By Lisa Crewe, Senior Product Marketing Manager, NetApp
What if you had a one stop shop for quickly assessing and making decisions about the storage in your virtual infrastructure or cloud environment. Now you do! In the last post, titled It’s the Elasticity, Stupid, we talked about the value of a Computing Elasticity Dashboard for your Hybrid Cloud environment. In this post, we’ll cover a dashboard designed for Storage Managers and Business Analysts that helps you better plan capacity and quickly identify cost saving opportunities. This dashboard is new and out of the box in Insight v7.0.1, which is now available from the NetApp support site for existing Insight customers.
The objective of this dashboard was to provide a single executive level view of current and future issues in the storage infrastructure, whether the storage is supporting a virtual infrastructure, private cloud or hybrid cloud environment. The information can we viewed on a single page via your web browser, on a laptop, desktop, or mobile device so you have access wherever you are.
The information is refreshed daily and stored within the data warehouse. The collected data is also augmented with information that you can configure to give context and allow comparisons. The dashboard is broken into seven widgets each containing contextual information on certain aspects of the storage environment. Let’s take a look at each widget and what it tells you.
The top “Summary” widget details the capacity of storage, the switch ports and how they trend over time for the last 12 months. It also tells you the current state of capacity, predicted capacity over time and gives context by allowing the entry of a target. This allows you to track usage and compare and contrast what the system says to where you want to be.
For example, you could configure the system to show the target and forecast for the end of the financial year so you can track how storage capacity is growing compared with the forecast. This enables you to adjust you trajectories and budget allocation.
The other information displayed in this first pane is the Fibre Channel (FC) Bandwidth Utilization. There is no target for this data set however there are acceptable ranges. The graph displays the actual utilization compared against these ranges. There is a range for low utilization, mid utilization and high utilization.
Storage Tiers Capacity
The next section displays Storage Tiers Capacity. The concept is the same as described for the prior Summary widget. The widget is configurable to display however many tiers is required. The capacity usage is shown for the last 12 months. It displays the value alongside a graph that represents the actual usage, the usage forecast by OnCommand Insight, and a target.
There is then a histogram showing the total capacity. This allows you to understand how much of a share of the environment is distributed amongst the tiers.
There is also a histogram that shows the number of months until that particular tier is full. This works similar to a fuel gauge - the less capacity, the smaller the bar displayed.
Daily Storage Traffic
The bottom, left section is a Daily Storage Traffic chart. It displays three lines that allow you to compare and contract yesterday’s traffic (the black line), to the last 7 days daily mean (dark grey line), and then to the daily mean for the last 6 months (light greay line). This enables you to visualize any abnormalities in the way the infrastructure has been performing, as it gives information, which would highlight both cyclical (last 7 days) and seasonal variations (last 6 months).
Data Centers Time to Full
The next section, Data Centers Time to Full, gives you an easy to read view into how much capacity is left in each data center for each tier of storage. The tiers could also be configured to track capacity in a cloud environment. The darker the blue color, the less space there is. The lighter the color, the less concern there should be. The darkest blue is less than 3 months, lighter blue is 3-6 months and the very pale blue is greater than 6 months.
Top 10 Applications
The Top 10 Consumers (Physical Hosts only), shown in the Top 10 Applications sections, lists the consumers in order of greatest capacity and displays how much capacity they have used over the last 12 months including showing the current amount, and then a histogram of the share of the infrastructure.
The Response Time graph shows the previous day’s minimum response time, the first quartile, the third quartile, the maximum response time and the white line being the median. With this information, you get a visualization of the range of user experience for the previous day. There is also a grey block that the box whisker is overlaid on that represents an acceptable service level range. This allows you to see if the consumers will be experiencing acceptable or, more importantly, unacceptable response times. If the box whisker graph completely resides to the far left of the acceptable range this could suggest the consumer is on storage that is delivering more performance than required and there could be an opportunity to move the consumer to a different type of storage.
Storage Tiers Daily Performance
The bottom widget, Storage Tiers Daily Performance, shows a summary of each tier’s previous days performance in the box whisker format for response time and throughput. These box whiskers are overlaid on top of the acceptable service range for that tier. This allows the you to see if there is an opportunity to consolidate tiers, realign workloads delivered from those tiers, and highlight issues with particular tiers.
Orphaned Capacity is the final section. There are two visualizations in this widget. The first is the horizontal stacked bar. The grey colors here are predefined and show the acceptable 25 and 75 percentile ranges. The darker the grey means the more unacceptable the situation is. The actual situation is shown with the narrow bronze bar that shows the actual capacity that is orphaned. The second visualization is the orphaned capacity detailed by tier of storage. This allows you to understand which categories to investigate to find savings.
An important consideration in this widget is the actual definition of orphaned storage. The first category is storage orphaned by configuration, this means there is storage allocated to a host but the configuration has not been done properly and the host cannot access the storage. For example, mapped but not masked, masked but not mapped, or mapped and masked but not zoned properly. The second category is orphaned by performance. This is storage that is correctly configured to, but has zero “0” storage traffic over the entire time the asset has been monitored.
This dashboard requires the OnCommand Insight Discover, Perform and Plan Modules. Discover enables the collection of all of the inventory. Perform is the performance engine. And Plan handles the capacity and reporting. If you’re an Insight customer, we’d love for you to download the new version and check it out. If you’re looking for a solution like this, ask me for a demonstration or better yet – visit us at VMworld at the NetApp booth #1205.
Hope to see you there!