WHITE PAPER EXECUTIVE SUMMARY Synopsis: IBM

---

WHITE PAPER EXECUTIVE SUMMARY Synopsis: IBM Power Systems and Their Support for Business Resilience in Challenging IT Environments Sponsored by: IBM Jean S. Bozman

Randy Perry

April 2012

Global Headquarters: 5 Speen Street Framingham, MA 01701 USA

P.508.872.8200

F.508.935.4015

www.idc.com

The Challenge The increasing competitiveness of the global business environment puts direct pressure on IT organizations — and the businesses they support. As never before, IT must deliver consistent, secure, and fast services that "don't go offline." These real-time businesses require what IDC refers to as business resilience, which is the ability to deliver high levels of availability to protect the business against outages. Business resilience keeps business services flowing to end users and end customers alike. Downtime disrupts the workflow within a business — and it costs an organization dearly in terms of lost revenue, resources to address the outages, reduced employee productivity, and, eventually, damage to the company's brand with loss of end customers. Downtime can be divided into two categories: planned (for maintenance) and unplanned (resulting from outages). The realities of today's global business environment — which runs 24 hours a day, 365 days a year — make it difficult for organizations to justify even small windows of planned downtime. IDC surveys indicate that most organizations (over 70%) feel that even one hour of downtime can impact the business severely (cause "severe" or "moderate" impact). These same organizations, according to the surveys, are investing in highly available servers and systems to protect business continuity. The following figure shows that downtime costs can increase exponentially (note the logarithmic plotting of revenue loss due to downtime), depending on the industry, and the relative levels of downtime (circle size) reflect these increasing costs. 100,000

Financial Services 3.57

10,000

1,000 Retail 7.75

Revenue Loss Per Downtime Hour Log Scale ($K)

Healthcare 21.7 100

Manufacturing 8.01

10

Public Sector 9.41

1 -

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

Productivity Loss Per Downtime Hour ($K)

The financial services vertical market sees nearly $10 million in revenue loss per hour and $3,640 in lost productivity per hour as a result of downtime; however, the investments made by this vertical market in reliability, availability, and serviceability (RAS) have paid off. As we can see, financial services has the lowest amount of downtime across industries at 3.6 hours per year. At the other end of the spectrum is healthcare, which has relatively low $157,000 revenue loss per hour and $1,250 productivity loss per hour but the highest amount of downtime at 21.7 hours per year. Public sector organizations suffer the least productivity loss per hour due to downtime, but they experience 9.4 hours of downtime per year, which is similar to the levels of downtime experienced by two commercial verticals studied: retail and manufacturing.

Reducing Downtime Through Resilience Organizations ensure availability in different ways: by duplicating datacenters, paying more for RAS features on server hardware, using clustering software, replicating production-level data, and more. Increasingly, customers are leveraging virtualization to restart applications with cloned virtual machines (VMs) running on other servers. However, most IT organizations surveyed prefer built-in server features that ensure high availability. IBM Power Systems, based on IBM POWER7 processor technology, address this requirement. They provide high performance and reliability in a server system that is offered in many form factors — from blades and rack-optimized servers to high-end supercomputers. These systems integrate RAS features across the chipset, server, and software levels.

Results Research indicates that these features reduce operational costs for the IT organization and the business it supports. Seventeen IT organizations that consolidated workloads, which had previously been deployed on large numbers of small servers, onto IBM Power Systems found that their combined system downtime dropped by as much as 84% over a period of three years or more (IDC Business Value Research). The costs of downtime, as discussed in this paper, are based on the IDC Business Value consulting team's model for calculating the business impacts of downtime and how downtime affects companies across different vertical markets. The model is based on IDC's Business Value Database, which is populated with customer survey data that is gathered and analyzed by the IDC Business Value team and addresses a variety of impacts on downtime, including IT productivity, enduser productivity, maintenance costs, and revenue loss due to downtime. This IDC Business Value data has been collected over a period of more than five years from over 3,000 companies in 43 countries and more than 25 industries.

Resilience at the Hardware Level Many organizations choose to build redundancy into the computing environment to achieve business resilience, but many approaches can be taken to achieve this goal. One way is to build duplicate datacenters with the ability to provide failover between the two. This can be seen as a form of insurance against natural disasters or outages that might disable one datacenter — but not others. Another, less costly, approach is to make sure that each server node is resilient as a standalone computing component. This is particularly important for servers that are running large databases or important enterprise applications that are being accessed by hundreds or even thousands of end users. Any downtime for these types of servers would have far-reaching effects across the organization. IDC's Mission-Critical Workloads Study determined that the majority of respondents (59.1%) prefer to have highavailability solutions for their server systems at the hardware level rather than at the software or middleware layer. Embedding RAS features in the hardware platform means that less IT staff time will be devoted to one-off customization of software scripts to manage a failover or restart of a critical application on alternate server resources.

IBM Power Systems Resilience Features IBM Power Systems have a number of RAS features designed in to provide high levels of business resilience. These features start with the fact that Power Systems are built on POWER7, the latest generation of IBM's POWER chipsets. POWER7 includes a number of built-in RAS features to provide resilience at the silicon level. In addition, Power Systems servers contain critical RAS features at the server level. All Power Systems servers support redundant/hot-swap fans and blowers, hot-swap DASD/media/PCI adapters, redundant/hot-swap power supplies, and dual disk controllers with a dual backplane. They also support concurrent firmware updates, processor instruction retries, alternate processor recoveries, and storage keys. Finally, IBM PowerVM micro-partitioning, which is available as an option on all IBM Power Systems servers, provides the ability to run up to 10 partitions per server core and to dynamically move processor, memory, and I/O resources between partitions to support changing workload requirements. Active partitions can be moved between servers, reducing the amount of downtime for planned systems maintenance.

Conclusion In an increasingly global business environment in which employees, customers, and business partners need to access critical applications at any time of the day or night, downtime of any type — planned or unplanned — can be extremely costly. Downtime can start with a small problem and end up costing an organization millions of dollars a day in lost revenue. That's why many organizations are looking for server solutions that offer high-availability features, built-in hardware-based RAS features, and support for disaster recovery to help them maintain business continuity. IBM Power Systems, based on IBM POWER7 technology, incorporate a number of RAS features at the chipset, server, and software levels and are designed to ensure high levels of availability for business workloads. The business resilience features and offerings found in IBM AIX, IBM i operating environments, IBM Power Systems, and IBM PowerHA high-availability software provide a comprehensive technical approach that is designed to protect the continuity and availability of business-critical applications and services.

Copyright Notice External Publication of IDC Information and Data — Any IDC information that is to be used in advertising, press releases, or promotional materials requires prior written approval from the appropriate IDC Vice President or Country Manager. A draft of the proposed document should accompany any such request. IDC reserves the right to deny approval of external usage for any reason. Copyright 2012 IDC. Reproduction without written permission is completely forbidden. 2

#233283ES

©2012 IDC