Are we really powerless to spend on soaring data center energy consumption?

Jeff Klaus

through IT Architecture agility and high controllability to meet rapidly rising energy costs

today, every aspect of data center management is booming, and operating budgets are hard to keep pace with. According to statistics, the total amount of data will be doubled every month, the application end load will be doubled every 2 years, but the operating budget will take 8 years to double.

people have been asking for The IT industry is doing more at a lower cost, and in recent years data centers have been increasingly demanding dynamic features. Smart devices, big data, virtualization, and cloud technologies are changing the way data centers deliver services, so flexibility, resiliency, and scalability are becoming increasingly important.

as a result of these rapid changes, the various aspects of data center management have become increasingly complex. Thousands of devices on the internal network are constantly connected and disconnected. At the same time, the mobile computing resource pool is automatically assigned to manage. So, in this hyper-dynamic environment, is it not that the IT management team can effectively identify idle and overloaded resources?

If that is the case, energy consumption in data centers will continue to soar. According to a new report by the Natural Resources Conservation Commission, the energy consumption of data centers is now staggering to the 10% of the global total energy consumption .

Fortunately, even if the challenges facing the data center are so severe, It industry is not powerless.

The discovery of secret weapons

Ironically, today's data centers are mostly made up of software-defined resources, but the secret weapon that controls the cost of energy is dependent on the hardware. Rack-and-blade servers, switches, power distribution units, and many other data center devices provide significant energy and temperature information during operation. Given the sheer size and variety of hardware devices in the data center that makes it cumbersome to manually collect and apply information resources, it also facilitates the development of energy management solution providers.

As a result, data center managers now have a variety of options. They can take advantage of the built-in energy management management console to enable integrators to integrate energy management middleware software into existing management consoles or to independently deploy energy management intermediate software scenarios to obtain the required functionality.

regardless of which approach is deployed, a comprehensive set of energy management solutions allows the IT management team views, records and analyzes the energy and temperature operating states of the entire data center. Automatic acquisition and concentration of energy and temperature information can be further drawn into the data center for each space of the image information, the data can be used to analyze and identify the corresponding operating laws, while understanding the data center workload and other changes.

Clearly visible, archived power information enables data center managers to have a basic understanding of power usage to make better decisions about data center planning and optimization.

The best energy management solutions take optimization to a higher level by combining automatic monitoring of recording systems with real-time control functions. For example, if conditions allow, or at the right time, some servers or racks can be set to limit the power to a certain extent. For servers that are idle for longer than a certain length of time, they can be automatically converted to a more power-saving sleep mode. The use of electricity can be flexibly distributed according to the business center of gravity, and this function can also effectively prolong the maintenance time of the standby power supply during power outage. An excellent energy-efficient solution can even dynamically adjust the frequency of the server to minimize energy consumption without negatively impacting service quality and application performance.

Data centers that care about energy levels take advantage of these capabilities to achieve a range of operational goals, such as accurately planning workloads, saving operating costs, prolonging data center equipment life, and achieving relevant environmental regulations.

Application situation and effect proof

In practice, customers reveal several common deployment motivations and provide insight into how the overall energy solution and its best practices can be achieved, and to what extent.

Energy consumption Monitoring:

identifying and understanding peak energy usage is a source of power for many companies to introduce energy management solutions. By understanding this information, customers can reduce the power consumption of 15% during peak hours , and even in the event of a surge in power demand during peak hours, it is possible to reduce the monthly electricity costs of the data center. Energy monitoring is also used to accurately charge hosting services and other service users.

Lifting Enclosure Density:

The room area is also one of the factors restricting the expansion of many data centers. In the absence of real-time information, the traditional fixed power supply method relies on the rated value of the power supply, or based on the non-rated value of the experimental data. Real-time power monitoring generally proves that the actual power consumption is generally far below the rated value. With the monitoring and power capping mechanism in place, the data center can increase the number of racks more proactively, 60% or even 80% the server cabinet density at the same power range .

Timely discovery of idle or low-utilization servers:

similar "ghost" servers consume up to half of their power consumption during peak workloads. Energy management Solutions show that at any point in time, such servers account for 10% to 15% of all servers in the data center, which helps data center managers better merge and virtualize to avoid this power and space waste.

Early detection of potential failures:

power monitoring and control not only monitors dangerous hot spots and automatically alerts, but also increases 15% standby power uptime and extends 25% business continuity during power outages .

Advanced Temperature Control:

the real-time temperature information acquisition function helps to draw an intuitive thermal image of the data center without the need for an expensive temperature sensor. The use of temperature images can significantly improve the monitoring of the data center, which can effectively cover the entire floor, or even every device. . These temperature maps can also improve capacity planning, helping to avoid insufficient refrigeration or excess. With more intuitive images and limited values, data center management can increase the operating environment temperature of the engine room more safely. The ambient temperature of 1 degrees Celsius, you can save 5% to 10% cooling costs.

Effective balance of energy consumption and performance:

The data center can effectively reduce the power consumption of 15% to 20% by using a smarter-designed processor instead, without impacting or virtually compromising performance .

It's time to take the energy problem seriously.

the hardware facilities of the data center remain pivotal. The evolving software that allocates resources to applications and services requires real-time granular monitoring of hardware devices. The introduction of energy management solutions makes this monitoring mechanism, as well as energy and temperature control possible, and allows it management to effectively control power resources, which have long been the largest single expenditure on the operating budget table.

Software and middleware solutions enable data center managers to effectively monitor hardware and their health, making automated processes faster, safer, and more cost-effective, avoiding a spike in power costs. In today's energy-scarce data centers, there is a need for energy-aware virtual machine migrations and job scheduling to be an industry standard.

For more product information, please log in:

Http://www.intel.cn/content/www/cn/zh/software/data-center-overview.html,

or send an email to [email protected] Contact us.

Author Introduction:

Jeff · Mr. Klaus is the United States Intel Corporation data Center Manager ( DCM ) Solutions general Manager. At Intel Corporation, Jeff • Klaus has - years of team management experience. His team is developing a middleware product for energy and thermal management. Jeff · Mr. Klaus holds an MBA from Boston University MBA degree.

This article is from the Intel Data Center Management specialist blog, so be sure to keep this source http://inteldcm.blog.51cto.com/11451124/1767527

Are we really powerless to spend on soaring data center energy consumption?