Data Center Intelligence ™ D ata C e n t e r I n t e l l I g e n C e ™ Virtualization in the Enterprise The Challenge of Managing Efficiency in Cloud & Virtualized Infrastructure Virtualization and consolidation have been widely embraced to reduce capital and operating costs. Initial deployments have delivered some savings, but unsure of how to increase VM densities without increasing risk, most organizations have significantly over-provisioned infrastructure. The resulting infrastructure is inefficient and resources are wasted. Cloud operational models only exacerbate these challenges. As organizations strive to increase agility, they over-provision demand buffers; they experiment with internal and external clouds without knowing how to accurately define infrastructure requirements for either environment. Additional complexities arise in determining which infrastructure is optimal for a given workload - physical, internal virtualization, internal Cloud or external Cloud. These challenges require organizations to apply analytics to environments that enable them to understand where to place workloads in order to safely maximize use of existing infrastructure. The Disruptive Element of Workload Placements Virtualization technologies introduce the opportunity to maximize efficiency of infrastructure by strate- gically placing workloads. Done right, placements can maximize the use of infrastructure. Done poorly, placements will result in excess capacity, performance and failure risks. In order to maximize efficiency, organizations require a new approach to determining infrastruc- ture requirements that is based on the optimal placements of workloads given their unique utiliza- tion patterns and personalities. A simple analogy that illustrates the requirement to consider placements is the game of Tetris™. Simply adding up the areas of the falling shapes and comparing it to the total area of the playing field is a gross oversimplification of the problem. Only by taking into account the sizes, shapes and orientations of the objects, and how they can potentially fit together, can you properly optimize the use of available space. Understanding How Your Infrastructure is Utilized A critical element in maximizing efficiency is accurately determining infrastructure requirements. Most organizations lack insight into exactly how much is required and for what purpose, so they over- provision for safety with guesstimates. Organizations that understand their fully-burdened utilization requirements - actual utilization in addition to requirements for business policies, growth or demand buffers, HA and DR - are able to control and minimize the amount of spare or wasted capacity in an environment. In Cloud environments, this means organizations are able to accurately size workloads and achieve greater efficiency. Strategically placing workloads is the key to utilizing spare capacity and minimizing it in hardware refresh cycles. CiRBA Data Center Intelligence For Efficiency Management CiRBA Data Center Intelligence (DCI) is analytics software that • Maximize density in infrastructure by determining optimal enables IT organizations to maximize the efficiency of Cloud- workload placements based, virtualized, and physical infrastructure. Only CiRBA’s • Automate to free up staff time by reducing time spent policy-driven analytics provide accurate workload placements analyzing environments and distributing information and resource allocations so that organizations can drive down costs by buying the right infrastructure, maximizing density, • Reduce risk by determining placements and allocations and automating key management processes. according to corporate policies, capacity thresholds, and technical configuration best practices. The CiRBA Difference CiRBA’s enterprise data center analytics platform is unique Efficiency management addresses the organization’s ability in its ability to provide accurate, actionable answers for to make optimal use of resources. To achieve this goal, maximizing efficiency across all major platforms for physical, CiRBA’s analytics support infrastructure transformation and virtual, and Cloud infrastructure. CiRBA’s cross-functional modernization to refresh hardware and software or migra- views of efficiency enable organizations to: tion to new platforms or Cloud models. CiRBA’s analytics also • Choose the right host infrastructure for workloads to continuously run in an environment to provide a comprehen- ensure efficiency, performance and SLA requirements are sive platform for managing efficiency. CiRBA’s intelligence met, and reduce risk. has widespread application from providing executive level views of efficiency metrics through to ongoing monitoring • Right-size infrastructure for maximum efficiency by of health and risk, supporting daily efficiency decisions, as sizing according to true workload requirements, demand well as, short and long term planning scenarios. forecasts, risk tolerance, HA & DR requirements, policies, and service level commitments. Putting the Analytics to Work Transformation E ciency & Modernization Optimization Analysis and Answer Distribution Automation D ata C e n t e r I n t e l l I g e n C e ™ CiRBA’s Advanced Analytics Engine CiRBA’s analytics simultaneously examine detailed configuration data, business policies, and utili- zation patterns and personalities to detect risks and determine the optimal workload placements and resource allocations. It is only through looking at the intersection of these three factors or constraints that organizations can control risk while optimizing the efficiency and manageability of infrastructure. Con guration Optimal Answers Business Utilization CiRBA’s template-driven analysis applies rule-based policies, advanced workload algorithms, and multi-dimensional What-If analysis to accurately model and proactively manage environments in a way not otherwise possible. CiRBA doesn’t move or motion workloads or provision capacity, rather it provides the optimal answers so that better decisions are made in the management of place- ments and allocations. CiRBA provides cross- platform, template-driven analysis that continuously runs in an environment to assess health, risk, and placement optimization opportunities. Infrastructure Transformation & Modernization Modernization is the transformation of existing infrastructure onto new technology and represents an ongoing, iterative task in IT. Whether you are virtualizing existing workloads, migrating to Cloud environments (internal or external), migrating to new platforms, or refreshing hardware, CiRBA’s analytics provide a comprehensive approach to transformation analysis and planning. CiRBA identifies which servers are good candidates for virtualization. When analyzing for VMware, Citrix, Microsoft’s Hyper-V, Zones, IBM LPARS and WPARS or mainframes, internal or external Cloud models, CiRBA quickly finds the lowest risk, maximum impact transformation. To achieve this CiRBA provides packaged and fully configurable analysis templates that incorporate best practices to streamline analysis processes. Qualifying Candidates CiRBA identifies which hardware models and CiRBA is the only virtualization solution that analyzes all the critical configura- configurations are best. tion and non-technical constraints within the environment in addition to utili- zation patterns and personalities. This ensures that organizations are able to assess transformation candidates and determine the optimal host infrastruc- ture, whether it’s physical, virtual, internal Cloud or external Cloud. Optimizing Workload Placements CiRBA’s multi-dimensional analysis ensures that workload placements respect critical concerns such as business processes, security and compliance rules, and are placed far from their disaster recovery or cluster/replication counter- parts. CiRBA relies on template-driven analysis of empirical configuration and utilization data and business policy information. By assessing these parameters against rules and algorithms that capture the preferences and constraints in a given environment, it is possible to solve for the optimal workload placements CiRBA enables you to evaluate the suitability and resource allocations. of external Cloud options vs. internal virtualization for placement of workloads. Optimizing Technology Decisions through What-if Analysis The ability to perform What-If analysis is critical to the ongoing moderni- zation process. CiRBA enables you to tune each potential scenario such as evaluating the impact of different risk levels, technologies, platforms and Cloud options. CiRBA’s analysis includes the following: • Determining the optimal host environment, hardware assets (including existing), and configurations to minimize infrastructure cost and ensure performance • Analyzing potential cross-platform migration options • Analyzing actual resource demand patterns to optimize specifications • Analyzing virtual workloads against blade vs. rack-mount vs. mid-range CiRBA enables you to choose the best options vs. mainframe vs. Cloud when leveraging multiple Clouds to determine • Assessing hypervisor technologies, versions, and configurations which workloads should go into internal vs. external Clouds. D ata C e n t e r I n t e l l I g e n C e ™ Analytics for Efficiency Management CiRBA’s analytics continuously run in an environment to provide a comprehensive platform for managing efficiency. CiRBA enables organizations to safely maximize the efficiency of Cloud and virtual infrastructure by determining the optimal host infrastructure for a workload, how much capacity is truly required, how it should be allocated, and where to place workloads in order to maximize utilization. Efficiency Assessment CiRBA enables organizations to assess efficiency and detect potential risks in Cloud, virtualized and physical infrastructure. By providing role-based views of infrastructure health, CiRBA enables capacity managers, application owners, operations teams, and executives to understand when changes need to be made to prevent and resolve capacity or configuration issues. CiRBA assesses infrastructure health through the ongoing monitoring of environ- ments to detect capacity risks, abnormal workload patterns, issues with resource allocations, and more. CiRBA enables organizations to: • Analyze utilization requirements vs. infrastructure supply to detect over and under-provi- sioning at the VM, host, cluster, and data center level • Analyze actual resource utilization vs. VM allocations to determine potential issues • Identify the fully-loaded utilization of infrastructure, which is the total of actual utili- zation and policy constraints and requirements • Identify utilization risks through workload pattern analysis such as spinning processes, high context switching, unusually low or high peak utilization levels • Identify improperly configured resources that impact performance such as over- CiRBA’s templates are based on best practices allocations or under-allocations of memory, virtual CPUs, resource pool settings for health monitoring, rebalancing and optimization analysis. • Identify idle systems and VM sprawl Forward-Looking Demand Analysis CiRBA enables organizations to proactively manage resource demands and potential risks by observing trends in historical data and modeling anticipated demands based on forecasted activity. CiRBA’s notifications and recommended actions, such as rebalancing or new placements, provide complete solutions for planning scenarios. CiRBA enables organizations to: • Monitor future state infrastructure health and requirements through trend analysis, business-led growth factors, forward looking scenarios (30 / 60 / 90 day look ahead) to prevent perform- ance degradation and outages • Add/remove physical hosts and VMs from virtual clusters • Off-line physical resources for maintenance, such as evacuating server frames • Ensure sufficient capacity is reserved to service ad-hoc capacity requests Efficiency Optimization CiRBA provides analytics for acting on efficiency assessments and forecasts by recom- mending where to place new workloads into existing infrastructure and how to rebal- ance and defrag infrastructure to optimize efficiency. CiRBA can be used to: • Defrag environments by optimizing workload placements on existing infra- structure to make room for new workloads or reduce infrastructure footprint • Proactively rebalance including placing workloads for forecasted changes • Control VM motioning by integrating affinity and anti-affinity rules into VMware DRS • Design and manage tiered services according to SLAs in order to increase cost-effectiveness • Enable ITIL compliant management of resources by planning VM moves in advance • Achieve business-centric views of efficiency by measuring efficiency and reporting on risks from a business service perspective CiRBA identifies where new workloads can fit into existing infrastructure. Operational Risk Analysis CiRBA provides comprehensive risk analysis including the ability to: • Control over resource over-commitment through resource contention probability and analyzing to risk tolerance levels • Model and plan for DR & HA scenarios at the server, blade enclosure, cabinet, PDU, room, site, power grid levels • Detect potential configuration risks such as technical incompatibilities, insufficient resource allocation, divergence from gold standard / peer / comparative environments • Predict potential contention for resources by workloads through probability-based risk analysis Customizable dashboards like this Efficiency & Risk Dashboard provide valuable role- based analytics offering actionable place- Customizable Views of Analytics & Automated ment and allocation recommendations Answer Distribution through detailed drill down reports. The sheer number of moving parts in a modern data center requires unprecedented levels of visibility. CiRBA provides the ability to create role-based views of relevant answers and efficiency metrics through customizable dashboards, dashboard templates, and bundled reports. In addition, CiRBA enables organizations to broaden the distribution of analysis results and key metrics without exposing detailed system information and workspaces to untrained users or those that do not have the appro- priate permissions through configurable UI views. Cross-functional teams can use these views to achieve business centric views of IT. This visibility provides aggregate views of the resources allocated to and used by a particular business function, thereby increasing transparency and improving decision making. An Infrastructure Manager Dashboard provides Results can also be distributed via asynchronous notifications, machine-readable extracts, views of optimization opportunities and risks at API-driven interaction, scheduled reports, and interactive portal-based delivery. a cluster, host, and VM level. D ata C e n t e r I n t e l l I g e n C e ™ 45 Vogell Road, toll Free: +1.866.731.0090 Suite 600 telephone: +1.905.731.0090 Richmond Hill, ON Fax: +1.905.770.4082 L4B 3P6 Online: www.cirba.com 58 Parker Street telephone: +44 20 7193 2047 Covent Garden Online: www.cirba.com London WC2B 5PZ England Copyright © 2011, CiRBA Inc. All rights reserved.