IBM Tivoli Monitoring for Virtual Environments

• Reduce hardware and related energy and software license costs by improving utilization and deferring purchases, while providing workload placement guidance.

• Reduce labor costs spent on manual capacity analysis and planning through interactive what-if capacity analysis tools

• Accurately alert on performance problems with dynamic thresholding, which alerts on abnormal behavior rather than static thresholds

• Predict days, weeks and months into the future when resources will reach capacity limitations

• Reduce mean time to resolution with integrated monitoring across physical and virtual servers, network and storage resources, including configuration change histories

• Low TCO with one tool for robust functionality, and platform coverage to include multiple hypervisor support; extensible to OS and application environments.

Businesses today are challenged to improve IT service delivery and reduce the costs of delivering those services. Virtualization is a key initiative businesses deploy to reduce infrastructure costs by improving server utilization and to improve services by having the ability to rapidly scale for increases in application usage.

While virtualization can produce significant cost savings as a result of reducing infrastructure overhead, it does not address the single-largest cost element for most data centers—the labor to manage this environment—which can be as high as 40 percent of the overall cost. If not controlled, management costs can negate the cost savings realized through virtualization.

Specific challenges around managing a virtual environment include:

• Capacity management and planning, controlling virtual server sprawl, proactively identifying when critical resources will reach capacity, and identifying opportunities for further consolidation of resources.

• Proactive identification and isolation of performance problems across server and storage environments.

• Accurate identification of performance problems in a dynamic environment; detecting changes from normal performance.

• Understanding the relationships between physical and virtual resources to help isolate performance problems to the correct resource.

• Going beyond simple physical capacity calculations, to position workloads in an intelligent way and avoid mistakes in co-location that could diminish service levels in pursuit of deeper consolidation.

IBM® Tivoli® Monitoring for Virtual Environments (formerly named IBM Tivoli Monitoring for Virtual Servers) is designed to address the specific challenges described above. Capacity planning and reporting are vital to optimizing the virtual environment; i.e. improving density and workload placement decisions. Predictive alerts identify resources that are trending negatively, and threatening to run out of capacity, days and weeks in advance. In addition, availability monitoring can be used to help find problems with critical resources and alert operations teams before users are adversely affected.

For customers with mature virtualization practices, advancing their posture toward full-featured private cloud deployments, ITM for Virtual Environments provides integration with complementary Tivoli tools that will help them view the infrastructure from a service delivery perspective, adding a dimension of capacity planning and analytics, as well as health monitoring, to cloud consumers. By integrating with tools like IBM Director, Tivoli Storage Productivity Center and Tivoli Application Dependency & Discovery Manager (TADDM), customers can have even deeper insight into the physical components that are part of their virtual/cloud infrastructure, helping drill into the lower levels of the environment. By integrating with products like Tivoli Business Service Manager, customers can ensure that their virtualized resources can be viewed in a business application context. Whether a customer is just starting out with virtualization, or is managing a mature virtualized infrastructure, ITM for Virtual Environments provides the tools to ensure the health, performance and capacity of that environment, while helping maximize IT cost savings.

Specific benefits include:

• Reduction in hardware and related energy and software license costs by improving utilization and deferring purchases, while providing workload placement guidance

• Reduced labor costs spent on manual capacity analysis and planning through enhanced IBM Cognos-based, what-if analysis

• Improved application availability, using proactive problem identification features, such as predictive trending, and dynamic thresholding

• Reduced mean time to resolution by separating intermittent problems from recurring problems and by monitoring key resources across a broad set of platforms, in the same context; for example, you can visualize data IO performance side-by-side with supporting volume performance.

• Lower operating costs for management by using one tool to monitor multiple platforms.

To effectively manage capacity requirements, to ensure high availability and maximize resource utilization, capacity planning analysis and reports are necessary. Capacity planning analysis includes:

• Virtual machine right sizing – Virtual machines are initially provisioned with allocated CPU and memory. By understanding real usage of computing resources over time, customers can adjust the allocated computing resources to match the virtual machine’s actual requirements. In addition, customers can view virtual machines that are no longer active and de-provision these VMs to free up additional computing resources.

• Predicting physical and virtual resource capacity bottlenecks – Preventing performance problems related to capacity constraints includes understanding actual usage, allocated capacity and available capacity for CPU, Memory, Network and Storage resources for clusters, hosts, VMs, datastores, volumes, HBA and network adapters.

• Workload balancing – This activity includes comparing usage and available capacity across clusters and hosts to determine over or under-utilized clusters and hosts.

• Determining how many more customers or virtual machines can be serviced with existing resources (including a specified buffer factor) in a cluster or host by using an average or peak VM profile or by inputting the size of the virtual machine(s) that are needed.

• Determining if there is enough capacity, and what the delta gaps are, to add more virtual machines to a server or cluster by using an average or peak VM profile, or by inputting the size of the virtual machine(s) that are needed.

IBM Tivoli Monitoring for Virtual Servers helps organizations by collecting key performance metrics important for capacity planning such as CPU, Memory, Storage and Network utilization, warehousing this data and reporting on capacity trends using Cognos-based Tivoli Common Reporting. Tivoli Common Reporting provides out of the box what – if analysis reports for capacity planning in VMware environments but also provides drag and drop capabilities for ad-hoc reporting.

In addition to capacity planning reports, this offering provides recommendations on where to place workloads, based on business and technical policies, to balance workloads across clusters or to minimize server resource consumption. Key features include:

• Benchmarking data for analysis: SPEC benchmarking data is included for what – if analysis on various hardware platforms.

• Custom fields for analysis: Customers can associate virtual machines with other data sources such as application association data from a CMDB, customer or line of business data from a self service provisioning tool.

• Workload modeling: This feature assesses workload patters – stable, bursty – and provides reservation recommendations based off historical data and growth projections.

• Technical & business policies: The policy-based framework provides the ability for customers to apply custom or out-of-the-box policies to the analysis. Examples include growth rates, co-location/anti-location parameters, locating VMs with the same OS on a server, and providing a reservation cushion for critical workloads.

• Results simulation: Provides a report to answer – “If I place workloads in this way, what is the benefit in terms of numbers of servers saved?”

• Optimization recommendation report: Provides specific recommendations on reservations and where to place workloads in the environment to achieve the greatest benefit.

Dynamic thresholds:

Static thresholds are acceptable for monitoring certain performance metrics. For example, I always want to be alerted when excessive swapping occurs or % ready is critical. However, many performance statistics do not have static bounds. For example, it is important to be altered that a host exceeds a CPU utilization threshold of 90%, but it is equally important to understand when CPU utilization falls below the normal pattern. Tivoli monitoring provides the capability to monitor resources, establish a baseline and alert when resources deviate from normal behaviour (2 standard deviations for example).

Virtual environments by their very nature are dynamic and that is why it is important to have the ability to change monitoring thresholds dynamically. Dynamic thresholds also provide customers with the convenience of being alerted only when status changes from normal. As we all know, normal patterns can change throughout the day, depending on application demand. Dynamic thresholds offer the flexibility of maintaining thresholds that vary by the hour - each hour of the day can feature a different resource specific threshold if desired. In addition, the user can automatically change thresholds based on future patterns. For example, a host or cluster has a threshold today based on recent history of workload. If a virtual machine is added to a host, a host added to a cluster or if virtual machines move from host to host, the baseline that was established is meaningless and needs to change with the new workload characteristics. In this example, ITM for Virtual Environments can reestablish thresholds to more accurately alert operators that there is a problem in the environment.

Predictive trending:

Because of their dynamic nature, virtual environments can scale up or down at a moment’s notice. For this reason, it is essential to have a handle on when critical resources will near capacity so additional physical resources can be provisioned before a service interruption occurs.

The performance analytics capability, which is integrated into the Tivoli monitoring infrastructure, extends the existing resource monitoring capabilities by providing trending and forecasting capabilities. These capabilities allow IT Operations and capacity and performance analysts to monitor resource consumption trends, anticipate future performance issues – to avoid or more effectively resolve problems or bottlenecks - and helps align system capacity to current and future demand. Users can be notified 7, 30 or 90 days in advance of a resource bottleneck.

Multiple hypervisor coverage:

A large percentage of customers have deployed more than one hypervisor in their data center, and this percentage will likely grow as hypervisor technologies continue to mature. IBM Tivoli Monitoring for Virtual Environments provides broad hypervisor coverage to help reduce the cost of managing multiple hypervisors. This offering includes the following agents for comprehensive monitoring of virtual environments:

Virtual Infrastructure Agent (VI): Remote performance and availability monitoring of VMware ESX, ESXi and vCenter Server environments. This agent has been certified by VMware in the VMware Ready program.

KVM Agent: Remote performance and availability monitoring to visualize availability, performance and capacity trends for Kernel Based Virtual Machines and hosts. This agent remotely monitors KVM by connecting to each host.

Citrix XenServer Agent: Remote performance and availability monitoring for XenServer pools, hosts and virtual machines to include pool master transitions and XenServer license expiration notifications. Each agent instance connects to a XenServer™ Pool containing 1 to 16 XenServer™ hypervisors or optionally a stand-alone XenServer™ hypervisor. This agent has been certified by Citrix in the Citrix Ready program.

Citrix XenApp Agent: Monitoring of Citrix XenApp environments. This agent is installed on the server or virtual machine running Citrix XenApp.

Citrix XenDesktop Agent: Performance and Availability monitoring for Citrix XenDesktop virtual desktop infrastructure.

NetApp Storage Agent: Remote performance and availability monitoring to visualize capacity, latency, and throughput performance metrics of NetApp and IBM N series storage systems. This agent monitors the DataFabric Manager Server (DFM) and remotely connects to the DFM server to obtain performance metrics.

Cisco UCS Agent: Performance and availability monitoring of key components in a Cisco UCS environment, including chassis, blades and network fabric, featuring environment and component health scores, power and cooling metrics, and key configuration information. Customers deploying a UCS system as a special-purpose application delivery platform, such as a Virtual Desktop Infrastructure host, can use one tool set to monitor, report, and plan the future health of the UCS environment.

The Network Devices agent: This agent monitors SNMP-enabled devices. The agent identifies and notifies you of common network problems as reported by SNMP MIB-II-enabled devices. The software includes the following features: SNMPv1, SNMPv2c, SNMPv3, MIB-II network monitoring standards, historical data collection for further analysis, and automatic sampling of data and notification when certain conditions occur.

Each of these agents includes best-practice situations and expert advice, customized workspaces, historical data gathering and reporting. In addition, these agents can send application-specific events to IBM Tivoli Enterprise Console®. With these powerful capabilities, you can more effectively and efficiently manage their complete, end-to-end infrastructure from a single customizable interface.

To learn more about IBM Tivoli Monitoring for Virtual Environments and other solutions for cloud & virtualization management, please contact your IBM marketing representative or IBM Business Partner, or visit the following website: https://www.ibm.com/developerworks/servicemanagement/cvm/

Tivoli software from IBM helps organizations efficiently and effectively manage IT resources, tasks and processes to meet ever-shifting business requirements and deliver flexible and responsive IT service management, while helping to reduce costs. The Tivoli portfolio spans software for security, compliance, storage, performance, availability, configuration, operations and IT lifecycle management, and is backed by world-class IBM services, support and research.

© Copyright IBM Corporation 2010

IBM Corporation
Software Group
Route 100
Somers, NY 10589 U.S.A.

Produced in the United States of America
February 2011
All Rights Reserved

IBM, the IBM logo, ibm.com and Tivoli are trademarks or registered trademarks of International Business Machines Corporation in the United States, other countries, or both. If these and other IBM trademarked terms are marked on their first occurrence in this information with a trademark symbol (® or ™), these symbols indicate U.S. registered or common law trademarks owned by IBM at the time this information was published. Such trademarks may also be registered or common law trademarks in other countries. A current list of IBM trademarks is available on the web at “Copyright and trademark information” at ibm.com/legal/copytrade.shtml

Java and all Java-based trademarks and logos are trademarks of Sun Microsystems, Inc. in the United States, other countries, or both.

Linux is a registered trademark of Linus Torvalds in the United States, other countries, or both.

Microsoft and Windows are trademarks of Microsoft Corporation in the United States, other countries, or both.

Other product, company or service names may be trademarks or service marks of others.

References in this publication to IBM products and services do not imply that IBM intends to make them available in all countries in which IBM operates.

Product data has been reviewed for accuracy as of the date of initial publication. Product data is subject to change without notice. Any statements regarding IBM’s future direction and intent are subject to change or withdrawal without notice, and represent goals and objectives only.

IBM products are warranted according to the terms and conditions of the agreements (e.g. IBM Customer Agreement, Statement of Limited Warranty, International Program License Agreement, etc.) under which they are provided.

The customer is responsible for ensuring compliance with legal requirements. It is the customer’s sole responsibility to obtain advice of competent legal counsel as to the identification and interpretation of any relevant laws and regulatory requirements that may affect the customer’s business and any actions the customer may need to take to comply with such laws. IBM does not provide legal advice or represent or warrant that its services or products will ensure that the customer is in compliance with any law or regulation.