Ramanjaneya Sarma Burugula

Web applications and dynamic reconfiguration in UNIX servers

In recent years, several large UNIX SMP Servers have added support for dynamic resource management through partitioning and dynamic resource reconfiguration. In this paper we study the ability of Dynamic Reconfiguration (DR) to accommodate fluctuating workloads and changes in operational priorities for a commercial web application. We use a WebSphere HTTP server, a WebSphere Application Server, and a DB2 Database for the application. This combination represents a popular platform for commercial computing deployments, and supports a number of common web-based application scenarios. In our study, we treat this application as a black box to provide a realistic measurement of the efficacy of the DR technology in UNIX Servers. We also use nonparametric estimation techniques to obtain an ab initio and unbiased study of the jitters in our experimental data. Our main conclusions are: (1) Resource allocations for the application (even for a complex and function-rich middleware system such as WebSphere) can be efficiently managed by DR, without the need for explicit accommodation of the DR features by the application, and (2) To obtain efficient resource utilization, the resource management system has to empirically monitor the throughput obtained from the application, rather than rely primarily on long time-scale estimations.

Dynamic reconfiguration of CPU and WebSphere on IBM pSeries servers

This paper studies the effects that dynamic reconfiguration (DR) has on a WebSphere workload while CPUs are dynamically added to and removed from the underlying AIX instance. DR is a new technology available in AIX 5.2. This study shows that the resource allocations for a complex and function‐rich middleware system such as WebSphere can be efficiently and dynamically managed by the DR technology, without WebSphere having to explicitly accommodate for the DR features of the operating system. Copyright

AHAFS subsystem for enhancing operating system health in the cloud computing era

Cloud computing is becoming an important paradigm for information technology (IT) infrastructures. A key advantage of cloud computing is its ability to decrease the cost of computing for a variety of applications. For the infrastructure layer, which includes servers and operating systems (OSs), cost reduction is achieved through the consolidation of many OS instances onto a single physical server and through significant improvements in system-administrator productivity. However, these productivity improvements require a scalable easy-to-use OS event-management subsystem so that the higher level software can reliably and effectively automate operations. This paper presents the design and implementation of an OS event-monitoring subsystem called Autonomic Health Advisor FileSystem (AHAFS) that is scalable, has little system-performance overhead, and provides instantaneous event notifications with useful information, for improving the robustness, security, and performance of an OS instance. AHAFS is extensible and can monitor various events for a wide variety of event consumers. AHAFS does not require the use of a new application programming interface (API) but uses the ubiquitous filesystem API; hence, AHAFS is usable by monitoring applications written in any of the languages commonly used for monitoring (e.g., C, C++, Perl®, and Java®) without requiring additional runtime modules or packages.

Automation of Cloud Node Installation for Testing and Scalable Provisioning

Motivated by the lack of automated provisioning and testing tools for physical systems in modern cloud infrastructures, we developed a toolset to automate the installation of a commercial platform virtualization software: the IBM PowerVM NovaLink software. This toolset automates numerous manual installation steps and post-installation tests. The toolset uses Python programs to prepare the target system and to generate installation-configuration files through interaction with a web-interface of the POWER firmware. Expect scripts are used to drive the rest of the installation via interactions through a remote text console. With different scenarios of applying this automation toolset, we demonstrate its use in driving regression testing for the development of the NovaLink installer in a DevOps environment. Furthermore, we discuss its use in provisioning a large number of IBM POWER systems to be ready for cloud management software.