Yiting Xia, T. S. Eugene Ng Rice University

Slides:



Advertisements
Similar presentations
Data Center Networking with Multipath TCP
Advertisements

COMPUTER NETWORK TOPOLOGIES
Computer Network Topologies
Software-defined networking: Change is hard Ratul Mahajan with Chi-Yao Hong, Rohan Gandhi, Xin Jin, Harry Liu, Vijay Gill, Srikanth Kandula, Mohan Nanduri,
A Novel 3D Layer-Multiplexed On-Chip Network
PortLand: A Scalable Fault-Tolerant Layer 2 Data Center Network Fabric. Presented by: Vinuthna Nalluri Shiva Srivastava.
Cognitive Publish/Subscribe for Heterogeneous Clouds Šarūnas Girdzijauskas, Swedish Institute of Computer Science (SICS) Joint work with:
Improving Datacenter Performance and Robustness with Multipath TCP Costin Raiciu, Sebastien Barre, Christopher Pluntke, Adam Greenhalgh, Damon Wischik,
Data and Computer Communications Ninth Edition by William Stallings Chapter 12 – Routing in Switched Data Networks Data and Computer Communications, Ninth.
60 GHz Flyways: Adding multi-Gbps wireless links to data centers
Reconfigurable Network Topologies at Rack Scale
Datacenter Network Topologies
ProActive Routing In Scalable Data Centers with PARIS Joint work with Dushyant Arora + and Jennifer Rexford* + Arista Networks *Princeton University Theophilus.
FireFly: A Reconfigurable Wireless Datacenter Fabric using Free-Space Optics Navid Hamedazimi, Zafar Qazi, Himanshu Gupta, Vyas Sekar, Samir Das, Jon.
A Scalable, Commodity Data Center Network Architecture Mohammad Al-Fares, Alexander Loukissas, Amin Vahdat Presented by Gregory Peaker and Tyler Maclean.
Ji-Yong Shin * Bernard Wong +, and Emin Gün Sirer * * Cornell University + University of Waterloo 2 nd ACM Symposium on Cloud ComputingOct 27, 2011 Small-World.
A Scalable, Commodity Data Center Network Architecture Mohammad AI-Fares, Alexander Loukissas, Amin Vahdat Presented by Ye Tao Feb 6 th 2013.
A Scalable, Commodity Data Center Network Architecture
Layer-3 Routing Natawut Nupairoj, Ph.D. Department of Computer Engineering Chulalongkorn University.
Quasi Fat Trees for HPC Clouds and their Fault-Resilient Closed-Form Routing Technion - EE Department; *and Mellanox Technologies Eitan Zahavi* Isaac Keslassy.
Network Support for Cloud Services Lixin Gao, UMass Amherst.
Capacity Scaling with Multiple Radios and Multiple Channels in Wireless Mesh Networks Oguz GOKER.
Network Aware Resource Allocation in Distributed Clouds.
Routing & Architecture
LAN Switching and Wireless – Chapter 1 Vilina Hutter, Instructor
Intro to Network Design
VLSI Physical Design: From Graph Partitioning to Timing Closure Chapter 6: Detailed Routing © KLMH Lienig 1 What Makes a Design Difficult to Route Charles.
© 2006 Cisco Systems, Inc. All rights reserved.Cisco Public 1 Version 4.0 Introducing Network Design Concepts Designing and Supporting Computer Networks.
VL2: A Scalable and Flexible Data Center Network Albert Greenberg, James R. Hamilton, Navendu Jain, Srikanth Kandula, Changhoon Kim, Parantap Lahiri, David.
© 2008 Cisco Systems, Inc. All rights reserved.Cisco ConfidentialPresentation_ID 1 Chapter 1: Introduction to Scaling Networks Scaling Networks.
Department of Computer Science A Scalable, Commodity Data Center Network Architecture Mohammad Al-Fares Alexander Loukissas Amin Vahdat SIGCOMM’08 Reporter:
Software Defined Networks for Dynamic Datacenter and Cloud Environments.
© 1999, Cisco Systems, Inc. 1-1 Chapter 2 Overview of a Campus Network © 1999, Cisco Systems, Inc.
© 2006 Cisco Systems, Inc. All rights reserved.Cisco PublicITE I Chapter 6 1 Introducing Network Design Concepts Designing and Supporting Computer Networks.
Dual Centric Data Center Network Architectures DAWEI LI, JIE WU (TEMPLE UNIVERSITY) ZHIYONG LIU, AND FA ZHANG (CHINESE ACADEMY OF SCIENCES) ICPP 2015.
Performance, Cost, and Energy Evaluation of Fat H-Tree: A Cost-Efficient Tree-Based On-Chip Network Hiroki Matsutani (Keio Univ, JAPAN) Michihiro Koibuchi.
Subways: A Case for Redundant, Inexpensive Data Center Edge Links Vincent Liu, Danyang Zhuo, Simon Peter, Arvind Krishnamurthy, Thomas Anderson University.
Logically Centralized? State Distribution Trade-offs in Software Defined Networks.
R2C2: A Network Stack for Rack-scale Computers Paolo Costa, Hitesh Ballani, Kaveh Razavi, Ian Kash Microsoft Research Cambridge EECS 582 – W161.
Wireless sensor and actor networks: research challenges Ian. F. Akyildiz, Ismail H. Kasimoglu
XFabric: a Reconfigurable In-Rack Network for Rack-Scale Computers Sergey Legtchenko, Nicholas Chen, Daniel Cletheroe, Antony Rowstron, Hugh Williams,
VL2: A Scalable and Flexible Data Center Network
Data Center Architectures
Multi Node Label Routing – A layer 2.5 routing protocol
Instructor Materials Chapter 1: LAN Design
Data Center Network Topologies II
Architecture and Algorithms for an IEEE 802
Lecture 2: Leaf-Spine and PortLand Networks
Interconnect Networks
Data Center Network Architectures
A Survey of Data Center Network Architectures By Obasuyi Edokpolor
ECE 544: Traffic engineering (supplement)
Improving Datacenter Performance and Robustness with Multipath TCP
Datacenter Interconnection Network Design
Improving Datacenter Performance and Robustness with Multipath TCP
FAR: A Fault-avoidance Routing Method for Data Center Networks with Regular Topology Please send.
NTHU CS5421 Cloud Computing
A Scalable, Commodity Data Center Network Architecture
BCube: A High Performance, Server-centric Network Architecture for Modular Data Centers Chuanxiong Guo1, Guohan Lu1, Dan Li1, Haitao Wu1, Xuan Zhang2,
Network Layer Path Determination.
Chuanxiong Guo, Haitao Wu, Kun Tan,
Dingming Wu+, Yiting Xia+*, Xiaoye Steven Sun+,
Degree-aware Hybrid Graph Traversal on FPGA-HMC Platform
Physical Network Topology
Jellyfish: Networking Data Centers Randomly
Internet and Web Simple client-server model
Data Center Architectures
Replica Placement Heuristics of Application-level Multicast
In-network computation
Towards Predictable Datacenter Networks
Presentation transcript:

Yiting Xia, T. S. Eugene Ng Rice University Flat-tree A Convertible Data Center Network Architecture from Clos to Random Graph Yiting Xia, T. S. Eugene Ng Rice University

Clos Topology 3-stage folded Clos - standard data center network architecture Core Switches Aggregation Switches Edge Switches Pods 1

Clos Topology Implementation friendly - central wiring - flexible scale and oversubscription - Pod modular design Suboptimal performance - long paths - congested network core 2

Random Graph Good performance Hard to implement - low average path length - rich bandwidth - optimal throughput for uniform traffic Hard to implement - neighbor-to-neighbor wiring complicated [Jellyfish NSDI’12] 3

Can we combine the best of both worlds? Why fixed topology? Tree Network Flat Network vs. Easy implementation Good performance Can we combine the best of both worlds? 4

Why fixed topology? Fluid data center traffic Fat-tree SIGCOMM’08 BCube SIGCOMM’09 DCell SIGCOMM’08 HyperX SC’09 Easy implementation Good performance Fluid data center traffic - each topology has sweet spots - one-size-fit-all topology impossible Cloud service constantly changing - fixed topology not adaptive to new demands 5

Convertible Network Flat-tree Tree Network Flat Network 6

Design Highlights Flat-tree starts from a Clos network and converts the topology to approximate random graphs. Challenges: Relocate servers from edge switches to aggregation and core switches Connect edge and core switches directly Easy peer-wise wiring between switches Random graphs of different scales Combinations of different topologies Packaging in Pods 7

Converter Switch Small port-count Low cost Physical layer device A B C - as packet switch * simple switching logic * no bandwidth contention * no expensive processor/buffering - as circuit switch * not sensitive to delay * small scale Physical layer device A B C D 8

Converter Switch Configurations 9

Flat-tree Example Clos Pod 10 Core Switch Edge Switch Aggregation Switch Server 10

Flat-tree Example Flat-tree Pod 11 Core Switch Edge Switch Converter Switch Aggregation Switch Server 11

Clos Network 12 Core Switch Converter Switch Aggregation Switch Server Edge Switch 12

Approximate Random Graph Core Switch Converter Switch Aggregation Switch Server Edge Switch 13

Approximate Local Random Graph Core Switch Converter Switch Aggregation Switch Server Edge Switch 14

Flat-tree Pod Blade B 15

Flat-tree Pod 16

Pod-Core Wiring 17

Server Distribution Choice of m and n Network profiling - how many servers per switch of different types - flat-tree maintains structure  not purely random * Clos connections between edge and aggregation switches * Pod-core connections * peer-wise connections between adjacent Pods - place servers to leverage shorter paths Network profiling - vary m and n - minimize average path length 18

Inter-Pod Wiring Simple shifting wiring pattern No repeated connection - <i, j> in Pod p  <i, (d/2-1-j+i)%(d/2)> in Pod p+1 No repeated connection Same number of “side” and “cross” connections Multi-link connectors - streamline the connection between adjacent Pods - hide wiring complexity 19

Evaluation Compared networks Metric - fat-tree - random graph - two-level random graph - flat-tree global (approximated global random graph) - flat-tree local (approximated pod-level random graph) - flat-tree hybrid (part flat-tree global and part flat-tree local) Metric - average path length - throughput * optimal routing * server links unbounded * linear programming solution

Evaluation Traffic patterns Locality - hot spots: broadcast/incast traffic in 1000-server clusters - clusters: all-to-all traffic in 20-server clusters Locality - (strong) locality * workload placed continuously across servers - weak locality * workload placed randomly in Pods - no locality * workload placed randomly in the entire network

Summary of Simulation Results Global Average Path Length Flat-tree Random Graph Clos ~4.75 ~4.6 ~5.9 Pod-Level Average Path Length Flat-tree Two-Level Random Graph Random Graph Clos ~3.4 ~3.6 ~4.6 ~3.9 20

Summary of Simulation Results Throughput of hot-spot traffic - flat-tree ≈ random graph - flat-tree = 1.5x Clos Throughput of small-clustered traffic - flat-tree > two-level random graph for 1/3 cases - flat-tree >= 91% two-level random graph - flat-tree = 1.15x random graph - flat-tree = 1.6x Clos 21

Global Average Path Length

Pod-Level Average Path Length

Throughput of Hot Spots Traffic

Throughput of Clustered Traffic

Conclusion Flat-tree converts between Clos topology and random graphs of different scales Low cost - inexpensive converter switches Easy implementation - changes packaged in Pods - regular Pod-core wiring patterns - multi-links between adjacent Pods Hybrid mode - network zones with different topologies Performance similar to random graphs - < 5% longer average path length - < 9% lower throughput 22

Impact and Inspiration Flat-tree is one design point of convertible network Motivate further study of relationship between different topologies Traffic optimization - joint optimization with routing and workload placement Network management - self recovery from failures - automatic up/down scale network at busy/idle time 23