26 Sep 2003 Transparent Adaptive Resource Management for Distributed Systems Department of Electrical Engineering and Computer Science Vanderbilt University, Nashville Jaiganesh Balasubramanian Ossama Othman Dr. Douglas C. Schmidt {jai, ossama,

Balasubramanian et allTransparent Resource Management Vanderbilt University Distributed Systems Typical issues with distributed systems –Heterogeneous environments –Concurrency –Large bursts of client requests –24/7 availability –Stringent QoS requirements Examples of distributed systems –E-commerce –Online trading systems –Mission critical systems Defense –Ship systems

Balasubramanian et allTransparent Resource Management Vanderbilt University Motivation Development and maintenance of QoS-enabled distributed systems –Non-trivial –Requires expertise that application developers often lack Solution: Middleware (e.g. CORBA) –Can shield distributed system developers from the complexities involved with developing distributed applications –Can facilitate manipulation of QoS requirements and management of resources

Balasubramanian et allTransparent Resource Management Vanderbilt University Distributed System Resource Management Middleware can alleviate resource management difficulties Doing so in a transparent and efficient manner is difficult Managing resources transparently is important for legacy DRE systems –Often cannot be easily modified to introduce support for improved distributed resource management –May not be feasible to do so Middleware in use may need to be enhanced to support this functionality –Such an enhancement can be found in a middleware- based load balancing service

Balasubramanian et allTransparent Resource Management Vanderbilt University Middleware-Based Load Balancing Load balancing service can be used to manage resources for middleware-based distributed systems –Can improve the efficiency and overall scalability of a distributed system –Allows additional components to be added to distributed systems with minimal impact to performance –Improves availability due to inherent redundancy –Can be composed with other services (e.g. fault tolerance, security, etc) –Can take into account request content

Balasubramanian et allTransparent Resource Management Vanderbilt University Basic Scenario Multiple clients making request invocations –Potentially non-deterministic Members –Multiple instances of the same object implementation Object groups –Collections of members among which loads will be distributed equitably –Logically a single object Load balancer –Transparently distributes requests to members within an object group

Balasubramanian et allTransparent Resource Management Vanderbilt University TAO Load Balancer (Cygnus) Cygnus makes it easier to develop distributed applications in heterogeneous environments providing application transparency, scalability, flexibility, adaptability and interoperability. Uses the following strategies: - RoundRobin - Random - LeastLoaded

Balasubramanian et allTransparent Resource Management Vanderbilt University Load Balancing Strategies Client binding granularity –Per-session Client permanently forwarded to a replica –Per-request Requests forward on client’s behalf –On-demand Client can be rebound to another replica whenever necessary Balancing policy –Non-adaptive No load feedback used when binding clients –Adaptive Load feedback taken in to account

Balasubramanian et allTransparent Resource Management Vanderbilt University Load Balancing Architectures Load balancing architecture comprised of a combination of client binding granularity and balancing policy Given the strategies just described, there are six possible architectures Three common architectures –Non-adaptive per- session –Adaptive per-request –Adaptive on-demand

Balasubramanian et allTransparent Resource Management Vanderbilt University Load Balancer Components Load Monitor –Provides load feedback Load Analyzer –Determines location and member load conditions Member Locator –Binds client to appropriate object group member Load Alert –Facilitates load control (load shedding) Load Manager –Mediates all interactions between load balancer components

Balasubramanian et allTransparent Resource Management Vanderbilt University Load Monitor Facilitates feedback –Monitor and report loads Load monitoring is location- oriented –The loads at a given location are monitored and reported –Contrast with loads on a given object group member

Balasubramanian et allTransparent Resource Management Vanderbilt University Load Analyzer –Decides which member will receive the next client request –Determines load condition at each location –Induces load shedding –Extensible load balancing strategies Set via the PropertyManager interface Each object group may use a different strategy

Balasubramanian et allTransparent Resource Management Vanderbilt University Member Locator Implements the Interceptor design pattern Transparently forwards client requests to member retrieved from the load analyzer –In CORBA, redirection induced via standard GIOP LOCATION_FORWARD message –Conforming client side ORB will transparently re-issue request to member chosen by load balancer

Balasubramanian et allTransparent Resource Management Vanderbilt University Load Alert Facilitates “control” aspect of load balancing –Load shedding –Only used for adaptive load balancing strategies –Forwards client requests back to load balancer

Balasubramanian et allTransparent Resource Management Vanderbilt University Load Manager Mediates interactions between all load balancer components Manages object groups Manages load monitor and load alert location maps Acts as a specialized event channel –Load events are published by load monitors –Load events are consumed by load analyzers

Balasubramanian et allTransparent Resource Management Vanderbilt University Load Balancing Instructions

Balasubramanian et allTransparent Resource Management Vanderbilt University Future Work Define stateful load balancing model Decentralized load balancing –Reduced network overhead –Improved reliability Integrate multicast support Examine compatibility with real-time systems Examine similarities and parallels to dynamic scheduling

Balasubramanian et allTransparent Resource Management Vanderbilt University Concluding Remarks Load balancing architecture and model is –Flexible Extensible load balancing strategies –Allows both non-adaptive and adaptive (including self- adaptive) strategies to be used Freedom to implement in a variety of ways –Centralized –Decentralized / Federated / Cooperative –Hierarchical –Generic Supports multiple object groups Not application-specific –Familiar Uses many of the same group management concepts found in existing fault tolerance models

Balasubramanian et allTransparent Resource Management Vanderbilt University References Ossama Othman, Jaiganesh Balasubramanian, Douglas C. Schmidt “ The Design of an Adaptive Load Balancing and Monitoring Service” Ossama Othman, Jaiganesh Balasubramanian, Douglas C. Schmidt “Performance Evaluation of an Adaptive Load Balancing and Monitoring Service”