Dynamic Performance Management in Multi-tenanted Business Process Servers Using Nonlinear Control

This conference paper "Dynamic Performance Management in Multi-tenanted Business Process Servers Using Nonlinear Control" is our latest publication which discusses the detail on our work on performance isolation aspect in a multi-tenanted business process execution runtime. It was recently published in 10th International Conference on Service Oriented Computing (ICSOC 2012), pp. 206-221. Authors of this paper are Tharindu Patikirikorala, Indika Kumara, Alan Colman, Jun Han, Liuping Wang, Denis Weerasiri and Waruna Ranasinghe.

Abstract of the paper is
The methodologies to develop multi-tenanted architectures have been investigated in the recent literature due to the popularity of cloud computing. A number of challenges need to be overcome if multi-tenanted architectures are to be effective and efficient. Among the challenges is the management of performance properties while effectively sharing the limited resources between the tenants. This work presents an approach to design such a management system for a multi-tenanted business process server. This approach not only enables performance to be maintained at different levels for different tenants depending on their priorities, but also autonomously detects the overloads of aggressive tenants and dynamically changes the control objectives to safeguard the business operations of other tenants. The novelty of the proposed approach is the use of the nonlinear feedback control. The experiment results indicate that the proposed nonlinear control approach achieves the objectives much better compared to the existing fixed and linear control techniques.

This paper can be accessed from here.

Relavant articles

• A Scalable Multi-tenant Architecture for Business Process Executions  
• Horizontal Format Data Mining with Extended Bitmaps - Extended version 
• Horizontal Format Data Mining with Extended Bitmaps 
• Mooshabaya: Mashup Generator for XBaya

The New Beginning

Two years back on 13th of September, I joined WSO2 after my undergraduate studies from University of Moratuwa. WSO2 was a wonderful place for me (as a fresh graduate) to start my career. And I have been to several software companies in different parts of the world and I haven’t experience such a friendly culture and a freedom from any other places.

Yesterday, after exactly two years I officially started my PhD studies at University of New South Wales with an induction for postgraduate researchers. I still feel like I am entering to a pitch black tunnel and I am still not sure when/how I will find the other end of the tunnel :).

Anyway my life is not all about computer engineering. There are several other stuff on my bucket list like being a photographer, traveler, licensed scuba diver, etc. The following picture of me is a tryout of all above :).

Scuba diving at Hikkaduwa

So looking forward to pursue the happiness with a new set of challenges :).

A Scalable Multi-tenant Architecture for Business Process Executions

The journal paper “A Scalable Multi-tenant Architecture for Business Process Executions” is our latest publication which discusses in detail our work on extending scalability and multi-tenancy support to Business Processes. It was recently published in International Journal of Web Services Research (IJWSR), vol. 9(2), pp. 21-41, April-June 2012. Authors of this paper are Milinda Pathirage, Dr. Srinath Perera, Indika Kumara, Denis Weerasiri and Dr. Sanjiva Weerawarana.

Abstract of the paper is
Cloud computing, as a concept, promises cost savings to end-users by letting them outsource their non-critical business functions to a third-party in pay-as-you-go style. However, to enable economic pay-as-you-go services, the end-users need Cloud middleware that maximizes sharing and support near-zero cost for unused applications. Multi-tenancy, which let multiple tenants to share a single application instance securely, is a key enabler for building such a middleware. On the other hand, Business processes capture Business logic of organizations in an abstract and reusable manner, and hence play a key role in most organizations. This paper presents the design and architecture of a scalable Multi-tenant Workflow engine while discussing in detail the potential use cases of such architecture. Primary contributions of this paper are motivating workflow multi-tenancy, and the design and implementation of a scalable multi-tenant workflow engine that enables multiple tenants to run their workflows securely within the same workflow engine instance without modifications to the workflows. Furthermore, the workflow engine supports process sharing and process variability across the tenants and discusses its ramifications.

Paper can be accessed from here.

A journey through the history of computers

Computer History Museum, a place I like a lot in Mountain view, California. It’s the world’s largest history museum which explores the computing revolution and its impact on our lives. The journey starts from earliest calculating devices like "Abacus" to every landmark in the modern digital era.

Some of the pictures from the visit.

At the entrance

ENIGMA

Who invented the computer?

The evolution of programming languages

An early IBM server

An early Google server

Some thoughts about .com boom and bust

On the way to the museum 

WSO2 BPS v3.0.0 Beta is out

WSO2 BPS v3.0.0 Beta is out The release candidate of WSO2 BPS v.3.0.0 was released with lot of long awaited features and with a load of bug fixes.

Some of the new features in the Beta release are

• WS- Human Task support 
• WS- BPEL4People support 
• BPEL Instance Illustrator 
• BPEL deployment descriptor editor 
• Improved BPS management console and BPS documentation 

You can download the binary distribution from here.

1st Google+ Hangout

This is cool and runs smooth with an average 3G cellular connection.

Configuring a log4j logger to WSO2 ESB main sequence

Configuring separate loggers to different proxy services is a useful feature in WSO2 ESB.
You can do this by setting a log appender as follows.

Here “__SynapseService” is the SERVICE_LOGGER for the main sequence. Similarly you can use the name of your proxy service as the SERVICE_LOGGER to manage logs separately for each of those proxies.