Service-level agreements have always been challenging to meet. Often, the SLAs imposed on an IT department are...
unrealistic. Even if an SLA is not overly ambitious, there are three major challenges that can stand in the way of meeting operational availability goals.
1. Unforeseen events. A natural disaster such as a flood or a fire, for example, could potentially bring down mission-critical systems. Most IT shops plan for these types of situations in hopes of keeping critical workloads online throughout the duration of the disaster. Even so, it doesn't always take something as dramatic as a fire or flood to impact operational availability.
Suppose an organization suffers a security breach or a widespread malware infection. While these types of events might not always force mission-critical systems offline directly and interrupt operational availability, such systems may need to be manually taken offline to prevent further damage or to perform a forensic analysis.
2. System maintenance. Windows-based failover clusters support cluster-aware updates that allow one node to be patched at a time, while the other nodes -- and your workloads -- remain online. As such, it may be tempting to establish an SLA that doesn't leave room for maintenance. Keep in mind, however, that maintenance does not always center on servers or cluster nodes. You might, for example, need to momentarily take a system offline to expand its storage or replace an aging switch.
3. Conditions that give the perception a problem has occurred even when operational availability remains. Suppose your organization's Internet service provider loses connectivity. Although the workloads in your data center remain functional, those attempting to access these resources from the Internet will receive an error message and may assume that the servers hosting the resources do not have system availability.
Case study: System availability key at high-traffic times
VM clusters and high availability help protect data
SLA template for disaster recovery
Dig Deeper on Disaster recovery facilities - operations
Related Q&A from Brien Posey
A poorly timed Windows 10 update can cause a major headache for an end user. Fortunately, users can pause updates or IT can take steps to halt ... Continue Reading
With tape storage still playing a role for such use cases as offline protection from ransomware, what are some key improvements in the media ... Continue Reading
If you don't need to access old data much, cloud archiving could work out well. That's just one consideration, though, for deciding which archival ... Continue Reading
Have a question for an expert?
Please add a title for your question
Get answers from a TechTarget expert on whatever's puzzling you.