Cross-Layer Scheduling in Cloud Systems Hilfi Alkaff, Indranil Gupta, Luke Leslie Department of Computer Science University of Illinois at Urbana-Champaign.

Slides:



Advertisements
Similar presentations
Path Splicing with Network Slicing
Advertisements

Data Center Networking with Multipath TCP
SDN + Storage.
CloudWatcher: Network Security Monitoring Using OpenFlow in Dynamic Cloud Networks or: How to Provide Security Monitoring as a Service in Clouds? Seungwon.
COS 461 Fall 1997 Routing COS 461 Fall 1997 Typical Structure.
Jaringan Komputer Lanjut Packet Switching Network.
Designing An g Ad-hoc Network for Multimedia Communication Chung-Wei Lee & Jonathan C.L. Liu Presented By: Mahendra Kumar.
Esma Yildirim Department of Computer Engineering Fatih University Istanbul, Turkey DATACLOUD 2013.
Gossip Scheduling for Periodic Streams in Ad-hoc WSNs Ercan Ucan, Nathanael Thompson, Indranil Gupta Department of Computer Science University of Illinois.
Towards Energy Efficient Hadoop Wednesday, June 10, 2009 Santa Clara Marriott Yanpei Chen, Laura Keys, Randy Katz RAD Lab, UC Berkeley.
Receiver-driven Layered Multicast S. McCanne, V. Jacobsen and M. Vetterli SIGCOMM 1996.
Datacenter Network Topologies
Towards Energy Efficient MapReduce Yanpei Chen, Laura Keys, Randy H. Katz University of California, Berkeley LoCal Retreat June 2009.
December 20, 2004MPLS: TE and Restoration1 MPLS: Traffic Engineering and Restoration Routing Zartash Afzal Uzmi Computer Science and Engineering Lahore.
A General approach to MPLS Path Protection using Segments Ashish Gupta Ashish Gupta.
1 Emulating AQM from End Hosts Presenters: Syed Zaidi Ivor Rodrigues.
A Scalable, Commodity Data Center Network Architecture Mohammad Al-Fares, Alexander Loukissas, Amin Vahdat Presented by Gregory Peaker and Tyler Maclean.
A General approach to MPLS Path Protection using Segments Ashish Gupta Ashish Gupta.
Multipath Routing CS 522 F2003 Beaux Sharifi. Agenda Description of Multipath Routing Necessity of Multipath Routing 3 Major Components Necessary for.
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.
Camdoop: Exploiting In-network Aggregation for Big Data Applications Paolo Costa, Austin Donnelly, Antony Rowstron, Greg O’Shea Presenter – Manoj Kumar(mkumar11)
The Structure of Networks with emphasis on information and social networks T-214-SINE Summer 2011 Chapter 8 Ýmir Vigfússon.
On Availability of Intermediate Data in Cloud Computations Steven Y. Ko, Imranul Hoque, Brian Cho, and Indranil Gupta Distributed Protocols Research Group.
Cross-Layer Scheduling in Cloud Computing Systems Authors: Hilfi Alkaff, Indranil Gupta.
U.S. Department of the Interior U.S. Geological Survey David V. Hill, Information Dynamics, Contractor to USGS/EROS 12/08/2011 Satellite Image Processing.
Network Support for Cloud Services Lixin Gao, UMass Amherst.
Lecture 3-1 Computer Science 425 Distributed Systems CS 425 / CSE 424 / ECE 428 Fall 2010 Indranil Gupta (Indy) August 31, 2010 Lecture 3  2010, I. Gupta.
Link State Routing Protocol W.lilakiatsakun. Introduction (1) Link-state routing protocols are also known as shortest path first protocols and built around.
ON DESIGING END-USER MULTICAST FOR MULTIPLE VIDEO SOURCES Y.Nakamura, H.Yamaguchi, A.Hiromori, K.Yasumoto †, T.Higashino and K.Taniguchi Osaka University.
Introduction to Hadoop and HDFS
Topology aggregation and Multi-constraint QoS routing Presented by Almas Ansari.
DARD: Distributed Adaptive Routing for Datacenter Networks Xin Wu, Xiaowei Yang.
© Copyright 2010 Hewlett-Packard Development Company, L.P. 1 Jayaram Mudigonda, HP Labs Praveen Yalagandula, HP Labs Mohammad Al-Fares, UCSD Jeff Mogul,
CloudNaaS: A Cloud Networking Platform for Enterprise Applications Theophilus Benson*, Aditya Akella*, Anees Shaikh +, Sambit Sahu + (*University of Wisconsin,
Chi-Cheng Lin, Winona State University CS 313 Introduction to Computer Networking & Telecommunication Chapter 5 Network Layer.
Scalable Multi-Class Traffic Management in Data Center Backbone Networks Amitabha Ghosh (UtopiaCompression) Sangtae Ha (Princeton) Edward Crabbe (Google)
Load-Balancing Routing in Multichannel Hybrid Wireless Networks With Single Network Interface So, J.; Vaidya, N. H.; Vehicular Technology, IEEE Transactions.
Grid Computing at Yahoo! Sameer Paranjpye Mahadev Konar Yahoo!
InterConnection Network Topologies to Minimize graph diameter: Low Diameter Regular graphs and Physical Wire Length Constrained networks Nilesh Choudhury.
Networking Fundamentals. Basics Network – collection of nodes and links that cooperate for communication Nodes – computer systems –Internal (routers,
A Utility-based Approach to Scheduling Multimedia Streams in P2P Systems Fang Chen Computer Science Dept. University of California, Riverside
Intradomain Traffic Engineering By Behzad Akbari These slides are based in part upon slides of J. Rexford (Princeton university)
Cpr E 308 Spring 2005 Process Scheduling Basic Question: Which process goes next? Personal Computers –Few processes, interactive, low response time Batch.
Large-scale Virtualization in the Emulab Network Testbed Mike Hibler, Robert Ricci, Leigh Stoller Jonathon Duerig Shashi Guruprasad, Tim Stack, Kirk Webb,
Scale up Vs. Scale out in Cloud Storage and Graph Processing Systems
DynamicMR: A Dynamic Slot Allocation Optimization Framework for MapReduce Clusters Nanyang Technological University Shanjiang Tang, Bu-Sung Lee, Bingsheng.
1 A Cross-Layer Architecture to Exploit Multi-Channel Diversity Jay A. Patel, Haiyun Luo, and Indranil Gupta Department of Computer Science University.
Virtual-Channel Flow Control William J. Dally
Peter Pham and Sylvie Perreau, IEEE 2002 Mobile and Wireless Communications Network Multi-Path Routing Protocol with Load Balancing Policy in Mobile Ad.
Querying the Internet with PIER CS294-4 Paul Burstein 11/10/2003.
Stela: Enabling Stream Processing Systems to Scale-in and Scale-out On- demand Le Xu ∗, Boyang Peng†, Indranil Gupta ∗ ∗ Department of Computer Science,
Spark on Entropy : A Reliable & Efficient Scheduler for Low-latency Parallel Jobs in Heterogeneous Cloud Huankai Chen PhD Student at University of Kent.
Network Layer COMPUTER NETWORKS Networking Standards (Network LAYER)
Yiting Xia, T. S. Eugene Ng Rice University
Data Center Network Architectures
Chris Cai, Shayan Saeed, Indranil Gupta, Roy Campbell, Franck Le
CS 425 / ECE 428 Distributed Systems Fall 2016 Nov 10, 2016
CS 425 / ECE 428 Distributed Systems Fall 2017 Nov 16, 2017
Fisheye Routing protocol
ISP and Egress Path Selection for Multihomed Networks
MapReduce Computing Paradigm Basics Fall 2013 Elke A. Rundensteiner
On the Physical Carrier Sense in Wireless Ad-hoc Networks
Boyang Peng, Le Xu, Indranil Gupta
湖南大学-信息科学与工程学院-计算机与科学系
COS 561: Advanced Computer Networks
Jellyfish: Networking Data Centers Randomly
Pradeep Kyasanur Nitin H. Vaidya Presented by Chen, Chun-cheng
2019/5/13 A Weighted ECMP Load Balancing Scheme for Data Centers Using P4 Switches Presenter:Hung-Yen Wang Authors:Peng Wang, George Trimponias, Hong Xu,
In-network computation
2019/9/14 The Deep Learning Vision for Heterogeneous Network Traffic Control Proposal, Challenges, and Future Perspective Author: Nei Kato, Zubair Md.
Presentation transcript:

Cross-Layer Scheduling in Cloud Systems Hilfi Alkaff, Indranil Gupta, Luke Leslie Department of Computer Science University of Illinois at Urbana-Champaign 1 Distributed Protocols Research Group:

Inside a Datacenter: Networks Connecting Servers Tree Fat Tree [Leiserson 85] Jellyfish [Singla 12] Clos [Dally 04] VL2 [Greenberg 09] 2

Tree Fat Tree [Leiserson 85] Jellyfish [Singla 12] Clos [Dally 04] VL2 [Greenberg 09] Structured NetworksUnstructured Networks and/or routing Inside a Datacenter: Networks Connecting Servers 3

SDN Software Defined Networking For any end-host pair, multiple routes available SDN Controller helps to choose one of these routes – Configures switches accordingly Which route is the “best”? 4

SDNs and Applications Which route is the “best”? Our approach – Best network routes should really be decided based on the application that is using the network To minimize interference (and thus congestion) and to optimize bandwidth use Today: SDN routes selected application-agnostic way – But the application itself can help, by placing tasks at servers Today: Applications schedule tasks in network-agnostic way, leading to bad bandwidth utilization – SDN Controller and Application Scheduler should coordinate with each other This is our cross-layer scheduling approach 5

Applications: Short Real-Time Analytics Jobs Batch Processing: MapReduce, Hadoop Stream Processing: Storm 6

Tasks Storm Tasks Hadoop 7

Tasks and Flows Storm Tasks Hadoop Flows 8

Challenges Two large state spaces to explore 1.Set of Possible Routes for each end-to-end flow – Large numbers of flows and possible routes 2. Set of Possible Task to Server Placements – Large numbers of servers and tasks 9

Our Strategy To explore state space, use simulated annealing – At application level scheduler – And separately at routing (SDN) level Simulated Annealing – probabilistic approach – avoids getting stuck in local optima with some non-zero probability of jumping away – probability of jumping away decreases quickly over time (annealing process for steel) 10

Pre-computation For all pairs of servers, pre-compute the k shortest paths – Store it in a hash table, indexed by server pair – Compact storage by merging overlapping routes (for a server pair) into a tree Small in size and Quick to compute – 1000 servers, k=10 – 50 M entries – After compaction, 6 MB – 3 minutes to generate 11

When a Job Arrives Don’t change the allocations or routes of existing jobs – Non-intrusive – Reduces state space to explore Simulated Annealing is run offline, and the resultant schedule is used to schedule new job’s tasks and flows Primary Simulated Annealing (SA) runs at Application level – Calls Routing level SA 12

Simulated Annealing Steps Start from an arbitrary state – Tasks to servers, and routes to flows Generate next-state S’ (At Application Level) 1.De-allocate one task Prefer tasks that affect computation more, e.g., closer to beginning or end of topology 2.Allocate this task to random server 3.Call Routing level SA 13

Simulated Annealing Steps (2) … 3.Call Routing level SA 4.(At Routing Level) 5.De-path one route Select random server pair Remove its worst path – Prefer higher number of hops, and break ties by lower bandwidth 6.Allocate Path: Change this route to a better path – Prefer lower number of hops, and break ties by higher bandwidth 14

Simulated Annealing Steps (3) After generating next-state S’ – Calculate utility(S’) – Utility function considers all jobs in cluster (not just new job) – Utility function accounts for bottlenecked paths from source tasks to sink tasks If utility(S’) > utility(current state) – Transition from current state to S’ If utility(S’) ≤ utility(current state) – Transition with probability e (utility(S’)-utility(current state))/t – Non-zero probability of transitioning even if S’ is a worse state – Probability decreases over time (t) Wait until convergence Re-run entire simulated annealing 5 times, and take best result 15

Experiments Implemented into Apache Hadoop (YARN) Implemented into Apache Storm Deployment experiments on Emulab: up to 30 hosts – Emulated network using ZeroMQ and Thrift – Emulated Fat-Tree and Jellyfish Larger scale simulation experiments – Upto 1000 hosts 16

Experimental Settings 10 hosts, 100 Mbps, 5 links per router, #links selected via scaling rules – 3 GHz, 2 GB RAM Hadoop cluster workload – Facebook’s SWIM benchmark – Shuffle ranges from 100 B to 10 GB – 1 job per second Storm cluster workload: Random tree topologies – Topologies constructed as randomly with number of children selected by Gaussian (mean = sd = 2) – 100 B tuples – Each source generate 1 MB – 100 MB of data – 10 jobs per minute Each experimental run is 10 minutes 17

Tree Fat Tree [Leiserson 85] Jellyfish [Singla 12] Clos [Dally 04] VL2 [Greenberg 09] Structured NetworksUnstructured Networks and/or routing Inside a Datacenter: Networks Connecting Servers 18

Storm on Jellyfish Topology App+Routing SA: 34.1% improvement in throughput at 30 hosts Application-only SA: 21.2% Routing-only SA: 23.2% Performance improves with scale 19

Hadoop on Fat-Tree Topology App+Routing SA: 26% improvement in throughput at 30 hosts Application-only SA & Routing-only SA Smaller than combining both Performance improves with scale 20

Other Experimental Results Similar results for other combinations Hadoop on Jellyfish – App+Routing SA: 31.9% improvement in throughput at 30 hosts – Performance improves with scale – Application-only SA: 18.8% – Routing-only SA: 25.5% Storm on Fat-Tree – App+Routing SA: 30% improvement in throughput at 30 hosts – Performance improves with scale – Application-only SA: 21.1% – Routing-only SA: 22.7% 21

Other Experimental Results (2) Scheduling time is small – Time to schedule a new job in a 1000 server cluster – Fat-Tree: 0.48 s (Hadoop) to 0.53 s (Storm) – Jellyfish: 0.67 s (Hadoop) to 0.74 s (Storm) No starvation – Worst case degradation in completion time for any job is 20% in Hadoop, 30% in Storm – Outliers are large jobs (rare in real-time analytics with short jobs) Fault-recovery is fast – Upon failure, re-run simulated annealing once – Recovery occurs within 0.35 s to 0.4 s 22

Takeaways Today: Application schedulers and SDN scheduler are disjoint – Leads to suboptimal placement and routing Our approach: coordinated cross-layer scheduling – Explore small state spaces – Use simulated annealing At 30 hosts, gives between 26% to 34% improvement in Hadoop and Storm for both structured/unstructured networks – Other networks will fall between these two numbers Overheads are small, and improvement gets better with scale 23 Distributed Protocols Research Group:

Ongoing/Future Work Our work opens the door: Explore other heuristics, e.g., data affinity for tasks, congestion Explore other non-SA approaches Available bandwidth estimation OpenFlow integration Batching multiple jobs into scheduling 24 Distributed Protocols Research Group: