IBM Systems Magazine, Power Systems - January 2018 - SE18
ARTICLE: HA/DR BACKUP
Preparation Is Vital to
Crafting an HA/DR Plan
-suite executives know that high availability (HA) and
disaster recovery (DR) are two key requirements for their IT
systems. In some industries, such as banking, legal compliance
issues dictate HA and DR requirements. As executives often
assume that HA and DR structures are in place, it's best for them
and systems exist.
Enterprise customers, as well as small-and medium-sized
businesses, have many options to ensure their HA/DR programs
are robust. Money-saving options include cloud-hosted HA
and subscription-based software offerings and DR as a Service
(DRaaS). DRaaS involves renting a partition from a managed
service provider (MSP), which the customer uses to host the DR
replicate data. The MSP lends a hand with setup, monitoring,
backups, role swap testing and other necessary tasks.
Not Just Storms
When it comes to HA/DR, most IT shops think in terms of natural
disasters. While Mother Nature can wreak havoc, other threats
can be just as disruptive to operations, says Peter Laz, managing
consultant at Forsythe Solutions Group Inc. These include local
power outages, a squirrel chewing through wires, human error
or a cyberattack. (See "Expert Recommends a 5-Step Process
to Develop a Disaster Recovery Plan," bit.ly/2yHh4Sv and
No matter the cause of the disruption, the methods for
handling it remain the same. First, establish a corporate crisis
management team (CMT). The CMT has "the responsibility to
direct and/or manage the full lifecycle of any event that could
impact business for an acceptable period of time," Laz says. The
CMT will call on subject matter experts, depending on the type
Effective communication is important when handling DR
situations. Giving an accurate description of the problem will be
helpful when communicating with outside entities, Laz says.
Creating a DR Plan
1. Establish guidance. This involves
the development of standards and
18 // 2018 Learn more at: ibmsystemsmag.com/buyersguide
policies to guide DR and the creation of governance rules.
2. Determine requirements. This includes looking at the risks
to the company, the recovery time objective, the recovery
point objective and assessing your company's capabilities to
3. Select a strategy. A company will want to review strategic
Once an option has been chosen, the company will want to
create an implementation roadmap.
4. Implement capability. This step centers on creating a
implementation plan and should include validating the
5. Govern the program. The chosen DR plan is put into place.
The company also needs to run validation exercises to
ensure the DR capabilities work. This is also the stage at
which compliance and risk reporting is implemented.
Any DR plan must be reviewed and revised regularly. Plan and
program maturity will determine how often such a review must
occur. Plans likely will go through several versions before all of
the glitches have been addressed. "Program maturing is also a
factor because the higher-quality programs have operationalized
the business continuity and disaster recovery environment;
reviews and updates to the plan are built into the daily operation
of the business," he says.
DR processes are necessary and can even help companies
save money. Laz notes that a full-service bank implemented
a multitiered recovery program that included virtualization
consolidation. The bank saved $600,000 over three years.
Hurricane Sandy in 2012 convinced Chilewich Sultan LLC,
a designer and manufacturer of high-end woven goods, to
revamp its DR situation. The company worked with Maxava to
develop a streamlined DR solution that involved the installation
of Power Systems* hardware as well as moving some data to the
cloud. HA was part of the company's revamp and resulted in realtime backups as well as reduced rollover times from 24 hours to
just four hours. (See "Chilewich Sultan
Weaves Together a Custom Disaster
Recovery Solution," bit.ly/2yM6sUw.)