"You can fool some of the people all of the time, and all of the people some of the time, but you cannot fool all of the people all of the time."
Abraham Lincoln
Some of the outage figures quoted by organizations look ludicrously small to me. Without casting aspersions on the veracity of these figures (or availability statistics), I do feel that some examination of them is needed.
The non-availability of a system is often quoted thus:
where MTTR is the Mean Time To Repair a particular outage.
”We recognized we'd run the wrong job and restarted correctly in just 3 minutes, thus our MTTR = 3 minutes."
If you substitute the word "recover" for "repair" in the above definition you will be closer to the truth. However, your database(s) are almost certainly on Planet Zog as far as consistency is concerned and the "repair" of that will often take much longer. The correct definition of MTTR should be "mean time to recover" and the equation then looks as above but with a new MTTR:
The last item in this equation I call the ramp up time, the time to get the show back on the road. This can be small but is often much larger than repair time, as shown in the diagram below. A decent Service Level Agreement (SLA) will opt for this definition of "fixed" for an issue and will include the ramp up time in the recovery time specification.
The recovery of a database or other data and metadata corrupted by human error or malware can take a considerable time to restore to the working status demanded by the end users.
This is borne out by several “Never Again” cases outlined in the Availability Digest (under the heading: Never Again) where financial bodies — banks, stock dealings — have repaired faults but taken many hours to recover normal working conditions again. ”The system was repaired at 11am and trading commenced normally at 2:30pm" is a typical (hypothetical) report on such situations.
The final point to make is that there are several viewpoints of an “outage” or period of "downtime", depending on your place in an organization. The end user's view will be that the outage lasts as long as he/she is prevented from using IT to do the job they are supposed to do. The server specialist's view might be that the outage of his hardware was a mere minute or two before it was fixed whereas the network person will say" “what's all the fuss about; everything on the network is working fine?”
It all depends on your viewpoint and I know what viewpoint the company CEO, the users and the board will take. Do you?
Dr. Terry Critchley is the Author of “High Availability IT Services” ISBN 9781482255904 (CRC Press).
The Latest
In MEAN TIME TO INSIGHT Episode 5, Shamus McGillicuddy, VP of Research, Network Infrastructure and Operations, at EMA discusses the network source of truth ...
A vast majority (89%) of organizations have rapidly expanded their technology in the past few years and three quarters (76%) say it's brought with it increased "chaos" that they have to manage, according to Situation Report 2024: Managing Technology Chaos from Software AG ...
In 2024 the number one challenge facing IT teams is a lack of skilled workers, and many are turning to automation as an answer, according to IT Trends: 2024 Industry Report ...
Organizations are continuing to embrace multicloud environments and cloud-native architectures to enable rapid transformation and deliver secure innovation. However, despite the speed, scale, and agility enabled by these modern cloud ecosystems, organizations are struggling to manage the explosion of data they create, according to The state of observability 2024: Overcoming complexity through AI-driven analytics and automation strategies, a report from Dynatrace ...
Organizations recognize the value of observability, but only 10% of them are actually practicing full observability of their applications and infrastructure. This is among the key findings from the recently completed Logz.io 2024 Observability Pulse Survey and Report ...
Businesses must adopt a comprehensive Internet Performance Monitoring (IPM) strategy, says Enterprise Management Associates (EMA), a leading IT analyst research firm. This strategy is crucial to bridge the significant observability gap within today's complex IT infrastructures. The recommendation is particularly timely, given that 99% of enterprises are expanding their use of the Internet as a primary connectivity conduit while facing challenges due to the inefficiency of multiple, disjointed monitoring tools, according to Modern Enterprises Must Boost Observability with Internet Performance Monitoring, a new report from EMA and Catchpoint ...
Choosing the right approach is critical with cloud monitoring in hybrid environments. Otherwise, you may drive up costs with features you don’t need and risk diminishing the visibility of your on-premises IT ...
Consumers ranked the marketing strategies and missteps that most significantly impact brand trust, which 73% say is their biggest motivator to share first-party data, according to The Rules of the Marketing Game, a 2023 report from Pantheon ...
Digital experience monitoring is the practice of monitoring and analyzing the complete digital user journey of your applications, websites, APIs, and other digital services. It involves tracking the performance of your web application from the perspective of the end user, providing detailed insights on user experience, app performance, and customer satisfaction ...
