Succeeding with Business Agility: Five forward-thinking Organizations

Part 1: Nationwide Insurance Saves Money, Increases Agility with IBM's Cloud Delivery Models
An InformationWeek Webcast Sponsored by

Webcast Logistics

Todays Presenters
Jeff Imholz, Senior Architect - Middleware Technologies, Nationwide

Alex Hay, Senior Engineer - Middleware Technologies, Nationwide

Don Boulia, Product Management Director, IBM

WebSphere Increases Agility in IBMs Cloud Delivery Model

Don Boulia Product Management Director, IBM

A Smarter Planet creates new challenges for the IT infrastructure 5 Weeks

Average time needed to deploy a new application

75%
Our World is Becoming:
Capital utilization that can be gained from cloud computing

interconnected intelligent

1 second delay
In response time results in 7% in lost sales and lost revenue

70%
Of IT spending is for maintaining current infrastructures

1. Mobile Internet Research Report reveals Massive Mobile Internet Growth, December 23, 2010 2. Scale Up with Elastic Caching Webcast, Forrester Research, February 2010

3. Global Internet Trends you should know: From the Meeker Files, December 23, 2010 4. The Performance of Web Applications: Customers Are Won or Lost in One Second, Bojan Simic, Aberdeen Group, November 2008

Business Agility is a necessity for organizations to dynamically adapt and respond to these challenges

Rising consumer expectations compel improvements in speed and reliability

Increasing IT demands with flat or decreasing budgets require greater returns on IT investments

Greater uncertainty in the marketplace requires the ability to respond to change faster

Sources: Predicting SOA Business Agility through Metrics, IBM Academy Best Practices Conferences, 2010

Delivering and scaling services quickly is key to Business Agility

Deliver and scale new services leveraging private clouds by:

Deploying new applications and SOA services faster Reducing response time of existing applications with elastic caching

Reduce Costs! Leveraging rapid cloud deployment and enhanced scalability through elastic caching

Leverage private clouds to deliver new applications and SOA services faster
Deliver new services faster with self-service cloud management and the convenience of an appliance Reduce complexity with consistent & repeatable deployment of hardened WAS patterns into a cloud of virtualized hardware running a hypervisor Integrate with existing infrastructure through programmable REST APIs

IBM Workload Deployer

Improves time to value by enabling development and test teams to roll out new applications and SOA services faster

IBM Workload Deployer v3.0 New Capabilities!

Virtual Application Support

Enhanced patterns with full lifecycle management

Autonomic Elasticity VM Mobility IP & Naming Control

Intelligent Mgmt Pack enabled topology patterns scale up/down according to observed demand
User may request that a virtual machine is moved from its present location in order to optimize placement

Users have more control over IP address and name assignment of dispensed nodes in topology patterns.
The appliance can hold more images and patterns with over 6x the storage capacity of the WCA v2.x appliance

Larger Storage

Expanded WebSphere Application Server Hypervisor Edition Support

Red Hat z/Linux Support

64 bit Linux support

An image is now available on Red Hat z/Linux

Linux images are now available atop a 64-bit OS. Note that on X86 Linux, WAS is still 32 bit.

Enhance scalability and reduce response time with Elastic Caching

Reduce response time by placing data close to running transactions Speed application performance for a powerful distributed cache for access to data Dynamically adapt to changing infrastructure demands

IBMs innovative Elastic Caching solutions

Data Oriented
Session management DataPower XC10 Appliance Drop-in cache solution optimized and hardened for data oriented scenarios Elastic DynaCache Web side cache

Worldwide cache
Data buffer Event Processing Petabyte analytics In-memory OLTP In-memory SOA eXtreme Scale Ultimate flexibility across a broad range of caching scenarios In-memory capabilities for application oriented scenarios

High density, low footprint improves datacenter efficiency

Application Oriented

Elastic caching for linear scalability High availability data replication Simplified management, monitoring and administration

Simplify your Application Infrastructure environment with Elastic Caching

Web Server Tier App Server Tier
1
DataPower XC10 for simple data oriented scenarios:
HTTP Session Replication Elastic Dynacache Web Side Cache

Elastic Data Grid

XC10

Database Tier

2
eXtreme Scale for maximum flexibility covering data and application oriented scenarios

IBM HTTP Server

WebSphere Application Server

DB2 UDB

Challenges facing Nationwide prior to the IBM implementation

Speed to market Maintaining level of consistency Siloed data sources Inefficiency redundant data Cost and control issues

IBM business value

Thanks to our partnership with IBM, we are able to utilize our system resources to the top of the glass rather than only of it. Alex Hay, Senior Engineer Middleware, Nationwide Nationwide can now deploy a full test system in one week enabling a previously unrealized level of innovation

Other solutions considered

Nationwide considered options including VMwares Lab Manager solution but chose the IBM solution due to requirements for zLinux integration

15

WebSphere Strategy
A foundational approach to private cloud
Alex Hay and Jeff Imholz
8/24/2011

BUSINESS DRIVERS
Reduce the costs associated with design and provisioning.
Improve operational process and performance through consistent infrastructure delivery. Drive higher levels of availability by enabling dynamic capabilities.

8/19/2011

17

SERVER VIRTUALIZATION
Addressed these opportunities
Improved environmental sensitivity Repeatable system configuration Increase server admin efficiency
Created

Resource over allocation Server sprawl Share Scare

Reduced server deployment time Basic migration capabilities Increased hardware utilization

Virtual Servers
Physical Servers

8/19/2011

18

APPROACH
WebSphere strategy
Centralized configuration management Dynamic application placement Integrated topology deployment
Integrated image (OS & MW) Advanced application mobility Directed hardware utilization Reduced server deployment time Basic migration capabilities Increased hardware utilization Process refinement Highly scalable Increased performance

Virtual Platform

Virtual Servers Physical Servers

8/19/2011

19

GOAL STATE
WebSphere test environment
Metric Number of Cells Current State 32 Goal State 2

Running Instances
Number of Servers Instances per Admin Instance Overhead
8/19/2011

627
160 120 10%

275
60 150 3%
20

SOLUTION COMPONENTS
IBM
Workload Deployer

Intelligent Mgmt

Rational
Automation

Dynamic scaling Standardized delivery Configuration consistency

WebSphere Strategy

Framework

Hypervisors

DataPower XC10

8/19/2011

21

DYNAMIC SCALING
DataPower XC10

Scalable session cache Provide linear performance for web services

Intelligent Mgmt

Enable burst capacity for unanticipated demand Scale nodes to support workload demand

IBM Workload Deployer

Scale application servers for workload characteristics

8/19/2011

22

STANDARDIZED DELIVERY
Rational Automation Framework

Integrated platform image Pattern based topologies Solution level focus

Intelligent Mgmt

Repeatable infrastructure Enables low-touch deployment

IBM Workload Deployer

Replace current multi-day process

8/19/2011

23

CONFIGURATION CONSISTENCY
Automate application configuration Application deployment

Rational Automation Framework

Version control for all artifacts

IBM Workload Deployer

Centralized point of control Artifacts can be reconstructed as needed

8/19/2011

24

NEXT STEPS
In progress implementation Development servers Application and schema deployments

8/19/2011

25

Q&A Session
Jeff Imholz, Senior Architect - Middleware Technologies, Nationwide

Alex Hay, Senior Engineer - Middleware Technologies, Nationwide

Don Boulia, Product Management Director, IBM