The Ninja Architecture
Extended abstract
Pietu Pohjalainen
Department of Computer Science
Helsinki University

The Ninja Project is a research effort for exploring the problems and
possible resolving patterns of scalable and highly available
distributed applications. The problem area includes issues to be
resolved in areas of client/server-communication, thin client software
construction as well as server-side application distribution.

This paper concentrates on the distribution of server-side
applications: why should applications be distributed (motivation),
what different approaches have been made in order to resolve these
problems and what is the Ninja project approach to these problems.

The structure of this paper is divided to six chapters. Chapter one
presents the motivation for the paper: why should applications be
scalable and highly available and what are the specific problems when
distributing applications. Chapter two presents the traditional
approaches for achieving scalability while chapter three presents the
traditional model for high availability.

Then, chapter four offers the reader a high level view of the Ninja
architechture; how does this architecture offer support for
applications that need to be both highly available and
scalable. Chapter five drills more deeply into this
architecture. Last, but not least, chapter six presents a conclusion
of the whole paper.



References:
- The Ninja Architecture for Robust Internet-Scale Systems and Services.
  Steven D. Gribble, Matt Welsh, et al. Special Issue of Computer
  Networks on Pervasive Computing. Volume 25 Issue 4, March
  2001. pp. 473-497. 

- A Design Framework and a Scalable Storage Platform to Simplify
  Internet Service Construction. Steven D. Gribble. Ph.D. thesis,
  September 2000.  

- Scalable, Distributed Data Structures for Internet Service
  Construction. Steven D. Gribble et al. Proceedings of the 4th USENIX
  Symposium on Operating System Design and Implementation (OSDI
  2000), San Diego, California, USA. October 200.

- Blueprints for High Availability: Designing Resilient Distributed
  Systems. Evan Marcus, Hal Stern. John Wiley & Sons, 2000.

- Distributed Systems, Principles and Paradigms. Andrew S. Tanenbaum, Marten
  van Steen. Prentice-Hall, 2002.

- Enterprise Java Beans Technology. Sun Microsystems.
  http://java.sun.com/products/ejb/. 

- Java Remove Method Invocation - Distributed Computing for Java. Sun
  Microsystems. http://java.sun.com/.

- HP Application Server 8.0, A Technical White Paper. TechMatrix
  Research. 2001.

- HP Application Server 8.0, performance tuning guide. Hewlett-Packard
  Company. 2002.