The Bio-Networking Architecture: An Infrastructure of Autonomic Agents in Pervasive Networks

1. The Bio-Networking Architecture:An Infrastructure of Autonomic Agents in Pervasive Networks Jun Suzuki jsuzuki@ics.uci.edunetresearch.ics.uci.edu/bionet/School of Information and Computer ScienceUniversity of California, Irvine

2. The Bio-Networking Architecture • Requirements to pervasive networks • scalability in terms of # of objects and network nodes • adaptability to changes in network conditions • availability/survivability from massive failures and attacks • simplicity to design and maintain • Our solution:apply biological concepts and mechanisms to network application design • Biological systems have overcome the above features. • e.g. bee colony, bird flock, fish school, etc. • The Bio-Networking Architecture • a network application architecture that models autonomic adaptive agents (called cyber-entities) after biological concepts and mechanisms. • in hope that scalability, adaptability and availability/survivability are improved or they are provided differently from existing ways.

3. Biological Concepts Applied • Decentralized system organization • No centralized administration and coordination among CEs (cyber-entities) • for scalability and survivability • Lifecycle • Each CE stores and expends energy for living. • gains energy in exchange for providing its service to other CEs • expends energy for performing its behaviors, utilizing resources (e.g. CPU and memory), and invoking another CE’s service. • Evolution • CEs evolve by generating behavioral diversity among them and executing natural selection. • Social networking • CEs are linked with each other using relationships.

4. The Bio-Networking Platform (bionet platform) Attributes Body Behaviors Bionet platform Devise cyber-entity Cyber-entities runningon a bionet platform users • The bionet platform is a middleware that • hosts the Bio-Networking Architecture on the real networks. • aids developing, deploying and executing cyber-entities by • providing a set of runtime services and components. • abstracting low-level networking and operating details.

5. Architecture A Cyber-entity (CE) is an autonomous mobile object. CEs communicate with each other using FIPA ACL. CE CE Platformrepresentative A platform rep keeps references to bionet services and container. External Helper Tools CE Context A CE context provides references to available bionet services. Bionet Services Bionet services are runtime services that CEs use frequently. Bionet Container Bionet container dispatches incoming messages to target CEs. Bionet Message Transport Bionet Class Loader Bionet message transport takes care of I/O, low-level messaging and concurrency. Bionet Platform Java VM Bionet class loader loads byte code of CEs to Java VM.

6. Bionet Services • Each bionet platform provides 8 bionet services. • CEs use bionet services to invoke their behaviors. • e.g. bionet lifecycle service when a CE replicates • Bionet Lifecycle Service • used to initialize and install a CE. • maintains a thread pool that contains threads assigned to CEs • allows a CE to replicate itself. • allows a CE to reproduce a child CE with a partner • Bionet Relationship Management Service • allows a CE to establish, examine, update and eliminate their relationships with other CEs. • Bionet Energy Management Service • keeps track of energy level of the CEs running on a local platform. • allows a CE to pay energy amounts for • invoking a service provided by another CE, • using resources • performing behaviors (i.e. invoking a bionet service).

7. Bionet Resource Sensing Service • allows CEs to sense the type, amount and unit cost of available resources. • CPU cycles and memory space • Bionet CE Sensing Service • allows a CE to discover other CEs running on the same platform. • used to establish relationships with neighboring CEs when a CE is created or completes migration • Bionet Migration Service • allows a CE to migrate to another bionet platform. • weak migration: transmits CE’s data state, CE’s class definition, and CE’s class name • Bionet Pheromone Emission/Sensing Service • allows a CE to leave its pheromone (or trace) on a local platform when it migrates to another platform • so that other CEs can find the CE at a destination platform • Bionet Social Networking Service • allows a CE to search other CEs through their relationships.

8. A Measurement Result • CEs implement a web server function on a bionet platform. • Request rate: 10 requests/sec • 500B, 5KB, 50KB, 500KB, and 5MB • These 4 types of file request are representative in a well-known web server performance profiling tool (WebStone). • 500B (35%), 5KB (50%), 50KB (14 %), 500KB (0.9%), and 5MB (0.1%) • Observations • When CPU utilization goes around 75%, the total CPU utilization reaches 100%. • The other 25% is used by the operating system. • 320 CEs can run before the CPU utilization reaches 75% (100% in total). • 290 CEs can run before the CPU utilization reaches 50%. • The bionet platform is scalable enough.

9. Standardization Effort at OMG • The OMG Super Distributed Objects (SDOs) working group • provide a standard computing infrastructure that incorporates massive numbers of objects (SDOs) including hardware devices and software components • deploy them in a highly-distributed and ubiquitous environment, and • allow them to seamlessly interwork with each other. • Status: • The SDO RFI issued (‘00), and responses gathered (‘01) • from 10 organizations including UCI • The SDO white paper published (‘01) • by Hitachi, GMD Fokus and UCI • The first RFP published (Jan. 02). • solicits resource data model for SDOs, discovery interfaces, etc. • The initial proposals submitted (Sept. 02) • by Hitachi, GMD Fokus and UCI • 28 organizations on the voting list • The revised joint proposal was submitted in March 2003. • by Hitachi, GMD Fokus and UCI • Several key designs have been reflected into the standard proposal, e.g. CE’s structures, CE’s profiles, and relationship between CEs.