Compaq

Reliability: The Ultimate Mission-Critical Mandate
(Continued)

Know Cost of Downtime

The Gartner Group considers five top actions items for high availability:
-Know the cost of downtime.
- Develop an availableility strategy and plan, justifying the investment based on the reduction of downtime.
- Invest in maturing IT management processes toward service-level management.
-Measure availability and response time from a user's perspective.

- Design application architectures for availability, reducing complexity whenever possible.

Compaq addresses these needs through its PDIM Lifecycle, which stands for plan, design, implement, and manage. The Lifecycle ranges from availability planning and assessment to availability monitoring and disaster recovery. Compaq availability experts will provide an in-depth analysis of a computer multivendor computing environment, comparing the current state of a firm's IT installation with business and availability goals.

The aAvailability rReview investigates all factors of an IT installation related to availability, calculates the cost of downtime per system, and simulates improvement scenarios to achieve the best ROI.

Availability solutions must be designed for an individual company's needs. "The main lesson we've learned over the last year is that 100 percent availability is achievable, but at a pretty high cost," says FTT's Grammer. "There are levels of availability that may more appropriately match what we're trying to achieve."

No question, though, that more and more firms are realizing the importance of having proactive and reactive services in place to ensure availability of mission-critical systems. "We spend a lot of time partnering with customers to understand their environment and prevent problems from happening, from upgrade planning to problem notification," Compaq's O'Neill says. "We might see an engineering change coming out and tell a customer they need a patch right away because it could cause problems down the road."

Having this level of familiarity requires a lot of diligence. Compaq conducts at least one annual site review annually for its premium customers. Data collectors also run on those customers' systems, putting essential information at their fingertips of Compaq technicians during emergencies. Technicians can click to a web site and immediately see the customer's current configuration, operating systems, and other technical information. Compaq also uses data collectors to monitor system availability. On a monthly basis, Compaq reports on system availability, uptime, and root cause failures.

Supporting Applications
The ever-increasing complexity of availability has also caused Compaq to move up the value chain and support the entire IT environment. of customers. "In the past, the services division focused on the hardware and operating system," O'Neill says. "We would get a customer up and running after a hardware or software failure, but the customer would still have to get his third-party applications running." Now Compaq is offering SAP/R3 Monitoring Services, extending the same level of protection and services to critical applications.

Availability issues goes well beyond purely technical issues. Compaq's premium service customers are assigned a technical account manager who becomes intimately knowledgeable about their customer’s environment and business.

"When you're having a problem, there is nothing more frustrating than talking to someone who doesn't appreciate your business pressures, such as your peak purchasing times," O'Neill notes.
For similar this reasons, customers calls are routed to their assigned technical account manger in usually talk to the same person in Compaq's cCustomer sSupport cCenter. "This kind of personal service helps us understand a company's strength and weaknesses," O'Neill says. "The other day, for example, a technical account manager realized that a firm had a crash because their third shift operator needed more training. You might be nervous to mention something like that to a customer if you hadn't built up a relationship based on personal credibility."

A well-designed highly-available system makes IT disruptions transparent to the world users. Using high-speed networks and OpenVMS clustering technology, Compaq offers disaster-tolerant solutions that link together two data centers up to 500 miles (800 kilometers) apart.

During normal operations, the two sites communicate and share resources. In the case of a mishap, one center can take over the entire operation for a company. For OpenVMS , cluster configurations can range from 2 to 96 systems, enabling massive scalability and availability.
FactSet maintains duplicate data systems in New York and Greenwich, Connecticut, each of which is powered by six Compaq AlphaServer GS140/525 systems running the OpenVMS V7.1-2 operating system. Both are hot sites, designed to run at less than 50 percent capacity, so that in the event of a shutdown one data center can take over the entire load of the other.
"The old disaster-recovery center model assumed you could afford to be down for one to three days while you rebuild your data center," Zorn says. "But business has changed so much that you can't afford an extended outage -- you'll be out of business. Our downtime should be invisible to clients."

Compaq offers 24x7 access to a fully configured backup system and office facilities at a secure recovery center. Mobile recovery systems are also available. Even with the utmost planning to ensure high availability, companies must prepare for all contingencies. "When talking about clustering, high availability, and uptime, we've even had people in our company refer to that as the disaster recovery plan," Grammer says. "But you still have disasters that come to you in different ways that high-availability systems don't help with, such as outside equipment failure and storm water damage. A sound backup strategy is still important."

What's most important, though, is keeping critical systems running -- that was the number one IT concern csited in a survey by Ernst & Young. That's why many availability experts no longer use the term "disaster recovery" but prefer to say "business continuity." Because business must be prepared to continue in all circumstances. And that's why Compaq offers availability services that take into account all circumstances.

"If an exchange stops because of a systems failure, the entire society is affected," Egervall says. "Anyone who is trying to start any kind of online exchange now can learn lessons from the fully automated financial exchanges, which have been living in a b-to-b ecommerce environment for 15 years."

Compaq technology has always formed the basis for OM's systems. "Maximum uptime is vital for the success of an electronic exchange," notes Anne von Corswant, OM Technology's vice president of market research for financial products. "Compaq OpenVMS provides a stable and disaster-tolerant platform that means our exchanges can rely on 100 percent system software uptime." That's vital, analysts concur, because most users will accept no less these days. (next)

 

1 2 3 4 5

 

Go Back to Samples