As we know, virtual environments consist of many moving pieces and are generally complex to setup. Typically, IT environments, depending on the size of the organization, can have several hundred VMs down to a handful of VMs. For such virtual infrastructure deployments, it helps to monitor the performance of VM and VM usage. It's also equally important to ensure the health of your virtual appliances are always in check and to immediately know when something goes wrong.
What you really don't want is to have alerts paging you 24/7, especially when they're not critical situations. Alert management can be a subtle, but dangerous activity. Additionally, manually setting alert thresholds can be an extremely time consuming task. Alternatively, using static thresholds that don't reflect real performance problems often result in alert storms, where administrators stop watching alerts carefully. This is where the "dangerous" part comes in and often true critical alerts can be lost in the noise and missed. As a result, intelligent, dynamic alerting can be critical for both staff efficiency and system reliability.
False Alerts: Reasons Why You Get Them and How to Avoid Them
Here are a few examples why your virtual environment may trigger alerts more frequently than normal:
■ Events that frequently occur, such as resource consumption can trigger alerts more often than most other virtual components.
■ You can get "spam" alerts from VMs or hosts that are no longer in use or that have been discharged.
■ Not properly tuning threshold levels can lead to a sudden spike in alerts.
Having intelligent alerting processes help ensure irrelevant alerts are not generated. This gives virtual admins time to look at "real" alerts and fix them. Here's what you can do to avoid alerting errors:
■ Set up alerts for specific VMs that you think are really going to impact your users or your business.
■ Leverage dynamic thresholds based on historical baseline trends whenever possible to set more realistic thresholds for your clusters, hosts, VMs, and datastore.
■ Establish well-defined threshold settings—this way you can optimize the kind of alerts you receive during the day and ensure that you're not bothered after work hours.
■ Set the right dependencies to significantly lower the amount of alerts you receive.
■ Forward specific alerts to the defined teams, since they understand the severity of the alert and can fix it right away.
Determine What to Monitor and Why
Most admins have to monitor hundreds of virtual appliances, which means you're probably dealing with plenty of alerts. Under these circumstances you'll have to determine a few things:
■ Go over each host to see if all VMs under the host must be monitored or if only a few critical VMs need to be monitored for alerts.
■ Talk to your business groups or users and understand what the impact will be. This will give you a sense of how many VMs and datastores have to be setup for alerts. They may have mission critical applications running inside them, which may affect business performance.
Statistical Thresholds: A Better Way to Set Baseline Values for your Virtual Environment
Normally, you would have to monitor the performance of hosts, VMs, and datastores for several weeks in order to know what the ideal or optimum baseline is to set warning and critical thresholds. However, integrated virtualization management tools can automatically calculate performance of clusters, hosts, VMs, and datastores and determine the baseline values.
IStatistical thresholds allow you to look at the following processes:
■ Applying thresholds to clusters, hosts, VMs, and datastores.
■ Understanding baseline statistics using standard deviation calculation for day and night system performance.
■ Gaining statistical insights into performance metrics and how they vary over time. Look at how stats are collected for higher and lower threshold values for individual VMs and hosts.
■ Calculating thresholds from historical performance data saves time in adjusting thresholds and provides more intelligent alerts.
■ Setting the right threshold values using the built-in baseline calculator. This calculates and applies the recommended threshold values for warning and critical stages for clusters, hosts, VMs, and datastores.
While this won't completely eliminate "spam" alerts, it will quickly let you get to a much smaller set for the administrator to deal with. In turn, it will let them spend more time and attention on striking that balance between monitoring your VM usage and hypervisor performance, and setting the right threshold values.
Karthik Ramachandran is Product Marketing Specialist at SolarWinds.
The enterprise WAN is unable to keep up with digital transformation demands, according to Foundation for Digital Transformation, a new research report, authored by Ensemble IQ and supported by InfoVista. This challenge was universal across all three vertical industries surveyed — retail, manufacturing, and banking/financial services ...
Achieving optimum Java Virtual Machine (JVM) performance is key to ensuring proper memory management and fast application processing. According to a Cornell University study, a 1-millisecond improvement in the performance of a trading application can be worth $100 million a year to a major brokerage firm. Because of this potential for loss, IT teams owning banking, financial, trading and other Java-based applications place a high value on having a proper JVM monitoring strategy in place ...
APM had to evolve to keep pace with development velocity and maintain the service quality for the modern applications born out of digital transformation. Automation and artificial intelligence (AI) technologies are critical to the next step in APM evolution, helping to address speed, scalability and intelligence demands ...
A worldwide survey by Gartner, Inc. showed that 91 percent of organizations have not yet reached a "transformational" level of maturity in data and analytics, despite this area being a number one investment priority for CIOs in recent years ...
Mobile app performance is still a significant issue. In a new report from PacketZoom, The Effect of Mobile Network Performance on Mobile App Users, 66% of consumers said reliable mobile app performance is "very important" — second only to mobile app security ...
IT departments that shift from reactionary fire fighters to becoming proactive business partners find their ticket counts reduced from 20 to 50 percent or more. The strategies outlined in Part 1 of this blog may all sound like a great way to turn IT into a strategic, proactive business-enabler, but how can companies turn strategy into reality? The following are three best practices ...
"We can't fix it if they don't call." I can't count how many times I've said those words in my IT career. We need users to call in their issues, while conversely we need our ticket volumes to decrease. How can IT lower the amount of call center tickets, quickly resolve those incidents that can't be avoided, and reduce their own costs in the process? Here are three key strategies ...
Today's network engineers have their work cut out for them. Bigger, more complex networks have created an environment where network engineers are forced to adapt and develop more effective ways to manage and troubleshoot their networks. This begins with better visibility, which has presented an issue traditionally as engineers struggle to create an accurate picture due to challenges with static maps ...
My last blog covered technology-oriented best practices that application management and IT help desks can use to optimize the performance of their applications and the IT teams that oversee them. Now I'll explore what IT professionals can do to optimize their team's time and resources — the people and processes — in pursuit of that same goal ...