Cloud Performance
Enterprises deploying either public or private clouds face unique cloud performance and capacity management challenges. These include the need for real-time resolutions to performance issues, unpredictable VM demand from self-service deployments and customer reporting requirements.
Capacity Management for the Cloud
Related Links
Private Cloud Capacity Management Resources
Virtualization Team: Capacity Management in a Virtualized Private Cloud
Peer 1 Hosting: Improve Your Visibility into the Cloud with VKernel
vOPS Reporting & Chargeback wins 2011 Gold Award for Cloud Computing Technology
Performance and Capacity Management for Cloud Environments
Performance and capacity management for cloud environments is unique for the following five reasons:
- Performance monitoring across the virtualized infrastructure is critical - VM performance issues need to be diagnosed and resolved in real time to maintain SLAs and end user satisfaction. Network and storage areas can be the cause of VM performance problems and must be monitored as well.
- Capacity planning accuracy drives return on assets - If assets are over-procured, a cloud operator's return on investments is lowered. Under-procure, and a cloud operator risks VM performance problems, lost sales, and a internal user migration to external cloud providers stemming from capacity shortages. VM demand from customers is unpredictable and cloud environments must be prepared for multiple scenarios.
- Reporting for customers and other stakeholders is needed - When tenants "pay" for resources, these users want to see usage, performance, and summary statistics for their accounts to make additional VM allocation decisions. Visibility into the environment must be secure, and reports must be distributed to users through the systems they use.
- Wastefinding improves storage efficiency - As self-service cloud environments and difficult to track, waste files such as abandoned VM images, long-unused snapshots and powered-off VMs from terminated accounts can accumulate. Removing this waste reclaims hardware that adds to the return for a cloud initiative.
- Chargeback is required for billing - Without chargeback, there is no way to meter for resource use and bill customers or limit use of perceived "free" virtual machines for internal private cloud.
- Data integration is necessary to streamline operations and enable workflow automation - Without back end access to capacity management data, operations involving various external systems such as self-service portals require manual data entry or fact checking.
To learn more, download the free VKernel whitepaper, Capacity Management for Infrastructure Clouds.
VKernel Cloud Solutions
VKernel is the leading provider of performance and capacity management solutions for public and private clouds. Specifically, VKernel's:
- Performance Monitoring - vOPSTM Performance Analyzer analyzes, diagnoses and resolves VM performance problems in real-time. Performance issues across multiple data centers and hypervisors can be easily visualized on one screen. Automated issue remediation is available in some cases. Alerting and visibility reaches into storage and network infrastructure.
- Capacity Planning - vOPSTM Capacity Manager allows virtualization administrators to model multiple VM deployment scenarios to assess the impact of VM additions to available capacity. Capacity status across multiple data centers and hypervisors can be easily visualized on one screen. Capacity for known VM deployments can be reserved and removed from further modeling consideration and resource requirements can be predicted to any future date.
- Reporting - vOPSTM Reporting & Chargeback provides advanced customer reporting to give secured visibility into resource usage and VM summary statistics. Reports can be embedded into web portals or other systems, or shared via SNMP or email.
- Wastefinding - vOPSTM Optimizer can identify Abandoned VM images, unused snapshots, templates, powered-off VMs and "Zombie" VMs.
- Chargeback - vOPSTM Reporting & Chargeback allows for rapid set-up of chargeback reports that profile VM utilization and can track back costs for resource allocations and actual usage based on CPU, memory, storage, and network tiering. Additionally, cost optimization opportunities across multiple data centers and hypervisors can be easily identified on a single screen.
- Data Integration - The vOPSTM API allows for programmatic access of all VKernel data to external systems. Additionally, management actions within VKernel's vOperations Suite can be tied to triggers from external systems.
Next Steps
- Learn about award-winning products that enable VKernel's performance and capacity management solutions for clouds in the VKernel vOperations Suite
- Learn about the vOPS API to enable integration of capacity management data into external systems
