Guest post by Inge De Maere, Storage Resource Management Functional Architect at HP Software
Performance and capacity management are becoming inseparable due to the dynamic nature of converged virtual infrastructures. Server virtualization and storage technologies can accelerate performance bottleneck problems and pose capacity planning concerns, due to their nature of overcommitting resources.
So the question is: Are storage administrators more likely to run out of IO performance or disk storage capacity? They need to carefully plan the number of virtual machines they can support per storage array controller port. They need to plan the disk's read/write performance, because virtual machines (VM) may make different types of requests depending on the applications they are running. One or two especially busy VMs on any given physical server can consume most of the CPU, memory, network bandwidth and disk I/O. If VMs are running especially I/O-intensive applications that are hammering on a disk, other VMs that use the same data store may suffer the effects of disk I/O contention. This problem is more difficult to identify than an overloaded system CPU or memory resources. Optimizing capacity and performance needs doesn’t mean simply allocating free storage space where it is available. Performance impact needs to be taken into account as well.
HP Virtualization Performance Viewer provides performance metrics for virtual systems and business processes. The customizable performance dashboard provides a view on the datastore, hosts and guest VMs, and underlying storage systems through metrics, graphs, and reports. A dashboard Treemap allows you to drill into datacenter, host and VM guest, datastore and storage views, to view trends and forecasts. A customizable performance workbench provides detailed metrics on datacenter, datastore, host and guest resource usage. A forecasting view provides graphs on CPU, memory and disk usage. Performance alerts highlight VM, datastore, host and storage performance bottlenecks, and provide recommendations to resolve root cause.
HP Storage Essentials Capacity Manager gives you the end-to-end view of your VMwareservers and your supporting storage environment. It provides current, historical, and projected capacity information for all supported business applications, guest OS, VMware ESX datastore, fabric switches, and storage systems. For each resource type, capacity information is presented in the appropriate context for that resource. For example, HP Storage Essentials provides:
Available, utilized, and free capacity (and percent calculations) for guest OS volumes and VMware VMFS volumes
Used and available ports for a fabric switch
Mapped, unmapped, unallocated, and unused raw capacity for disk pools in a storage system
Leveraging this knowledge, server and storage administrators can address capacity and performance planning questions for their environment:
How many more VM guests can be placed on existing datastores?
How much time do I have before I run out of IO?
How much time do I have before I run out of storage capacity?
What’s the total capacity consumed by all VMware farms?
For in-depth historical and real-time performance monitoring, HP Virtualization Performance Viewer provides detailed information about the guest VMs. It offers also a customizable performance dashboard, called Performance Workbench. For historic capacity monitoring, HP Storage Essentials Capacity Manager offers detailed views into the host and storage arrays capacity usage. For an overview on current capacity usage for both ESX hosts and storage arrays, the Storage Essentials Capacity Dashboard can be used. Both products offer the options to forecast into the future, based on historic data collection.
Storage Essentials is a storage resource management solution that helps you reduce total costs of storage operations and increase productivity.