The High Performance On Demand Solutions (HiPODS) team develops innovative
solutions to challenging on demand business issues, with an emphasis on
optimizing IT resources and helping IBM customers evolve into on demand
businesses. The HiPODS team is committed to sharing its work through a variety
of white papers describing best practices, lessons learned, and technology case
studies. This page presents the HiPODS papers.
Cloud Interoperability: Porting Google App Engine Applications to IBM
Middleware (March 2009)
Google App Engine is a cloud computing platform that hosts third party Web
applications. Application authors write their application front ends in a high-level
language like Python or Java™ Applications can use other Google services
like datastore and Google Accounts as their application back end. Authors upload
the application front ends through the Admin console to the Google App Engine.
We at IBM Cloud Labs have produced interoperability code/documentation to allow
GAE application developers to move easily to IBM middleware. This document shows
how to port the three major Java-based Google App Engine APIs/functionalities to
the IBM middleware environment where application function can be expanded. The
APIs discussed are datastore, Users, and URL fetch. (PDF file)
Building a Compute Grid on Apache Hadoop using Cloud Computing: A case study
at the University of Pretoria (December 2008)
Computational intelligence (CI) algorithms, such as those developed by the
University of Pretoria, South Africa, require a lot of computing resource to
complete simulation and development in reasonable time. In an environment with
many students, where each student??s workloads can take days to complete,
managing workloads and resources can become a full-time challenge. Cloud
computing dynamically provisions, configures, reconfigures, and deprovisions
resources as required. This paper discusses how cloud computing was applied to
solve the scheduling and resource management requirements of the University of
Pretoria, as well as provide grid computing capability for testing CI algorithms.
(PDF file)
From Cloud Computing to the New Enterprise Data Center (May
2008)
This paper describes how the HiPODS team has harvested workload patterns, best
practices, and reusable assets from our worldwide customer cloud centers to
build a solutions framework for the new enterprise data center. We describe a
high-level infrastructure framework and its underlying technologies. We also
present examples of value propositions for new enterprise data centers based on
cloud infrastructures that have been built as of today. (PDF file)
To receive information about future workshops and seminars about this proposed
solution, send an email to hipods@us.ibm.com.
Introducing HiGIG: The HiPODS Global Innovation Grid (August
2006)
This paper describes the High Performance On Demand Solutions (HiPODS) global
innovation grid (HiGIG). The HiGIG solution, used in production by the HiPODS
team, is based on a composite service-oriented architecture pattern and provides
IT infrastructure management services for multiple, globally distributed IT
infrastructures. HiGIG also includes a service grid component that can integrate
with service catalogs and be used for discovering and managing globally
distributed services, and a collaborative grid environment that includes wikis,
blogs, and social networking tools. (PDF file)
Virtualization and Automation: Best Practices for a WebSphere Application
Server Environment / An update (April 2005)
This paper is an update to the July 2003 paper Architecture for
Virtualization with WebSphere Application Server Version 5. It explores the
topic of providing virtual application server environments for the JavaTM
2 Platform, Enterprise Edition (J2EETM) applications by using the
best practices and technologies for virtualization and automation. The High
Performance On Demand Solutions (HiPODS) organization has developed best
practices through many customer engagements. This paper reviews these best
practices and tools, and considerations for using them.
Provisioning
Provisioning Web Applications with IBM Tivoli Provisioning Manager (June
2007)
This paper reviews the findings from a project that used IBM Tivoli Provisioning
Manager to automate the construction of the development environment for a major
Japanese manufacturer. It also describes how IBM Tivoli Provisioning Manager
was customized to fit the customer environment, and how the project team solved
the challenges of the customer environment.
Techniques to Configure IBM Tivoli Intelligent Orchestrator to Manage IBM
pSeries Servers (April 2006)
This paper describes the HiPODS team's experience of configuring the IBM Tivoli
Intelligent Orchestrator and Tivoli Provisioning Manager to manage IBM pSeries
servers. It provides examples that illustrate, clarify, and, in some cases,
give supplemental information to facilitate and accelerate the set-up and
configuration processes.
Provisioning for On Demand Data Centers (May 2004)
This paper introduces an end-to-end automated provisioning solution called Rapid
Provisioning for IBM Tivoli Provisioning Manager. It describes architecture,
trade-offs, and lessons learned during proofs of concept for IBM customers. For
a demonstration of the Rapid provisioning solution, see Rapid
Provisioning Demo.
Provisioning Best Practices for On Demand Data Centers (June 2004)
This paper describes identified best practices for implementing an end-to-end
automated provisioning solution. It describes architecture, trade-offs, lessons
learned and best practices derived from experiences during proofs of concept for
IBM customers. The paper complements the Rapid Provisioning solution and can be
used as a starting point for implementing automated provisioning.
Provisioning Case Study: pSeries and AIX LPARS (November 2004)
This paper is a case study for setting up a system to provision and service
pSeries AIX logical partitions (LPARs) using IBM Tivoli Intelligent ThinkDynamic
Orchestrator. It assumes knowledge of configuration and build methodologies for
IBM pSeries and Tivoli provisioning. The information is based on the
experiences of the HiPODS team with large IBM customers.
Provisioning Case Study: WebSphere Application Server Version 5 (November
2004)
Seamless server provisioning to a Java2 Platform, Enterprise Edition (J2EE)
application server environment involves installing the operating system software,
configuring the network, installing the required middleware, and finally
deploying and enabling the applications to process requests. These tasks
consume time and labor and are often error-prone. This paper describes a
solution that uses the IBM Tivoli Provisioning Manager component of the IBM
Tivoli Intelligent ThinkDynamic Orchestrator to provision servers to WebSphere
Application Server Version 5 environments. The solution is based on the
experiences and best practices that have emerged from several successful proofs
of concept provided by HiPODS for large IBM customers.
Orchestration
Orchestration for On Demand Data Centers (May 2004)
This paper introduces an end-to-end integration scenario that shows cross-tier
management of service level objectives for WebSphere and DB2 using the IBM
Tivoli Intelligent ThinkDynamic Orchestrator. The paper describes the
architecture and customer scenarios useful for understanding and implementing
this new technology. The solution showcases new hierarchical autonomic control
loop functions. For a demonstration of the orchestration solution, see Orchestration
Demo
Orchestration Case Study: Integrating WebSphere and DB2 Orchestration with
Enterprise Workload Manager (October 2004)
This paper describes a technology proof of concept led by IBM's HiPODS team to
demonstrate that the IBM Enterprise Workload Manager (EWLM) can be integrated
with the IBM WebSphere and DB2 Orchestration solution to leverage EWLM's
performance monitoring functions and to extend the orchestration solution with
the response time metrics that EWLM aggregates.
Implementing Multiple Network Deployment Managers on a Single Node (October
2004)
This paper describes how to implement multiple instances of WebSphere
Application Server Network Deployment Version 5 on a single node and how this
configuration helps optimize the use of corporate resources and isolate critical
applications.
Business-driven IT
Case Study: SOA-based On Demand Service Delivery Solution at Star Technology
Services Ltd. (March 2006)
This paper describes a proof of concept implementation of an automated service
delivery solution built on a service-oriented framework. The solution is an
integration of assets and products using Web services interfaces. The core
solution addresses IT Infrastructure Library (ITIL) processes and is a standards-based
framework of technologies, assets, and products fulfilling customer requirements.
On Demand Service Delivery for SOA Environments (May 2005)
This paper describes a self-service delivery solution that is built on a service-oriented
framework. The proposed solution is an integration of assets and products based
on a service-oriented architecture (SOA) and Web services interfaces. It
supports the full life cycle of a service delivery from order placement through
final reporting and billing.
Automating to Become an On Demand Business (December 2004)
An on demand business is flexible and reacts quickly to changing business
conditions. This paper introduces how a business could achieve more flexibility
when it automates the communication between the processes of making business
decisions and maintaining the IT infrastructure. It describes what
transformations are required to enable such automation, and the benefits
achieved. It concludes with the steps you can take to start your business on
its journey towards improved flexibility and automation.
Smart Card Transaction Processing, A Case Study: Applying best practices to
improve performance (February 2008)
The IBM High Performance On Demand Solutions India team helped a major IBM
client upgrade their smart card online transaction processing application to
achieve high performance. This paper summarizes the tuning methodology and
Java2 Enterprise Edition (J2EE) best practices that were key to the performance
improvements that exceeded the customer requirements.
Testing and Analyzing Performance from the User's Perspective: Analyzing
Performance for e-business (September 2006)
Many tests are available to help understand and improve the performance of a Web
site. Tests that are performed from a user's point of view have their benefits
and limitations. This paper reviews tests from the user side that are useful in
analyzing Web site performance; it includes a detailed discussion of the data
that needs to be collected and analyzed, as well as available tools.
Case Study: Performance Testing a Financial Application using WebSphere
Business Integration Server Foundation (February 2006)
IBM conducted tests to provide a major financial services company with tuning
and capacity planning recommendations for the WebSphere Business Integration
Server Foundation component of their Enterprise Services Management (ESM)
environment by testing selected ESM business processes. This paper summarizes
the methodology and results of performance testing completed by IBM's High
Performance On Demand Solutions (HiPODS) team. The test results demonstrated
that WebSphere Business Integration Server Foundation successfully scaled and
performed to meet the customer's performance target.
High Volume Performance Testing: Best Practices for WebSphere Applications
(October 2005)
This paper describes best practices for benchmarking the performance of
WebSphere applications in Java™ 2 Platform Enterprise Edition environments
characterized by high volumes and complex back ends. These techniques are the
product of IBM's experience working with customers seeking to improve the
performance and availability of some of the world's largest Web sites.
Best Practices for Deploying Applications to Highly Available Web-Based
Systems (September 2005)
This paper summarizes the methodology and results of a test of automated
deployment of online applications. The methodology demonstrated the application
of WebSphere Application Server's cloning facility to enable a major online
customer to maintain high availability of Web-based applications, while
maintaining application performance during application deployment.
Case studies
IBM BlogCentral Version 2: Architecting for Performance (November
2006)
BlogCentral is an intranet blogging platform that allows IBM employees around
the world to share opinions on work-related topics. The office of the IBM CIO
asked the HiPODS team to study BlogCentral performance and implement changes to
improve response times and throughput. This paper describes the HiPODS effort to
develop and deploy BlogCentral Version 2 with significantly improved performance.
It includes a description of the lessons learned, particularly in the areas of
database tuning, and options for further performance improvements.
Case Study: Best Practices for Testing and Tuning the Performance of a High-Volume
Stock Exchange EAI Messaging System (September 2006)
IBM conducted tests to help a major stock exchange customer assess its message
exchange system and evaluate a solution to improve performance using WebSphere
Business Integration components. The tests demonstrated that WebSphere's
middleware offerings can scale and perform to meet the customer's objectives.
This paper summarizes the methods used to tune the EAI system's performance, the
results, and the best practices involved in the tuning work.
Advanced Query Management for Optimizing Data Warehouse Performance (August
2006)
This paper describes a case study--the first in Japan--of the implementation of
IBM DB2 Query Patroller Version 8.2 in a development project for a data
warehouse system. It describes the issues the client, a major Japanese
manufacturer, had in their environment, and how the issues were addressed and
solved. It provides key information for readers who want to know how to monitor,
manage, and control queries on DB2 Universal Database.
Case Study on the Performance of WebSphere Workloads on 64-bit Systems
(July 2006)
IBM HiPODS evaluated an application for a major US customer using the IBM 64-bit
solutions. The objective was to meet the customer??s objective of migrating from
32-bit to 64-bit systems without degrading application performance or using more
memory. This paper highlights the project results and demonstrates that IBM 64-bit
solutions, including hardware, operating system, and middleware, provided
superior performance and total cost of ownership.
Case study: Reduce Web Site Bandwidth Demand through Workload Analysis
(June 2006)
This paper describes a case study for a customer seeking to reduce bandwidth in
order to reduce costs. The study analyzes bandwidth demand and identifies which
components use the most bandwidth and which factors contribute most to bandwidth
use. The site and application studied primarily transfer text, so there are
several choices for optimization. These methods do not change application
functionality and do not require changes to the application. You can achieve
major reductions in bandwidth use by selectively using the methods described
here.
A Retail Data Warehouse Solution using DB2 for Linux (November
2004)
This paper describes the results of a proof of concept IBM completed for a major
retailer with a large data warehouse. HiPODs demonstrated its capacity for
creating a representative customer environment and proving how IBM's products
meet customer requirements. The key requirements for this PoC were the database
extract/transform/load process, scalability and performance of DB2 queries when
multidimensional clusters are implemented on the Linux platform, and database
provisioning.
Case Study: Performance Test on Financial Information System using WebSphere
Business Integration (October 2004)
IBM conducted tests to help a major financial customer assess its middleware
environment and evaluate a solution to improve performance using WebSphere
Business Integration components. The tests demonstrated that WebSphere's
middleware offerings can scale and perform to meet the customer's objectives.
WebSphere
A Methodology for Migrating from Axis to WebSphere Web Services (April
2005)
This paper describes a methodology to migrate a Web services application from
Apache Axis Web services to WebSphere Web services. It includes examples of some
code changes that might be required. The actual code changes required to migrate
depend on the features and APIs that the application uses.
A technique for monitoring WebSphere Application Server workload management
(March 2004)
This article describes a technique for using standard Web server directives to
log workload management activity in the WebSphere Application Server cluster,
and a mechanism for summarizing the behavior of WebSphere Application Server
workload management in real time.
Systems Administration and Scalability of Application Server JVMs with
WebSphere Application Server, Version 5 (October 2004)
This paper discusses the architectural considerations for administering systems
and scaling application server Java virtual machines (JVMs) for WebSphere
installations. It explains the setup of a laboratory environment and the
results of tests conducted to explore different issues. The report should be
useful to those architecting solutions for large-scale deployment environments
using infrastructure components such as shared storage.
Miscellaneous
Using VMware ESX Server with IBM WebSphere Application Server (July
2006)
This paper looks at the best practices when running VMware ESX Server and IBM
WebSphere Application Server on a large symmetric multiprocessor (SMP) server to
achieve the best application performance (throughput and response time). Our
tests concluded that the use of VMware software is an excellent option for those
wanting to run WebSphere Application Server in a virtualized SMP environment. (PDF
file)
VMware ESX Server 2: Performance and Scalability Evaluation (January
2004)
Representatives of VMware and IBM's High Performance On Demand Solutions team
conducted joint tests to evaluate VMware's ESX Server 2 in terms of overhead,
scalability, and priority handling. ESX Server 2 proved to be an effective
virtualization technology. This paper summarizes the test configuration and
results. (PDF file)
Sonoma: Web Service for Estimating Capacity and Performance of Service-Oriented
Architecture (SOA) Workloads (October 2006)
This paper describes Sonoma, a Web service for estimating the performance and
capacity of service-oriented architecture (SOA) workloads. Sonoma is built using
the IBM-developed technology of the On Demand Performance Advisor (OPERA). OPERA
is a time-tested capacity planning tool that employs queuing theory, advanced
autonomic algorithms, and extensive measurements with IBM customers. Sonoma
provides IT architects with performance modeling and hardware configuration
recommendations by sizing four SOA workloads based on common usage scenarios
such as travel booking, banking, and insurance claims handling.
Case studies
HVWS Simulator: An eBay Case Study (February 2004)
This paper describes how IBM's HVWS Performance Simulator for WebSphere was used
to analyze the capacity of the code supporting one of eBay's high-volume and
revenue-earning features. The objective of the analysis was to assess whether
eBay's IT architecture could support current transaction throughput rates and
meet future growth needs.
HiPODS Model: An eBay Case Study (June 2004)
This presentation to the IBM Academy describes how IBM's HVWS Performance
Simulator for WebSphere, now referred to as OPERA, was used to analyze the
capacity of the code supporting one of eBay's high-volume and revenue-earning
features. The objective of the analysis was to assess whether eBay's IT
architecture could support current transaction throughput rates and meet future
growth needs. This presentation corresponds to the paper listed above, "HVWS
Simulator: An eBay Case Study."
Extreme Availability with WebSphere and DB2 (November 2004)
The New York Stock Exchange (NYSE) has extraordinary availability requirements.
It relies on the Securities Industries Automation Corporation (SIAC) to ensure
those requirements are met. SIAC chose IBM to provide the hardware and software
infrastructure for a new trading floor application. This paper describes this
unique challenge, the resulting availability partnership and solution that began
a successful deployment in April 2004, and enhancements to IBM software that
demonstrate IBM's strengths in the realm of extreme availability.
End-to-End Banking Transformation in the
Product Domain (May
2009)
Banking organizations face continuing pressure to transform and modernize their processes and applications to meet today's business demands and requirements. This paper describes the architeture, design, and implementation of a solution that uses SOA technology to enable banking transformation in the product domain. (PDF file)
Web 2.0 -- Advanced AJAX Applications with the Google Web Toolkit (February
2007)
This paper is a guide to using the Google Web Toolkit (GWT) to integrate rich
user interface components with remote services. It includes examples showing
how to create a paged table with editable fields and a personalized dialog box.
It discusses the management of events and service calls for data access, as well
as peculiar aspects of the API and tips about its use. GWT improves productivity
when developing AJAX applications for the Web. (PDF file)
IBM Tivoli Access Manager for e-business / Quick Prototyping with Version 5.1
(October 2005)
This article is a quick guide to installing and specifying a basic configuration
for IBM Tivoli Access Manager for e-business Version 5.1. It illustrates the
product's basic functions with two examples: securing access to several Web
applications and single sign-on. The goal of this article is to assist users
with prototyping and quickly deploying Tivoli Access Manager in conjunction with
the more comprehensive information that accompanies the product. (PDF file)
