1 04/18/2005 Flux Flux: An Adaptive Partitioning Operator for Continuous Query Systems M.A. Shah, J.M. Hellerstein, S. Chandrasekaran, M.J. Franklin UC.

Slides:



Advertisements
Similar presentations
Live migration of Virtual Machines Nour Stefan, SCPD.
Advertisements

Virtual Memory (II) CSCI 444/544 Operating Systems Fall 2008.
MapReduce Online Tyson Condie UC Berkeley Slides by Kaixiang MO
Scheduling in Web Server Clusters CS 260 LECTURE 3 From: IBM Technical Report.
Analysis of : Operator Scheduling in a Data Stream Manager CS561 – Advanced Database Systems By Eric Bloom.
Lecture 5 Memory Management Part I. Lecture Highlights  Introduction to Memory Management  What is memory management  Related Problems of Redundancy,
Managing Wire Delay in Large CMP Caches Bradford M. Beckmann David A. Wood Multifacet Project University of Wisconsin-Madison MICRO /8/04.
LIBRA: Lightweight Data Skew Mitigation in MapReduce
1 Storage-Aware Caching: Revisiting Caching for Heterogeneous Systems Brian Forney Andrea Arpaci-Dusseau Remzi Arpaci-Dusseau Wisconsin Network Disks University.
Building a Distributed Full-Text Index for the Web S. Melnik, S. Raghavan, B.Yang, H. Garcia-Molina.
SkewTune: Mitigating Skew in MapReduce Applications
Requirements on the Execution of Kahn Process Networks Marc Geilen and Twan Basten 11 April 2003 /e.
Study of Hurricane and Tornado Operating Systems By Shubhanan Bakre.
Data Warehousing 1 Lecture-25 Need for Speed: Parallelism Methodologies Virtual University of Pakistan Ahsan Abdullah Assoc. Prof. & Head Center for Agro-Informatics.
Tradeoffs in Scalable Data Routing for Deduplication Clusters FAST '11 Wei Dong From Princeton University Fred Douglis, Kai Li, Hugo Patterson, Sazzala.
Low-Cost Data Deduplication for Virtual Machine Backup in Cloud Storage Wei Zhang, Tao Yang, Gautham Narayanasamy University of California at Santa Barbara.
1 Routing and Scheduling in Web Server Clusters. 2 Reference The State of the Art in Locally Distributed Web-server Systems Valeria Cardellini, Emiliano.
Piccolo – Paper Discussion Big Data Reading Group 9/20/2010.
A Stratified Approach for Supporting High Throughput Event Processing Applications July 2009 Geetika T. LakshmananYuri G. RabinovichOpher Etzion IBM T.
VLDB Revisiting Pipelined Parallelism in Multi-Join Query Processing Bin Liu and Elke A. Rundensteiner Worcester Polytechnic Institute
ISCSI Performance in Integrated LAN/SAN Environment Li Yin U.C. Berkeley.
MS 9/19/97 implicit coord 1 Implicit Coordination in Clusters David E. Culler Andrea Arpaci-Dusseau Computer Science Division U.C. Berkeley.
An Adaptive Multi-Objective Scheduling Selection Framework For Continuous Query Processing Timothy M. Sutherland Bradford Pielech Yali Zhu Luping Ding.
1 stdchk : A Checkpoint Storage System for Desktop Grid Computing Matei Ripeanu – UBC Sudharshan S. Vazhkudai – ORNL Abdullah Gharaibeh – UBC The University.
1 Indirect Adaptive Routing on Large Scale Interconnection Networks Nan Jiang, William J. Dally Computer System Laboratory Stanford University John Kim.
Locality-Aware Request Distribution in Cluster-based Network Servers Presented by: Kevin Boos Authors: Vivek S. Pai, Mohit Aron, et al. Rice University.
Xen and the Art of Virtualization. Introduction  Challenges to build virtual machines Performance isolation  Scheduling priority  Memory demand  Network.
Optimizing Threaded MPI Execution on SMP Clusters Hong Tang and Tao Yang Department of Computer Science University of California, Santa Barbara.
Load distribution in distributed systems
Achieving Load Balance and Effective Caching in Clustered Web Servers Richard B. Bunt Derek L. Eager Gregory M. Oster Carey L. Williamson Department of.
©2001 Pål HalvorsenINFOCOM 2001, Anchorage, April 2001 Integrated Error Management in MoD Services Pål Halvorsen, Thomas Plagemann, and Vera Goebel University.
Department of Computer Science Engineering SRM University
施賀傑 何承恩 TelegraphCQ. Outline Introduction Data Movement Implies Adaptivity Telegraph - an Ancestor of TelegraphCQ Adaptive Building.
1 An SLA-Oriented Capacity Planning Tool for Streaming Media Services Lucy Cherkasova, Wenting Tang, and Sharad Singhal HPLabs,USA.
A Novel Adaptive Distributed Load Balancing Strategy for Cluster CHENG Bin and JIN Hai Cluster.
Parallel Programming Models Jihad El-Sana These slides are based on the book: Introduction to Parallel Computing, Blaise Barney, Lawrence Livermore National.
An Architectural Evaluation of SDN Controllers Syed Abdullah Shah, Jannet Faiz, Maham Farooq, Aamir Shafi, Syed Akbar Mehdi National University of Sciences.
Software Pipelining for Stream Programs on Resource Constrained Multi-core Architectures IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEM 2012 Authors:
1 University of Maryland Linger-Longer: Fine-Grain Cycle Stealing in Networks of Workstations Kyung Dong Ryu © Copyright 2000, Kyung Dong Ryu, All Rights.
Smita Vijayakumar Qian Zhu Gagan Agrawal 1.  Background  Data Streams  Virtualization  Dynamic Resource Allocation  Accuracy Adaptation  Research.
A Measurement Based Memory Performance Evaluation of High Throughput Servers Garba Isa Yau Department of Computer Engineering King Fahd University of Petroleum.
Fine-grained Partitioning for Aggressive Data Skipping Liwen Sun, Michael J. Franklin, Sanjay Krishnan, Reynold S. Xin† UC Berkeley and †Databricks Inc.
OPERETTA: An Optimal Energy Efficient Bandwidth Aggregation System Karim Habak†, Khaled A. Harras‡, and Moustafa Youssef† †Egypt-Japan University of Sc.
Abdullah Aldahami ( ) March 23, Introduction 2. Background 3. Simulation Techniques a.Experimental Settings b.Model Description c.Methodology.
1 An Adaptive File Distribution Algorithm for Wide Area Network Takashi Hoshino, Kenjiro Taura, Takashi Chikayama University of Tokyo.
Rassul Ayani 1 Performance of parallel and distributed systems  What is the purpose of measurement?  To evaluate a system (or an architecture)  To compare.
Online Data partitioning in distributed database systems
DynamicMR: A Dynamic Slot Allocation Optimization Framework for MapReduce Clusters Nanyang Technological University Shanjiang Tang, Bu-Sung Lee, Bingsheng.
Infrastructure for Data Warehouses. Basics Of Data Access Data Store Machine Memory Buffer Memory Cache Data Store Buffer Bus Structure.
Page 1 A Platform for Scalable One-pass Analytics using MapReduce Boduo Li, E. Mazur, Y. Diao, A. McGregor, P. Shenoy SIGMOD 2011 IDS Fall Seminar 2011.
1 Farm Issues L1&HLT Implementation Review Niko Neufeld, CERN-EP Tuesday, April 29 th.
1 Evaluation of Cooperative Web Caching with Web Polygraph Ping Du and Jaspal Subhlok Department of Computer Science University of Houston presented at.
Md Baitul Al Sadi, Isaac J. Cushman, Lei Chen, Rami J. Haddad
Optimizing Distributed Actor Systems for Dynamic Interactive Services
S. Sudarshan CS632 Course, Mar 2004 IIT Bombay
Memory Management.
Resource Management IB Computer Science.
Applying Control Theory to Stream Processing Systems
Self-Tuning Memory Management of A Database System
RT2003, Montreal Niko Neufeld, CERN-EP & Univ. de Lausanne
Memory Management for Scalable Web Data Servers
A Framework for Automatic Resource and Accuracy Management in A Cloud Environment Smita Vijayakumar.
Department of Computer Science University of California, Santa Barbara
Streaming Sensor Data Fjord / Sensor Proxy Multiquery Eddy
Smita Vijayakumar Qian Zhu Gagan Agrawal
Control Theory in Log Processing Systems
B. Stegmaier und R. Kuntschke TU München – Fakultät für Informatik
Performance And Scalability In Oracle9i And SQL Server 2000
Performance-Robust Parallel I/O
The Gamma Database Machine Project
Presentation transcript:

1 04/18/2005 Flux Flux: An Adaptive Partitioning Operator for Continuous Query Systems M.A. Shah, J.M. Hellerstein, S. Chandrasekaran, M.J. Franklin UC Berkeley Presenter: Bradley Momberger

2 04/18/2005 Overview ● Introduction ● Background ● Experiments and Considerations ● Conclusion

3 04/18/2005 Introduction ● Continuous query (CQ) systems – Create unbounded, streaming results from unbounded, streaming data sources. – May in the long run have scalability issues, due to the need for fast response times, the possibility of large numbers of users, and the management of potentially large histories. – Are only as fast as their constituent operators will allow.

4 04/18/2005 Parallelism ● Traditional parallelism techniques – Poor fit for CQ systems – Not adaptive ● CQ requires adaptability to changing conditions

5 04/18/2005 Overview ● Introduction ● Background ● Experiments and Considerations ● Conclusion

6 04/18/2005 Background ● Exchange – Producer-consumer pair – Ex-Prod: Intermediate producer instance connected to consumers – Ex-Cons: Intermediate consumer instance which polls inputs from all producers. – “Content sensitive” routing ● RiverDQ – “Content insensitive” routing – Random choice of Ex-Cons target

7 04/18/2005 Flux ● Flux, Fault-tolerant Load-balancing eXchange – Load balancing through active repartitioning – Producer-consumer pair – Buffering and reordering – Detection of imbalances

8 04/18/2005 Short Term Imbalances ● A stage runs only as fast as its slowest Ex-Cons – Head-of-line blocking – Uneven distribution over time ● The Flux-Prod solution – Transient Skew buffer ● Hashtable buffer between producer and Flux-Prod – Get new tuples for each Flux-Cons as buffer space becomes available. – On-demand input reordering

9 04/18/2005 Flux-Prod Design

10 04/18/2005 Long Term Imbalances ● Eventually overload fixed size buffers – Cannot use same strategy as short term ● The Flux-Cons solution – Repartition at consumer level – Move states – Aim for maximal benefit per state moved – Avoid “thrashing”

11 04/18/2005 Flux-Cons Design

12 04/18/2005 Memory Constrained Environment ● First tests were done with adequate memory – Does not necessarily reflect reality – Memory shortages ● Large histories ● Extra operators ● Load shedding with little memory – Push to disk – Move to other site – Decrease history size ● May not be acceptable in some applications

13 04/18/2005 Flux and Constrained Memory ● Dual-destination repartitioning – Other machines – Disk storage ● Local mechanism – Flux-Cons spills to disk when memory is low ● Retrieves from disk when memory becomes available ● Global Memory Constrained Repartitioning – Poll Flux-Cons operators for memory usage – Repartition based on results

14 04/18/2005 Memory-Adaptive Flux-Cons

15 04/18/2005 Overview ● Introduction ● Background ● Experiments and Considerations ● Conclusion

16 04/18/2005 Experimental Methodology ● Example operator – Hash-based, windowed group-by-aggregate – Statistic over fixed-size history ● Cluster hardware – CPU: 1000 MIPS – 1GB main memory ● Network simulation – 1K packet size, infinite bandwidth, 0.07ms latency – Virtual machines, simulated disk.

17 04/18/2005 Experimental Methodology ● Simulator – TelegraphCQ base system – Operators share physical CPU with event simulator – Aggregate evaluation and scheduler simulated ● Testbed – Single producer-consumer stage – 32 nodes in simulated cluster – Ex-Cons operator dictates performance

18 04/18/2005 Short Term Imbalance Experiment ● Give Flux stage a transient skew buffer – Compare to base Exchange stage with equivalent space ● Comparison statistics – 500ms load per virtual machine, round robin – Simulated process: 0.1ms processing, 0.05ms sleep – 16s runtime (32 machines  0.5s/machine)

19 04/18/2005 Short Term Imbalance Experiment

20 04/18/2005 Long Term Imbalance Experiment ● Operator stage – 64 partitions per virtual machine – 10,000 tuple (800KB) history per partition – 160KB skew buffer – 0.2μs per tuple for partition processing ● Network – 500mbps throughput for partitions – 250mbps point-to-point

21 04/18/2005 Balancing Processing Load

22 04/18/2005 Graceful Degradation

23 04/18/2005 Varying Collection Time

24 04/18/2005 Memory Constrained Experiments ● Memory “pressure” – 768MB initial memory load ● 6MB/partition  128 partitions/machine – Available memory => 512MB (down from 1GB) ● Change made after 1s of simulation – 14s required to push the remaining 256MB ● May be to disk or to other machines

25 04/18/2005 Throughput during Memory Balancing

26 04/18/2005 Avg. Latency during Memory Balancing

27 04/18/2005 Average Latency Degradation

28 04/18/2005 Hybrid Policy ● Combines previous policies – Memory-based policy when partitions are on disk ● Minimize latency – Load-balancing policy when all partitions are in memory ● Maximize throughput

29 04/18/2005 Comparative Review ┌ last 20 seconds of simulation ┐┌ Steady state ┐

30 04/18/2005 Overview ● Introduction ● Background ● Experiments and Considerations ● Conclusion

31 04/18/2005 Conclusions ● Flux – Is a reusable mechanism – Encapsulates adaptive repartitioning – Extends the Exchange operator – Alleviates short- and long-term imbalances – Outperforms static partitioning when correcting imbalances – Can use hybrid policies to adapt to changing processing and memory requirements.