This paper presents an overview of iSeries high availability to companies that are beginning to explore this powerful business continuity technology. The critical components of high availability solutions are detailed in this paper, including data replication engines, system monitors, role swap capabilities, and the importance of autonomic processes.
An Introduction to System i
High Availability
Get to know the core components
of HA solutions and important
installation considerations.W H I T E P A P E R
Introduction
Every company faces critical hours when system downtime is unwelcome-whether
it's planned or unplanned. One company's important hours might only be from
9-to-5, while for another it's 24x7. Increasingly, shops that were able to accommodate
some periods of downtime for backups and system maintenance are finding that this
window is quickly shrinking (or has disappeared altogether) because of increased
system demands-especially from global and online business. Because of the need
to keep systems available for increasing amounts of time, companies are realizing that
a system failure or a site disaster would create an enormous disruption and expense,
especially if it went on for longer than a few hours. For many companies, exposure to
this amount of potential downtime has become unacceptable. Shops that thought
they weren't candidates for a high availability solution are now realizing an urgent
need to start looking at their options.
The purpose of this white paper is to provide an introduction to System i high
availability for companies that want to understand this technology and evaluate
whether such a solution can become a cost-justifiable component of their backup
and recovery strategy.
Before looking at the details of System i high availability, lets take a quick look at the
cost of downtime and some of the primary disaster recovery strategies that are used to
mitigate this cost.
v i s i o n s o l u t i o n s . c o mW H I T E P A P E R
The Cost of Downtime
The rising cost of Management is often amazed when they add up all of the direct and indirect costsdowntime has caused of what downtime costs their company. At first, they may figure that if a system isdown for several hours or even a day, it is certainly a big inconvenience and a tolerablemany companies to risk-as long this kind of downtime is a rare occurrence. However, once numbers areconsider HA that plugged into the following rule-of-thumb formula, they are often shocked:may not have Take the average sales lost during an hour of system downtime during businessconsidered it before. hours, then add the total hourly wage (including benefits) of all employees that areidle during that hour of downtime. Now multiply this figure by the estimated numberof hours of system downtime during a year. Finally, multiply the result by 2 to takeinto account the costs of this lost employee productivity, lost business reputation, andlost business-both now and in the future-from your lost customers.
Unplanned Downtime vs. Planned Downtime
The IBM System i is considered the most reliable business system in the industry; somestudies put its reliability at 99.95%. But consider the following from the IBM Redbook,Clustering and IASPs for Higher Availability on the IBM eServer System i Server:
"According to one IBM study, the System i server averages 61 months betweenhardware failures. However, even this stellar record can be cause for availabilityconcerns. Stated another way, 61 months between hardware failures means thatnearly 67 percent of all System i servers can expect some type of hardware failurewithin the first five years."
Other sobering statistics to consider:It is the downtime from . A ccording to the Hurwitz Group, after the loss of a system for a week 43% of businesses planned system never reopen and another 29% close within two years. maintenance events . A ccording to Gartner, 93% of all companies that experience 'significant data loss' are out that cause IT shops of business within five years.the most pain. Despite the potentially dire consequences of unplanned downtime, less than 10% of alldowntime can be attributed to unplanned events, and only a fraction of that is due toa site disaster. The other 90+%-the kind that companies face on a regular basis-arethose caused by system maintenance tasks, including:
. D ata backups (nightly, weekly, and monthly saves). R eorganization of files-to reclaim disk space and improve performance. V endor software upgrades & data conversions. IBM software release upgrades & PTFs. New application software install... [download for more]
