Efficient Event-based Resource Discovery Wei Yan*, Songlin Hu*, Vinod Muthusamy +, Hans-Arno Jacobsen +, Li Zha* * Chinese Academy of Sciences, Beijing.

Slides:



Advertisements
Similar presentations
Network Resource Broker for IPTV in Cloud Computing Lei Liang, Dan He University of Surrey, UK OGF 27, G2C Workshop 15 Oct 2009 Banff,
Advertisements

C. Mastroianni, D. Talia, O. Verta - A Super-Peer Model for Resource Discovery Services in Grids A Super-Peer Model for Building Resource Discovery Services.
Dynamic Replica Placement for Scalable Content Delivery Yan Chen, Randy H. Katz, John D. Kubiatowicz {yanchen, randy, EECS Department.
Solving Manufacturing Equipment Monitoring Through Efficient Complex Event Processing Tilmann Rabl, Kaiwen Zhang, Mohammad Sadoghi, Navneet Kumar Pandey,
Opportunistic Multipath Forwarding in Publish/Subscribe Systems Reza Sherafat Kazemzadeh AND Hans-Arno Jacobsen Middleware Systems Research Group University.
P2PR-tree: An R-tree-based Spatial Index for P2P Environments ANIRBAN MONDAL YI LIFU MASARU KITSUREGAWA University of Tokyo.
Alex Cheung and Hans-Arno Jacobsen August, 14 th 2009 MIDDLEWARE SYSTEMS RESEARCH GROUP.
Scalable Content-Addressable Network Lintao Liu
Small-Scale Peer-to-Peer Publish/Subscribe
Transactional Mobility in Distributed Content-Based Publish/Subscribe Systems Songlin Hu*, Vinod Muthusamy +, Guoli Li +, Hans-Arno Jacobsen + * Chinese.
Small-world Overlay P2P Network
MIDDLEWARE SYSTEMS RESEARCH GROUP A Taxonomy for Denial of Service Attacks in Content-based Publish/Subscribe Systems Alex Wun, Alex Cheung, Hans-Arno.
Subscription Subsumption Evaluation for Content-Based Publish/Subscribe Systems Hojjat Jafarpour, Bijit Hore, Sharad Mehrotra, and Nalini Venkatasubramanian.
Applications over P2P Structured Overlays Antonino Virgillito.
©NEC Laboratories America 1 Hui Zhang Samrat Ganguly Sudeept Bhatnagar Rauf Izmailov NEC Labs America Abhishek Sharma University of Southern California.
Self-Correlating Predictive Information Tracking for Large-Scale Production Systems Zhao, Tan, Gong, Gu, Wambolt Presented by: Andrew Hahn.
Carnegie Mellon University Complex queries in distributed publish- subscribe systems Ashwin R. Bharambe, Justin Weisz and Srinivasan Seshan.
1 Peer-To-Peer-Based Resource Discovery In Global Grids: A Tutorial Rajiv Ranjan, Aaron Harwood And Rajkumar Buyya, The University Of Melbbourne IEEE Communications.
Matching Patterns Servers assemble sequences of notifications from smaller subsequences or from single notifications.This technique requires an advertisement.
XtreemOS IP project is funded by the European Commission under contract IST-FP XtreemOS WP3.2 - T3.2.3 Scalable Directory Service Design State.
Object Naming & Content based Object Search 2/3/2003.
Hermes: A Distributed Event- Based Middleware Architecture Peter Pietzuch and Jean Bacon 1st DEBS Workshop, Vienna,
1 Seminar: Information Management in the Web Gnutella, Freenet and more: an overview of file sharing architectures Thomas Zahn.
Wide-area cooperative storage with CFS
Project Mimir A Distributed Filesystem Uses Rateless Erasure Codes for Reliability Uses Pastry’s Multicast System Scribe for Resource discovery and Utilization.
Roger ZimmermannCOMPSAC 2004, September 30 Spatial Data Query Support in Peer-to-Peer Systems Roger Zimmermann, Wei-Shinn Ku, and Haojun Wang Computer.
Effects of Routing Computations in Content-Based Routing Networks with Mobile Data Sources Vinod Muthusamy, Milenko Petrovic, Hans-Arno Jacobsen University.
Word Wide Cache Distributed Caching for the Distributed Enterprise.
Publisher Mobility in Distributed Publish/Subscribe Systems Vinod Muthusamy, Milenko Petrovic, Dapeng Gao, Hans-Arno Jacobsen University of Toronto June.
MIDDLEWARE SYSTEMS RESEARCH GROUP Denial of Service in Content-based Publish/Subscribe Systems M.A.Sc. Candidate: Alex Wun Thesis Supervisor: Hans-Arno.
Gil EinzigerRoy Friedman Computer Science Department Technion.
Supporting Disconnected Operations in Publish/Subscribe Systems Vinod Muthusamy Joint work with Milenko Petrovic, Ioana Burcea, H.-Arno Jacobsen, Eyal.
Ramiro Voicu December Design Considerations  Act as a true dynamic service and provide the necessary functionally to be used by any other services.
Content-Based Routing in Mobile Ad Hoc Networks Milenko Petrovic, Vinod Muthusamy, Hans-Arno Jacobsen University of Toronto July 18, 2005 MobiQuitous 2005.
Scalable Web Server on Heterogeneous Cluster CHEN Ge.
MIDDLEWARE SYSTEMS RESEARCH GROUP Middleware A Policy Management Framework for Content-based Publish/Subscribe Middleware Hans-Arno Jacobsen Department.
Introduction to dCache Zhenping (Jane) Liu ATLAS Computing Facility, Physics Department Brookhaven National Lab 09/12 – 09/13, 2005 USATLAS Tier-1 & Tier-2.
DISTRIBUTED EVENT AGGREGATION FOR CONTENT-BASED PUBLISH/SUBSCRIBE SYSTEMS Navneet Kumar Pandey 1 Stéphane Weiss 1 Roman Vitenberg 1 Kaiwen Zhang 2 Hans-Arno.
MIDDLEWARE SYSTEMS RESEARCH GROUP MSRG.ORG Total Order in Content-based Publish/Subscribe Systems Joint work with: Vinod Muthusamy, Hans-Arno Jacobsen.
Distributed Automatic Service Composition in Large-Scale Systems Songlin Hu*, Vinod Muthusamy +, Guoli Li +, Hans-Arno Jacobsen + * Chinese Academy of.
Classification and Analysis of Distributed Event Filtering Algorithms Sven Bittner Dr. Annika Hinze University of Waikato New Zealand Presentation at CoopIS.
Flexible Distributed Business Process Management Vinod Muthusamy University of Toronto Thesis Defense September 23, 2011.
Networking 2012 On inter-domain name resolution for information-centric networks K.V. Katsaros, N. Fotiou, X. Vasilakos, C.N. Ververidis, C. Tsilopoulos,
S-Paxos: Eliminating the Leader Bottleneck
1 Peer-to-Peer Technologies Seminar by: Kunal Goswami (05IT6006) School of Information Technology Guided by: Prof. C.R.Mandal, School of Information Technology.
Parallel Event Processing for Content-Based Publish/Subscribe Systems Amer Farroukh Department of Electrical and Computer Engineering University of Toronto.
MIDDLEWARE SYSTEMS RESEARCH GROUP Adaptive Content-based Routing In General Overlay Topologies Guoli Li, Vinod Muthusamy Hans-Arno Jacobsen Middleware.
Information Services Andrew Brown Jon Ludwig Elvis Montero grid:seminar1:lectures:seminar-grid-1-information-services.ppt.
Minimal Broker Overlay Design for Content-Based Publish/Subscribe Systems Naweed Tajuddin Balasubramaneyam Maniymaran Hans-Arno Jacobsen University of.
ICDCS Beijing China Routing of XML and XPath Queries in Data Dissemination Networks Guoli Li, Shuang Hou Hans-Arno Jacobsen Middleware Systems Research.
Information-Centric Networks10b-1 Week 10 / Paper 2 Hermes: a distributed event-based middleware architecture –P.R. Pietzuch, J.M. Bacon –ICDCS 2002 Workshops.
Evaluation of Information Service Architectures for Grid.
Information-Centric Networks Section # 10.2: Publish/Subscribe Instructor: George Xylomenos Department: Informatics.
CS 6401 Overlay Networks Outline Overlay networks overview Routing overlays Resilient Overlay Networks Content Distribution Networks.
Peter R Pietzuch and Jean Bacon Peer-to-Peer Overlay Networks in an Event-Based Middleware DEBS’03, San Diego, CA, USA,
Distributed Automatic Service Composition in Large-Scale Systems Songlin Hu*, Vinod Muthusamy +, Guoli Li +, Hans-Arno Jacobsen + * Chinese Academy of.
MIDDLEWARE SYSTEMS RESEARCH GROUP MSRG.ORG Distributed Ranked Data Dissemination in Social Networks Joint work with: Mo Sadoghi Vinod Muthusamy Hans-Arno.
Community Clustering in Distributed Publish/Subscribe System Wei Li 1,2,Songlin Hu 1, Jintao Li 1, Hans-Arno Jacobsen 3 1 Institute of Computing Technology,
NCLAB 1 Supporting complex queries in a distributed manner without using DHT NodeWiz: Peer-to-Peer Resource Discovery for Grids Sujoy Basu, Sujata Banerjee,
Building Wireless Efficient Sensor Networks with Low-Level Naming J. Heihmann, F.Silva, C. Intanagonwiwat, R.Govindan, D. Estrin, D. Ganesan Presentation.
September 2003, 7 th EDG Conference, Heidelberg – Roberta Faggian, CERN/IT CERN – European Organization for Nuclear Research The GRACE Project GRid enabled.
Congestion Avoidance with Incremental Filter Aggregation in Content-Based Routing Networks Mingwen Chen 1, Songlin Hu 1, Vinod Muthusamy 2, Hans-Arno Jacobsen.
1 Towards Scalable Pub/Sub Systems Shuping Ji 1, Chunyang Ye 2, Jun Wei 1 and Arno Jacobsen 3 1 Chinese Academy of Sciences 2 Hainan University 3 Middleware.
Navneet Kumar Pandey1 Stéphane Weiss1 Roman Vitenberg1
Early Measurements of a Cluster-based Architecture for P2P Systems
Composite Subscriptions in Content-based Pub/Sub Systems
Foundations for Highly-Available Content-based Publish/Subscribe Overlays Young Yoon, Vinod Muthusamy and Hans-Arno Jacobsen.
Small-Scale Peer-to-Peer Publish/Subscribe
Load Balancing in BAD Project
Caching 50.5* + Apache Kafka
Presentation transcript:

Efficient Event-based Resource Discovery Wei Yan*, Songlin Hu*, Vinod Muthusamy +, Hans-Arno Jacobsen +, Li Zha* * Chinese Academy of Sciences, Beijing + University of Toronto July 9, rd Int’l Conference on Distributed Event- Based Systems (DEBS 2009) MIDDLEWARE SYSTEMS RESEARCH GROUP

MIDDLEWARE SYSTEMS RESEARCH GROUP Composite applications Mashups Service-oriented architectures Cloud computing A fundamental need is to discover resources and services. Many resources Distributed Dynamic attributes Real-time discovery

MIDDLEWARE SYSTEMS RESEARCH GROUP Related work DEBS ’09 Event-based Resource Discovery 3 SchemeComments Centralized index (Condor ’ s Matchmaker) Suffers from large scale and high dynamism. Hierarchical index (Globus ’ s MDS) Root node easily becomes a bottleneck. Federated UDDIExpensive to replicate frequently updated information among repositories. Discovery flooding (Gnutella) Creates a large volume of traffic. DHT (CAN, Chord, Pastry, etc.) Optimized for key-based discoveries. Range queries over DHTs (P-Ring, Mercury, etc.) Multi-attribute lookups can be expensive. Many approaches have limited expressiveness, or support for dynamic attributes or real-time discovery.

MIDDLEWARE SYSTEMS RESEARCH GROUP DEBS ’09 Event-based Resource Discovery 4 Contributions Event-based resource discovery  Distributed architecture  Leverage publish/subscribe system  Support dynamic resource updates  Allow for continuous discovery and real-time results Discovery similarity optimization  Share results among discoveries Evaluations  Tradeoffs of decentralized architecture  Benefits of sharing discovery results

MIDDLEWARE SYSTEMS RESEARCH GROUP DEBS ’09 Event-based Resource Discovery 5 Event-based resource discovery framework

MIDDLEWARE SYSTEMS RESEARCH GROUP DEBS ’09 Event-based Resource Discovery 6 Supported models One-time discovery Continuous discovery Static resource Dynamic resource Static (e.g., find weather service) Dynamic (e.g., find micro- generation power) Static continuous (e.g., monitor real estate) Dynamic continuous (e.g., monitor grid resources) Resources Discoveries Event-based

MIDDLEWARE SYSTEMS RESEARCH GROUP Architecture DEBS ’09 Event-based Resource Discovery 7 Resource providers act as publishers Discovery clients act as subscribers Advertise all attributes: system = linux memory <= 2000 disk <= 320 Publish updates of dynamic attributes: memory = 1500 disk = 80 Subscribe for resources: system = linux disk >= 200 B1 B4 B5 B2 B3 Distributed Content-Based Publish/Subscribe

MIDDLEWARE SYSTEMS RESEARCH GROUP Static model DEBS ’09 Event-based Resource Discovery 8 Discovery is performed locally by any single broker. Advertisement: system = linux, memory = 2, disk = 320 Subscription: memory > 1Publication B1 B4 B5 B2 B3

MIDDLEWARE SYSTEMS RESEARCH GROUP Dynamic model DEBS ’09 Event-based Resource Discovery 9 Resource update publication cached at the resource’s host broker. Discovery subscription routed to potentially matching resource host brokers. Advertisement: system= linux, memory <= 2, disk < 320 Subscription: memory > 1Publication: memory = 1, disk = 200 B1 B3 B4 B5 B2

MIDDLEWARE SYSTEMS RESEARCH GROUP Static continuous model DEBS ’09 Event-based Resource Discovery 10 Discovery is performed locally by any single broker (like static model). Discovery subscription stored at discovery host broker. Advertisement: system = linux, memory = 2, disk = 320 Subscription: memory > 1Publication B1 B4 B5 B2 B3

MIDDLEWARE SYSTEMS RESEARCH GROUP Dynamic continuous model DEBS ’09 Event-based Resource Discovery 11 Traditional pub/sub routing of messages. Discovery subscription is routed to and stored at matching resource host brokers. B5 B4 B3 B2 B1 Advertisement: system= linux, memory <= 2, disk < 320 Subscription: memory > 1Publication: memory = 1, disk = 200

MIDDLEWARE SYSTEMS RESEARCH GROUP Summary of models One-time discovery Continuous discovery Static resource Dynamic resource Static Dynamic Static continuous Dynamic continuous Discovery handled locally at discovery host broker Updates delivered only to interested clients No persistent subscription state Subscription state used to route back updates

MIDDLEWARE SYSTEMS RESEARCH GROUP DEBS ’09 Event-based Resource Discovery 13 Discovery similarity

MIDDLEWARE SYSTEMS RESEARCH GROUP Reuse results of similar discoveries DEBS ’09 Event-based Resource Discovery 14 Find machines with at least 1GB memory S1 Subscription: memory >= 1000 R1 R2 R3 More general Find machines with at least 2GB memory S1 Subscription: memory >= 2000 R1 R2 CoversSuperset

MIDDLEWARE SYSTEMS RESEARCH GROUP Similarity forwarding DEBS ’09 Event-based Resource Discovery 15 To retrieve old results: Send covered sub to the covering sub’s discovery host broker. To intercept new results: Store covered sub at the first broker with a covering sub. Adv: system= linux, memory <= 2, disk < 320 Sub2: memory > 2Pub: memory = 1, disk = 200 Sub1: memory > 1 B1 B5 B2 B3 B4

MIDDLEWARE SYSTEMS RESEARCH GROUP DEBS ’09 Event-based Resource Discovery 16 Evaluations

MIDDLEWARE SYSTEMS RESEARCH GROUP Experimental setup Algorithms implemented in Java  Based on PADRES content-based pub/sub system Run on a cluster of nodes with 1.86 GHz CPU and 4 GB memory Default workload  Topology Decentralized: 24 brokers Centralized: 1 broker  1000 resources Balanced and unbalanced spatial distributions  1000 discoveries Balanced and unbalanced spatial distributions Various degrees of similarity Metrics  Discovery time  Message overhead DEBS ’09 Event-based Resource Discovery 17

MIDDLEWARE SYSTEMS RESEARCH GROUP Discovery time Similarity forwarding optimization is faster Increased discovery similarity  Normal algorithm suffers More matching resources are found  Optimized algorithm benefits Reuse results Spatial clustering of resources  Normal algorithm benefits Smaller subscription propagation tree (more “multicast”)  Optimized algorithm benefits slightly Results are often retrieved from discovery host broker Spatial clustering of discoveries  Normal algorithm suffers Congestion of messages near discovery host brokers  Optimized algorithm suffers slightly Matching of cached results is relatively cheap DEBS ’09 Clustered spatial distribution of discoveries Balanced spatial distribution of discoveries Discovery similarity (%) Avg discovery time (s) Normal(B) Similarity(B) Normal(U) Similarity(U) Discovery similarity (%) Avg discovery time (s) Normal(B) Similarity(B) Normal(U) Similarity(U)

MIDDLEWARE SYSTEMS RESEARCH GROUP Similarity forwarding optimization propagates fewer subscriptions Increased discovery similarity  Normal algorithm suffers slightly More matching resources are found  Optimized algorithm benefits Covered subs only propagate to a single discovery host broker Spatial clustering of resources  Normal algorithm benefits Smaller subscription propagation tree (more “multicast”)  Optimized algorithm benefits (but less than normal algorithm) Covered subs are not affected Spatial clustering of discoveries  Normal algorithm has little effect Subscriptions still propagate to resource host brokers  Optimized algorithm has little effect Cost is dominated by the covering subs, which still need to propagate to resource host brokers DEBS ’09 Clustered spatial distribution of discoveries Balanced spatial distribution of discoveries Subscription messages

MIDDLEWARE SYSTEMS RESEARCH GROUP Decentralized architecture (one-time requests) DEBS ’09 Event-based Resource Discovery 20 Successive discovery groups match increasing number of resources Measure time to find (updated) resources Decentralized architecture distributes the load  Discovery handled locally by discovery host broker  Updates are propagated only to interested discovery host brokers

MIDDLEWARE SYSTEMS RESEARCH GROUP Decentralized architecture (continuous requests) DEBS ’09 Event-based Resource Discovery 21 Decentralized architecture better distributes the load  Results from similar discoveries are reused  Updates are propagated only to interested brokers

MIDDLEWARE SYSTEMS RESEARCH GROUP Conclusions Discovering resources and services is increasingly important in composite distributed applications A distributed event-based resource discovery framework was designed  Parallel discovery of static resources  Efficient dissemination of dynamic resource attributes  Real-time discovery of new resources Optimizations to exploit similarity among discoveries were developed  Find similar discoveries  Reuse results  Exploit publish/subscribe covering techniques Evaluations show that the distributed architecture achieves faster discovery at the expense of increased network traffic The similarity optimization benefits from more skewed spatial and interest distributions DEBS ’09 Event-based Resource Discovery 22

MIDDLEWARE SYSTEMS RESEARCH GROUP DEBS ’09 Event-based Resource Discovery 23 Efficient Event-based Resource Discovery padres.msrg.toronto.edu Open source soon! Q&A