Vulimiri, P. Brighten Godfrey, Radhika Mittal 2, Justine Sherry 2, Sylvia Ratnasamy 2, and Scott Shenker 2,3 UIUC 2 UC Berkeley 3 ICSI arxiv:36.377v [cs.ni] 6 Jun 23 ABSTRACT

More information

Improving Multipath TCP for Latency Sensitive Flows in the Cloud

Improving Multipath TCP for Latency Sensitive Flows in the Cloud 2016 5th IEEE International Conference on Cloud Networking Improving Multipath TCP for Latency Sensitive Flows in the Cloud Wei Wang,Liang Zhou,Yi Sun Institute of Computing Technology, CAS, University

More information

Oboe: Auto-tuning Video ABR Algorithms to Network Conditions

Oboe: Auto-tuning Video ABR Algorithms to Network Conditions Oboe: Auto-tuning Video ABR Algorithms to Network Conditions Zahaib Akhtar, Yun Seong Nam, Ramesh Govindan, Sanjay Rao, Jessica Chen, Ethan Katz-Bassett, Bruno Ribeiro, Jibin Zhan, Hui Zhang : Co-primary

More information

CSE 123A Computer Networks

CSE 123A Computer Networks CSE 123A Computer Networks Winter 2005 Lecture 14 Congestion Control Some images courtesy David Wetherall Animations by Nick McKeown and Guido Appenzeller The bad news and the good news The bad news: new

More information

A Mechanism Achieving Low Latency for Wireless Datacenter Applications

A Mechanism Achieving Low Latency for Wireless Datacenter Applications Computer Science and Information Systems 13(2):639 658 DOI: 10.2298/CSIS160301020H A Mechanism Achieving Low Latency for Wireless Datacenter Applications Tao Huang 1,2, Jiao Zhang 1, and Yunjie Liu 2 1

More information

TAIL LATENCY AND PERFORMANCE AT SCALE

TAIL LATENCY AND PERFORMANCE AT SCALE TAIL LATENCY AND PERFORMANCE AT SCALE George Porter May 21, 2018 ATTRIBUTION These slides are released under an Attribution-NonCommercial-ShareAlike 3.0 Unported (CC BY-NC-SA 3.0) Creative Commons license

More information

Congestion Control In the Network

Congestion Control In the Network Congestion Control In the Network Brighten Godfrey cs598pbg September 9 2010 Slides courtesy Ion Stoica with adaptation by Brighten Today Fair queueing XCP Announcements Problem: no isolation between flows

More information

Packet Switch Architectures Part 2

Packet Switch Architectures Part 2 Packet Switch Architectures Part Adopted from: Sigcomm 99 Tutorial, by Nick McKeown and Balaji Prabhakar, Stanford University Slides used with permission from authors. 999-000. All rights reserved by authors.

More information

Episode 5. Scheduling and Traffic Management

Episode 5. Scheduling and Traffic Management Episode 5. Scheduling and Traffic Management Part 3 Baochun Li Department of Electrical and Computer Engineering University of Toronto Outline What is scheduling? Why do we need it? Requirements of a scheduling

More information

CSE 124: TAIL LATENCY AND PERFORMANCE AT SCALE. George Porter November 27, 2017

CSE 124: TAIL LATENCY AND PERFORMANCE AT SCALE. George Porter November 27, 2017 CSE 124: TAIL LATENCY AND PERFORMANCE AT SCALE George Porter November 27, 2017 ATTRIBUTION These slides are released under an Attribution-NonCommercial-ShareAlike 3.0 Unported (CC BY-NC-SA 3.0) Creative

More information

Resource Sharing or Designing Access Network For Low Cost.

Resource Sharing or Designing Access Network For Low Cost. Resource Sharing or Designing Access Network For Low Cost www.broadcom.com Traffic Percentile What Drives Demand for Bandwidth? Usage In North America, the median usage is 4 GB per month, while the mean

More information

Deconstructing Datacenter Packet Transport

Deconstructing Datacenter Packet Transport Deconstructing Datacenter Packet Transport Mohammad Alizadeh, Shuang Yang, Sachin Katti, Nick McKeown, Balaji Prabhakar, and Scott Shenker Stanford University U.C. Berkeley / ICSI {alizade, shyang, skatti,

More information

Factors Affecting Performance of Web Flows in Cellular Networks

Factors Affecting Performance of Web Flows in Cellular Networks in Cellular Networks Ermias A. Walelgne, Kim Setälä, Vaibhav Bajpai, Stefan Neumeier, Jukka Manner, Jörg Ott May 15, 2018 - FP Networking, Zurich ntroduction ntroduction ntroduction Motivation 99% of the

More information

CS519: Computer Networks. Lecture 5, Part 4: Mar 29, 2004 Transport: TCP congestion control

CS519: Computer Networks. Lecture 5, Part 4: Mar 29, 2004 Transport: TCP congestion control : Computer Networks Lecture 5, Part 4: Mar 29, 2004 Transport: TCP congestion control TCP performance We ve seen how TCP the protocol works Sequencing, receive window, connection setup and teardown And

More information

Downton Abbey Without the Hiccups: Buffer-Based Rate Adaptation for HTTP Video Streaming

Downton Abbey Without the Hiccups: Buffer-Based Rate Adaptation for HTTP Video Streaming Downton Abbey Without the Hiccups: Buffer-Based Rate Adaptation for HTTP Video Streaming Te-Yuan Huang Ramesh Johari Nick McKeown Stanford University {huangty,ramesh.johari, nickm}@stanford.edu ABSTRACT

More information

Preprint. Until published, please cite as:

Preprint. Until published, please cite as: Preprint. To appear in Proc. Third IEEE Workshop on Telecommunications Standards: From Research to Standards. Part of IEEE Globecom 204 Until published, please cite

More information

Factors Affecting Performance of Web Flows in Cellular Networks

Factors Affecting Performance of Web Flows in Cellular Networks in Cellular Networks Ermias A. Walelgne, Kim Setälä, Vaibhav Bajpai, Stefan Neumeier, Jukka Manner, Jörg Ott October 17, 2018 - RIPE 77, Amsterdam Introduction Introduction Introduction Motivation 99%

More information

SECURED SOCIAL TUBE FOR VIDEO SHARING IN OSN SYSTEM

SECURED SOCIAL TUBE FOR VIDEO SHARING IN OSN SYSTEM ABSTRACT: SECURED SOCIAL TUBE FOR VIDEO SHARING IN OSN SYSTEM J.Priyanka 1, P.Rajeswari 2 II-M.E(CS) 1, H.O.D / ECE 2, Dhanalakshmi Srinivasan Engineering College, Perambalur. Recent years have witnessed

More information

CS 344/444 Computer Network Fundamentals Final Exam Solutions Spring 2007

CS 344/444 Computer Network Fundamentals Final Exam Solutions Spring 2007 CS 344/444 Computer Network Fundamentals Final Exam Solutions Spring 2007 Question 344 Points 444 Points Score 1 10 10 2 10 10 3 20 20 4 20 10 5 20 20 6 20 10 7-20 Total: 100 100 Instructions: 1. Question

More information

Lecture 18: Video Streaming

Lecture 18: Video Streaming MIT 6.829: Computer Networks Fall 2017 Lecture 18: Video Streaming Scribe: Zhihong Luo, Francesco Tonolini 1 Overview This lecture is on a specific networking application: video streaming. In particular,

More information

ISSN: (Online) Volume 3, Issue 6, June 2015 International Journal of Advance Research in Computer Science and Management Studies

ISSN: (Online) Volume 3, Issue 6, June 2015 International Journal of Advance Research in Computer Science and Management Studies ISSN: 2321-7782 (Online) Volume 3, Issue 6, June 2015 International Journal of Advance Research in Computer Science and Management Studies Research Article / Survey Paper / Case Study Available online

More information

Announcements. Congestion Control. A few words from Panda. Congestion Control Review. Load and Delay. Caveat: In this lecture

Announcements. Congestion Control. A few words from Panda. Congestion Control Review. Load and Delay. Caveat: In this lecture Announcements Project 3 is out! Congestion Control EE Fall 0 Scott Shenker http://inst.eecs.berkeley.edu/~ee/ Materials with thanks to Jennifer Rexford, Ion Stoica, Vern Paxson and other colleagues at

More information

Congestion Control. EE122 Fall 2012 Scott Shenker

Congestion Control. EE122 Fall 2012 Scott Shenker Congestion Control EE122 Fall 2012 Scott Shenker http://inst.eecs.berkeley.edu/~ee122/ Materials with thanks to Jennifer Rexford, Ion Stoica, Vern Paxson and other colleagues at Princeton and UC Berkeley

More information

Information-Agnostic Flow Scheduling for Commodity Data Centers

Information-Agnostic Flow Scheduling for Commodity Data Centers Information-Agnostic Flow Scheduling for Commodity Data Centers Wei Bai, Li Chen, Kai Chen, Dongsu Han (KAIST), Chen Tian (NJU), Hao Wang Sing Group @ Hong Kong University of Science and Technology USENIX

More information

/ Cloud Computing. Recitation 9 March 17th and 19th, 2015

/ Cloud Computing. Recitation 9 March 17th and 19th, 2015 15-319 / 15-619 Cloud Computing Recitation 9 March 17th and 19th, 2015 Overview Administrative issues Tagging, 15619Project Last week s reflection Project 3.2 This week s schedule Project 3.3 Unit 4 -

More information

Resolving Tensions between Congestion Control Scaling Requirements

Resolving Tensions between Congestion Control Scaling Requirements Discussion paper Resolving Tensions between Congestion Control Scaling Requirements Bob Briscoe Koen De Schepper 11 Jul 2017 Abstract Low Latency, Low Loss Scalable throughput (L4S) is being proposed as

More information

Aemon: Information-agnostic Mix-flow Scheduling in Data Center Networks

Aemon: Information-agnostic Mix-flow Scheduling in Data Center Networks Tao Wang Huazhong University of Science and Technology Hong Xu City University of Hong Kong Fangming Liu Huazhong University of Science and Technology ABSTRACT Data center networks carry a mix of flows,

More information

Lecture 22: Buffering & Scheduling. CSE 123: Computer Networks Alex C. Snoeren

Lecture 22: Buffering & Scheduling. CSE 123: Computer Networks Alex C. Snoeren Lecture 22: Buffering & Scheduling CSE 123: Computer Networks Alex C. Snoeren Lecture 23 Overview Buffer Management FIFO RED Traffic Policing/Scheduling 2 Key Router Challenges Buffer management: which

More information

Distributed Systems Exam 1 Review Paul Krzyzanowski. Rutgers University. Fall 2016

Distributed Systems Exam 1 Review Paul Krzyzanowski. Rutgers University. Fall 2016 Distributed Systems 2015 Exam 1 Review Paul Krzyzanowski Rutgers University Fall 2016 1 Question 1 Why did the use of reference counting for remote objects prove to be impractical? Explain. It s not fault

More information

Low Latency via Redundancy

Low Latency via Redundancy Low Latency via Redundancy Ashish Vulimiri, Philip Brighten Godfrey, Radhika Mittal, Justine Sherry, Sylvia Ratnasamy, Scott Shenker Presenter: Meng Wang 2 Low Latency Is Important Injecting just 400 milliseconds

More information

Packet-Based Load Balancing in Data Center Networks

Packet-Based Load Balancing in Data Center Networks Packet-Based Load Balancing in Data Center Networks Yagiz Kaymak and Roberto Rojas-Cessa Networking Research Laboratory, Department of Electrical and Computer Engineering, New Jersey Institute of Technology,

More information

This is a repository copy of Efficient UDP-Based Congestion Aware Transport for Data Center Traffic.

This is a repository copy of Efficient UDP-Based Congestion Aware Transport for Data Center Traffic. This is a repository copy of Efficient -Based Congestion Aware Transport for Data Center Traffic. White Rose Research Online URL for this paper: http://eprints.whiterose.ac.uk/89456/ Version: Accepted

More information

Xbox360 matchmaking & predictions

Xbox360 matchmaking & predictions Sharad Agarwal Chris Butcher (bungie) Youngki Lee (intern) Jitu Padhye (microsoft research) (microsoft research) Xbox360 matchmaking & predictions Xbox360 Internet games most games P2P Xbox Live server

More information

CS244a: An Introduction to Computer Networks

CS244a: An Introduction to Computer Networks CS244a: n Introduction to Computer Networks Handout 7: Congestion Control Nick McKeown Professor of Electrical Engineering and Computer Science, Stanford University nickm@stanford.edu http://www.stanford.edu/~nickm

More information

Buffer Sizing in a Combined Input Output Queued (CIOQ) Switch

Buffer Sizing in a Combined Input Output Queued (CIOQ) Switch Buffer Sizing in a Combined Input Output Queued (CIOQ) Switch Neda Beheshti, Nick Mckeown Stanford University Abstract In all internet routers buffers are needed to hold packets during times of congestion.

More information

CHOKe - A simple approach for providing Quality of Service through stateless approximation of fair queueing. Technical Report No.

CHOKe - A simple approach for providing Quality of Service through stateless approximation of fair queueing. Technical Report No. CHOKe - A simple approach for providing Quality of Service through stateless approximation of fair queueing Rong Pan Balaji Prabhakar Technical Report No.: CSL-TR-99-779 March 1999 CHOKe - A simple approach

More information

TCP Bandwidth Allocation for Virtual Networks

TCP Bandwidth Allocation for Virtual Networks TCP Bandwidth Allocation for Virtual Networks Shuoh-Ren Tsai Department of Computer and Communication Engineering, National Kaohsiung First University of Science and Technology, Taiwan shawn@nkfust.edu.tw

More information

Page 1. Outline / Computer Networking : 1 st Generation Commercial PC/Packet Video Technologies

Page 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 information

Measuring Over-the-Top Video Quality

Measuring Over-the-Top Video Quality Contents Executive Summary... 1 Overview... 2 Progressive Video Primer: The Layers... 2 Adaptive Video Primer: The Layers... 3 Measuring the Stall: A TCP Primer... 4 Conclusion... 5 Questions to Ask of

More information

Mobile Video Streaming with Video Quality and Streaming Performance Guarantees

Mobile Video Streaming with Video Quality and Streaming Performance Guarantees Mobile Video Streaming with Video Quality and Streaming Performance Guarantees Victor K. C. Wu, Yan Liu, and Jack Y. B. Lee Department of Information Engineering The Chinese University of Hong Kong Hong

More information

SPDY - A Web Protocol. Mike Belshe Velocity, Dec 2009

SPDY - A Web Protocol. Mike Belshe Velocity, Dec 2009 SPDY - A Web Protocol Mike Belshe Velocity, Dec 2009 What is SPDY? Concept SPDY is an application layer protocol for transporting content over the web with reduced latency. Basic Features 1. Multiplexed

More information

ETSF10 Internet Protocols Transport Layer Protocols

ETSF10 Internet Protocols Transport Layer Protocols ETSF10 Internet Protocols Transport Layer Protocols 2012, Part 2, Lecture 2.1 Kaan Bür, Jens Andersson Transport Layer Protocols Process-to-process delivery [ed.4 ch.23.1] [ed.5 ch.24.1] Transmission Control

More information

Real-Time Protocol (RTP)

Real-Time Protocol (RTP) Real-Time Protocol (RTP) Provides standard packet format for real-time application Typically runs over UDP Specifies header fields below Payload Type: 7 bits, providing 128 possible different types of

More information

Attaining the Promise and Avoiding the Pitfalls of TCP in the Datacenter. Glenn Judd Morgan Stanley

Attaining the Promise and Avoiding the Pitfalls of TCP in the Datacenter. Glenn Judd Morgan Stanley Attaining the Promise and Avoiding the Pitfalls of TCP in the Datacenter Glenn Judd Morgan Stanley 1 Introduction Datacenter computing pervasive Beyond the Internet services domain BigData, Grid Computing,

More information

ED STIC - Proposition de Sujets de Thèse. pour la campagne d'allocation de thèses 2017

ED STIC - Proposition de Sujets de Thèse. pour la campagne d'allocation de thèses 2017 ED STIC - Proposition de Sujets de Thèse pour la campagne d'allocation de thèses 2017 Axe Sophi@Stic : Titre du sujet : aucun Joint Application and Network Optimization of Big Data Analytics Mention de

More information

Quickly Starting Media Streams Using QUIC

Quickly Starting Media Streams Using QUIC Quickly Starting Media Streams Using QUIC Packet Video Workshop 2018 Şevket Arısu and Ali C. Begen Agenda Motivation and our goal Previous work and our contributions Approach, setup and evaluation Results

More information

Communication using Multiple Wireless Interfaces

Communication using Multiple Wireless Interfaces Communication using Multiple Interfaces Kameswari Chebrolu and Ramesh Rao Department of ECE University of California, San Diego Abstract With the emergence of different wireless technologies, a mobile

More information

/ Cloud Computing. Recitation 10 March 22nd, 2016

/ Cloud Computing. Recitation 10 March 22nd, 2016 15-319 / 15-619 Cloud Computing Recitation 10 March 22nd, 2016 Overview Administrative issues Office Hours, Piazza guidelines Last week s reflection Project 3.3, OLI Unit 4, Module 15, Quiz 8 This week

More information

SamKnows test methodology

SamKnows test methodology SamKnows test methodology Download and Upload (TCP) Measures the download and upload speed of the broadband connection in bits per second. The transfer is conducted over one or more concurrent HTTP connections

More information

An experimental study of the learnability of congestion control

An experimental study of the learnability of congestion control An experimental study of the learnability of congestion control Anirudh Sivaraman, Keith Winstein, Pratiksha Thaker, Hari Balakrishnan MIT CSAIL http://web.mit.edu/remy/learnability August 31, 2014 1 /

More information

DIBS: Just-in-time congestion mitigation for Data Centers

DIBS: Just-in-time congestion mitigation for Data Centers DIBS: Just-in-time congestion mitigation for Data Centers Kyriakos Zarifis, Rui Miao, Matt Calder, Ethan Katz-Bassett, Minlan Yu, Jitendra Padhye University of Southern California Microsoft Research Summary

More information

Internet Design Principles and Architecture

Internet Design Principles and Architecture Internet Design Principles and Architecture Venkat Padmanabhan Microsoft Research 2 April 2001 Venkat Padmanabhan 1 Lecture Outline A brief history of the Internet How is the Internet different from the

More information

Performance of UMTS Radio Link Control

Performance of UMTS Radio Link Control Performance of UMTS Radio Link Control Qinqing Zhang, Hsuan-Jung Su Bell Laboratories, Lucent Technologies Holmdel, NJ 77 Abstract- The Radio Link Control (RLC) protocol in Universal Mobile Telecommunication

More information

Transparent Transmission Segmentation for Software-Defined Networks NetSoft 2017

Transparent Transmission Segmentation for Software-Defined Networks NetSoft 2017 Transparent Transmission Segmentation for Software-Defined Networks NetSoft 2017 Andreas Schmidt, Thorsten Herfet Telecommunications Lab Saarland Informatics Campus - Saarbrücken July 04, 2017 Motivation

More information

T Computer Networks II Data center networks

T Computer Networks II Data center networks T-110.5116 Computer Networks II Data center networks 29.9.2014 Matti Siekkinen (Sources: S. Kandula et al.: The Nature of Datacenter: measurements & analysis, A. Greenberg: Networking The Cloud, M. Alizadeh

More information

IX: A Protected Dataplane Operating System for High Throughput and Low Latency

IX: 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 information

Low Latency via Redundancy

Low Latency via Redundancy Low Latency via Redundancy Ashish Vulimiri UIUC vulimir@illinois.edu P. Brighten Godfrey UIUC pbg@illinois.edu Radhika Mittal UC Berkeley radhika@eecs.berkeley.edu Justine Sherry UC Berkeley justine@eecs.berkeley.edu

More information

/ Cloud Computing. Recitation 8 October 18, 2016

/ Cloud Computing. Recitation 8 October 18, 2016 15-319 / 15-619 Cloud Computing Recitation 8 October 18, 2016 1 Overview Administrative issues Office Hours, Piazza guidelines Last week s reflection Project 3.2, OLI Unit 3, Module 13, Quiz 6 This week

More information

Salsify: Low-Latency Network Video Through Tighter Integration Between a Video Codec and a Transport Protocol

Salsify: Low-Latency Network Video Through Tighter Integration Between a Video Codec and a Transport Protocol Salsify: Low-Latency Network Video Through Tighter Integration Between a Video Codec and a Transport Protocol Sadjad Fouladi, John Emmons, Emre Orbay, Catherine Wu, Riad S. Wahby, Keith Winstein https://snr.stanford.edu/salsify

More information

Network Layer Enhancements

Network Layer Enhancements Network Layer Enhancements EECS 122: Lecture 14 Department of Electrical Engineering and Computer Sciences University of California Berkeley Today We have studied the network layer mechanisms that enable

More information

Data Center Performance

Data Center Performance Data Center Performance George Porter CSE 124 Feb 15, 2017 *Includes material taken from Barroso et al., 2013, UCSD 222a, and Cedric Lam and Hong Liu (Google) Part 1: Partitioning work across many servers

More information

Advanced Topics in Congestion Control

Advanced Topics in Congestion Control Advanced Topics in Congestion Control EE122 Fall 2012 Scott Shenker http://inst.eecs.berkeley.edu/~ee122/ Materials with thanks to Jennifer Rexford, Ion Stoica, Vern Paxson and other colleagues at Princeton

More information

Router Design: Table Lookups and Packet Scheduling EECS 122: Lecture 13

Router Design: Table Lookups and Packet Scheduling EECS 122: Lecture 13 Router Design: Table Lookups and Packet Scheduling EECS 122: Lecture 13 Department of Electrical Engineering and Computer Sciences University of California Berkeley Review: Switch Architectures Input Queued

More information

Transport Layer Protocols TCP

Transport Layer Protocols TCP Transport Layer Protocols TCP Gail Hopkins Introduction Features of TCP Packet loss and retransmission Adaptive retransmission Flow control Three way handshake Congestion control 1 Common Networking Issues

More information