Presentation is loading. Please wait.

Presentation is loading. Please wait.

Cluster-Based Scalable

Similar presentations


Presentation on theme: "Cluster-Based Scalable"— Presentation transcript:

1 Cluster-Based Scalable
Network Services Authors: Armando Fox, Steven D. Gribble, Yatin Chawathe, Eric A. Brewer (University of California, Berkeley) Paul Gauthier (Inktomi Corporation)

2 Contents Clusters & Load balancing
Problems in providing clustered services TACC - programming model Clustered-Based Scalable Network Architecture

3 Contents (cont..) TranSend & HotBot cluster implementations
TranSend performance results Current cluster servers

4 Clusters & Load balancing
Cluster : A domain where several homogeneous systems aim which behave as a single system to provide high performance service, availability, reliability & transparency of data over a network. Load Balancer r1 r3, r4 r1 r2 r3 r1,r2

5 Problems in providing clustered services
Scalability Availability Cost effective high performance Transparency Configurability & maintenance Extensibility

6 TACC - programming model
Transformation Aggregation Caching Customization

7 Conventional Transactional systems
ACID properties (Atomicity, Consistency, Isolation & Durability) Internet Services are requires BASE properties (Basic Availability, Soft state & Eventual Consistency)

8 Clustered-Based Scalable Network Architecture

9 TranSend version of SNS
(a) Front Ends HTTP Request accept Pairing HTTP requests with User’s preferences Assigning requests to distillers Return cached data

10 TranSend (Cont…) (b) Load Balancing Requests Front End MS Cache ($)
User Profile Front End MS Manager Cache ($) Distiller + WS Response Requests

11 (d) User Profile database
TranSend (Cont…) (c) Crash recovery & fault tolerance (d) User Profile database (e) Graphical Monitor (d) Caching

12 HotBot implementation (Inktomi work)
Front End Nodes – multiple threads put the connections in the queue Load balancer -- statistically partition the database among worker nodes Failure Management -- similar nodes are attached for a partition

13 TranSend – performance details
Distiller Performance

14 TranSend -- Cache performance
Average cache hit take 27 millisecs to serve the request 95% of cache hits takes less than 100 milliseconds Observed increased performance using LRU caching mechanism till the total users wont exceed the cache size. Latency due to many connections with Front-Ends

15 TranSend – scalability
Front-Ends scalability Distiller scalability

16 Interface layer for cluster object distribution
Current cluster servers (WebLogic, WebSphere, PRAMATI Web-Server) requests requests response response Load Balancer Load Balancer Interface layer for cluster object distribution Node3-r1 Node1-r1 Node1-r2 registering New Node Node2-r1

17 Scalability of web servers
K E T Connections Processing Threads (keep alive) Accepting Threads

18 Conclusions & Contributions
Easy implementation of stateless workers to achieve TACC for Internet Content Large scale network services can be achieved by BASE principles References: --- for cluster architecture PRAMATI web server , Resin, Tomcat for connection scalability


Download ppt "Cluster-Based Scalable"

Similar presentations


Ads by Google