|Read the Digest in
You need the free Adobe
BCP tells you how to recover from the effects of downtime.
CPA tells you how to avoid the effects of downtime.
In this issue:
Browse through our Useful Links.
Check our article archive for complete articles.
Sign up for your free subscription.
Join us on our Continuous Availability Forum.
Join Us for Discussions on Continuous Availability at the NonStop Symposium 2010
Our Managing Editor, Dr. Bill Highleyman, will present a preconference seminar on Achieving Continuous Availability with Active/Active Networks at the upcoming NonStop Symposium and Expo. The Symposium will be held in San Jose, California, from September 26th to September 29th. The seminar will be an all-day affair on September 26th and will cover availability theory, active/active architectures, data-replication techniques and products, comparisons to other high-availability technologies, cost/benefit analyses, and case studies. Included also will be a sprinkling of Bill’s educational and intriguing Never Again horror stories.
Bill also will be discussing fast-failover techniques for continuous availability. How can you achieve subsecond recovery times unless you can reroute traffic from a failed node to a surviving node in subseconds? Learn several techniques for doing this in the in-depth breakout session Active/Active Fast Failover.
You can register for the NonStop Symposium at
Dr. Bill’s preconference seminar is an example of his one- to five-day customized seminars that he presents onsite to companies such as yours. Check out our prototype syllabi at
It seems that no matter how hard we try, assured availability of our IT services continues to be evasive. In this article, we review a sampling of the many outages that have struck major companies in the past six months. Networking continues to be a top issue, causing over a third of all outages. Likewise, during this period, power and cooling failures accounted for another third. Interestingly, half of the power outages were caused by automatic transfer switches that did not cut over the data center to backup power. The rest of the outages were a mix of hardware faults, software bugs, and a denial-of-service attack.
A disturbing statistic is that half of these outages were experienced by large hosting, cloud, and SaaS providers such as Amazon, Hostway, Salesforce, Rackspace, and The Planet. This observation emphasizes the fact that shared services are not yet suitable for critical applications unless you have a good, tested failover plan.
Not everyone runs active/active systems for continuous availability. Many of you don’t even run a backup system. Downtime is simply a cost of business with which to be dealt.
But downtime is still costly. Even if you don’t run redundant systems and are willing to accept some amount of downtime, how can you reduce it through tuning up your operating procedures and purchasing decisions? At last month’s HP Technology Forum, held in Las Vegas, Keith Parris, a Systems/Software Engineer at HP, presented an excellent set of insights and tips for maximizing your availability no matter what kind of system you run. In this article, we review his observations and suggestions.
Many of the availability tips suggested by Keith may be obvious to you, but I am sure that there are nuggets of good ideas buried in this discussion. Improving availability is not just a matter of hardware and software choice and configuration. It is a process that must be planned and followed throughout the deployment of your application.
After cosponsoring a successful HP Technology Forum in Las Vegas last month (an event that attracted over 5,000 attendees), Connect will be hosting U.S.-based multiday events dedicated to the NonStop and OpenVMS user communities. The NonStop Symposium will take place this September from the 26th to the 29th in San Jose, California. The OpenVMS Bootcamp will be held in Nashua, New Hampshire, from September 13th to the 17th. Connect is HP’s largest and most engaged IT professional user community.
These NonStop- and OpenVMS-dedicated events disappeared for a year. But they’re back, so take advantage. By registering for one or both events, you’ll get the inside track about HP roadmaps, dive deep into the detail you can only get from HP technologists, and have the opportunity to network with fellow users.
In our December, 2009, issue of the Availability Digest, we reviewed Megaplex: An Odyssey of Innovation, a history prepared by The Standish Group. The history covers 35 years of unique technical innovations brought to us by Tandem computers, now HP NonStop servers. Standish defined the Megaplex as “a fabric of resources that will provide for application services for the next 35 years.” It goes on to predict that ”in the future, NonStop technology will be the basis for the Megaplex.”
The Standish Group has now published a roadmap of how this is going to happen. In its white paper, Roadmap to the Megaplex, Standish further defines the Megaplex as “a collection of server blades acting together as a single system using multiple types of operating systems, databases, and other computer resources.”
The Megaplex roadmap is a guide to modernizing NonStop applications and to making them a core element of the enterprise. By decomposing applications into services, each service can be run in the environment that matches its availability, data integrity, and security needs. The less critical the service, the less costly is its environment. Modernization provides the balance between cost and performance for all the applications in the enterprise portfolio.
Sign up for your free subscription at http://www.availabilitydigest.com/signups.htm
Would You Like to Sign Up for the Free Digest by Fax?
Simply print out the following form, fill it in, and fax it to:
+1 908 459 5543
The Availability Digest may be distributed freely. Please pass it on to an associate.
Managing Editor - Dr. Bill Highleyman firstname.lastname@example.org.
© 2010 Sombers Associates, Inc., and W. H. Highleyman