The nation’s data centers face a looming crisis caused by rising energy costs, supply shortages, and regulatory oversight. With annual energy consumption costing $4.5 billion (per the EPA’s landmark 2006 Report to Congress on Server and Data Center Energy Efficiency), what was once a benign “facilities expense” has emerged as a central issue in green computing.
The same EPA report reveals that data centers consume 2 percent of the nation’s electricity and are a primary contributor to greenhouse gases—44 million tons annually. Since 1999, energy costs have increased by almost 50 percent, and trends indicate consumption will double by 2011.
Initiatives such as modernizing data centers and applications, upgrading hardware and cooling systems, adopting virtualization and consolidation, and more extreme solutions—such as relocating data centers to areas where the energy supply is less expensive and more abundant—have been recommended. While these activities will eventually provide a “green ROI,” they’re long-term, potentially disruptive, and require a significant upfront investment. These initiatives also ignore the near-term benefits of leveraging low-cost, easy-to-implement, software-based solutions such as Workload Automation (WLA) tools.
Effective WLA solutions can influence power consumption and enforce energy management policies across the enterprise. Effectively implementing a modern WLA solution allows data centers to gain an immediate green ROI by managing a workload that:
- Dynamically balances workload and supports virtualization. In addition to reducing hardware energy requirements, this provides a corresponding reduction in cooling costs. According to IDC, by 2010, cooling costs will be 70 cents on the dollar for each dollar spent on new hardware.
- Reduces workload complexity and eliminates the need for multiple tools. A modern WLA tool simplifies workloads through consolidation on a single tool and seamless integration of batch and real-time processes (such as file transfers, Web services, and Java 2 Platform Enterprise Edition [J2EE] elements), which reduces processor load and execution times for the entire business process flow.
- Leverages innovative technologies such as agentless scheduling and job level virtualization. These enhance virtualization by routing workloads based on energy policies—even if the machine doesn’t have a WLA tool component.
- Reduces staff requirements. Staff can be redeployed to other modernization or continuous improvement initiatives.
- Enables Business Service Management (BSM) capabilities. The organization can proactively identify and avoid workload issues, thereby maximizing service uptime and quality.
- Governs execution using a diverse array of physical and logical resource controls to align workload with multiple energy policies. Logical resource elements tied to these categories can be dynamically adjusted by the tool or via external methods to suspend or “throttle down” submission of workload for each category. The tool also can further drive energy savings by dynamically shifting low priority work to off-peak hours or route it to locations where the energy costs are lower for that time period.
- Leverages capabilities such as enterprisewide forecasting. Sites can identify opportunities to shut down devices during periods when service level workload isn’t scheduled to run. Servers and I/O devices consume up to 40 percent of their maximum power when idle, according to Eaton Corp. and other IT electrical system and component vendors. Using the WLA tool to help identify when they can be powered down will generate significant savings and reduce cooling system demands.
- Facilitates comprehensive trend reporting. As the Uptime Institute indicates, an estimated 30 percent of corporate servers and I/O devices can be consolidated. The WLA tool’s database can generate workload trends and the utilization of the devices that support it. A modern WLA tool includes historical statistical details for every job to identify workload activity and execution location, statistical analysis of execution times, and mainframe and server utilization metrics.
- Positions the organization to leverage future innovations as these tools continue to evolve to manage additional elements and devices. The WLA tools directly control the power status of servers and other devices based on the real-time demands of production. Industry research shows that evolving from an “always on” to an “always available” paradigm can reduce energy consumption and related cooling costs of targeted devices by 50 percent.
Data centers that consolidate workloads on a modern WLA tool can accelerate their green ROI. The successful deployment of a WLA tool provides a greater ROI by reducing IT and business costs and improving the quality and availability of the organization’s critical business services. Z