EDM provides the ability to create views for displaying lists of objects in the UI. By creating views you can select the columns that will be displayed, and also specify filtering characteristics (conditions) for the items to be displayed or excluded from the display. EDM 3.3 provides one more configuration option: ACLs. On the new User tab of Edit View dialog box, you can select which users will be able to use that view, effectively limiting the users that can see Hosts, Connections, or any other screen in the EDM UI.
Alerting is one of the core capabilities of EDM. Alerting provides a collection of thresholds that are evaluated by EDM for each incoming data block, and defines the actions to be taken (email notifications, run external application, etc) for each of them in case a metric’s value rises above or falls below of the value defined in the threshold definition.
Alerts are groups into Alert Groups, and every Alert Group contains a set of configured alerts that can be assigned to one or more connections. How do you know whether an Alert Group has been assigned to a connection? In EDM 3.3, you can display this information by selecting the server icon in the Alert Groups display. This seems like a simple thing, but it can be a real time-saver when you need to double-check the your alert settings!
While working with one of our customers and analyzing some of their performance data we came across an interesting situation… one that administrators should be aware of and plan for … over-provisioning! If you’re not familiar with the term, over-provisioning is allocating excess resources to a group of guests systems within a Hypervisor. This could be AIX LPARS, Solaris Zones, or VMware guests.
Say I have a 4-cpu system. I could deploy three instances of Linux on that system, providing each with two cpus. At first this is AWESOME! I just made three 2-cpu servers out of four cpus! This is the value of virtualization — it allows us to better leverage finite resources and reduce costs. But remember, virtualization has a price. What if all three servers want resources at the same time? Performance suffers. This is the cost of over-provisioning.
With Sightline, we help identify and right-size the provisioning rules to give you a balance of physical resources and virtual performance. By analyzing the performance profile of both the physical and virtual workloads, administrators can keep mission critical workloads running efficiently and cost effectively!
Performance evaluation is the process of determining how well an existing or future computer system meets a set of alternative performance objectives. Arbitrarily selecting performance metrics, evaluation techniques, and workloads often leads to inaccurate conclusions.
How should one carry out a performance evaluation study? The answer is to follow a systematic approach. The methodology proposed here involves six steps:
The purpose of this article is to provide the performance analyst with a systematic approach to performance evaluation and point out some common mistakes that can be avoided.
Read the white paper
Questions? Comments? Suggestions? Ask John! If we use your input in a future newsletter, we’ll send you a $10 Amazon gift card.