|Read the Digest in
You need the free
The digest of current topics on Continuous Availability. More than Business Continuity Planning.
BCP tells you how to recover from the effects of downtime.
CA tells you how to avoid the effects of downtime.
Thanks to This Month's Availability Digest Sponsor
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.
Check out our seminars.
Check out our technical writing services.
Our Keynote Address on Spectacular Data Center Failures
I was pleased to have been invited to give the keynote presentation at BITUG, the British Isles HP NonStop User Group, this past December. My talk, entitled “Help! My Data Center is Down,” described spectacular outages experienced by data centers around the world, often for reasons we could never imagine – a battery-room explosion, alarm sirens destroying disks, an old lady digging up an Internet backbone cable.
I reviewed with the audience of over one hundred the lessons that can be learned from such failures. One is that humans are the least reliable of all data-center components. Another is that no matter how much protection you provide, some unforeseen event may take down your data center, leaving you to rely on your Business Continuity Plan (do you have one?).
These stories were taken from the annals of our Never Again series of articles, which are the most read in the Availability Digest. They also form the background for much of the content in our popular availability seminars.
If you are involved in an organization that would like to sponsor this educational and entertaining presentation, let us know. We’d be glad to share these experiences with your members.
Dr. Bill Highleyman, Managing Editor
Black Friday and Cyber Monday recently brought grief to many retail web sites. Several retailers lost online sales due to outages for times extending from a half hour to several hours.
But as serious as these outages were to brand loyalty and to revenues, they may be hiding an even more insidious problem – website performance. No one talks about this, but as performance expert Lenny Rachitsky says:
“Downtime is better for a B2C [business-to-consumer] web service than slowness. Slowness makes you hate using the service; downtime you just try again later.”
Several studies have shown that a two-second page load time is optimal. However, many major retail web sites experience load times exceeding twenty seconds during major shopping days. Insights into the impact on revenues and customer loyalty of poor website performance raise some critical questions. Do retailers understand the cost of poor performance? Do they have a stated performance goal for their web sites? Do they have a clear sense of how to go about making their sites faster?
It seems that there may be hidden gold that is worth mining in a poorly performing web site.
Since 1989, the Disaster Recovery Journal (DRJ) has sponsored the semiannual Spring World and Fall World conferences dedicated to Business Continuity and Disaster Recovery (BC/DR). Its 46th conference, Spring World 2012, will be held the week of March 25th at Disney’s Coronado Springs Resort in Orlando, Florida. The body of the four-day agenda (Sunday, March 25th, to Wednesday, March 28th) includes nine unopposed general sessions, 24 breakout sessions, twelve workshops, and a new invitation-only advanced practitioners track.
Several pre-conference and post-conference courses also are scheduled for an additional fee. Pre-conference courses are held all day on Saturday, March 24th, and Sunday morning, March 25th. Post-conference courses are one- to three-day courses held on Wednesday afternoon, all day Thursday, and Friday morning, March 28th to March 30th. Courses and examinations for certain BCP (Business Continuity Planning) certifications are offered following the conference.
DRJ’s Spring World 2012 conference continues twenty-three years of distinction in the fields of Business Continuity and Disaster Recovery. Lasting for over a week with informational sessions, workshops, certification exam preparation, and qualifying exams, it is the premier educational event for BC/DR professionals.
Today’s data centers are incredibly complex. No matter how much redundancy data-center designers build into their infrastructures, things fail. Data centers fail, often with disastrous consequences. Sometimes, a fault will take down a data center for hours. Sometimes, a fault will take down a data center for days.
In our previous articles, we related several major data-center disasters caused by power failures, storage subsystem faults, and in our last article, Internet outages.
A data center is no good to anyone if it cannot be accessed by its users. The Internet outages we described in our last article were faults external to the data centers and that seriously impacted their operations. But data centers also rely heavily on internal networks to interconnect their servers, to connect internal users, and to provide connectivity to the external Internet. These internal networks are called Intranets.
In this article, we explore some notable Intranet failures that rendered data centers useless even though they were otherwise fully operational.
MySQL is one of the most popular relational databases in use today. Schooner Information Technology, Inc., has made significant extensions to MySQL to improve its availability and replication performance. The goal of the resulting product, SchoonerSQLTM, is to achieve five 9s availability, or about five minutes of downtime per year on the average.
SchoonerSQL provides significant availability and performance advantages over standard MySQL implementations. Certified by Oracle as being fully compatible with MySQL Enterprise and InnoDB, it can be used without modification by any MySQL Enterprise application.
The product’s performance improvement is achieved by highly parallelized, multithreaded replication threads for use with multicore processors. Benchmarks have shown a performance improvement over MySQL configurations of two to five times.
SchoonerSQL’s high availability is achieved through the use of synchronously replicated multinode clusters that permit fast failover (within seconds) of a node failure. A SchoonerSQL cluster is scalable by configuring it with up to eight nodes. Additional read and write scalability is provided by adding more clusters synchronized via asynchronous replication. A multicluster environment can be distributed over unlimited geographical distances to provide full disaster tolerance.
Sign up for your free subscription at https://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 is published monthly. It may be distributed freely. Please pass it on to an associate.
Managing Editor - Dr. Bill Highleyman email@example.com.
© 2011 Sombers Associates, Inc., and W. H. Highleyman