Data Center Power & Cost Management

Computers drive progress in today’s world. Both individuals and industry depends on a spectrum of computing tools. Data centers are at the heart of many computational processes from communication to scientific analysis. They also consume over 3% of total power in the United States, and this amount continues to increase.[1]

Data centers service jobs, submitted by their customers, on the data center’s servers, a shared resource. Data centers and their customers negotiate a service-level agreement (SLA), which establishes the average expected job completion time. Servers are allocated for each job and must satisfy the job’s SLA. Job-scheduling software already provides some solutions to the budgeting of data center resources.

Data center construction and operation include fixed and accrued costs. Initial building expenses, such as purchasing and installing computing and cooling equipment, are one-time costs and are generally unavoidable. An operational data center must power this equipment, contributing an ongoing cost. Power management and the associated costs define one of the largest challenges for data centers.

To control these costs, the future of data centers is in active participation in advanced power markets. More efficient cooling also provides cost saving opportunities, but this requires infrastructure updates, which is costly and impractical for existing data centers. Fortunately, existing physical infrastructure can support participation in demand response programs, such as peak shaving, regulation services (RS), and frequency control. In demand-response programs, consumers adjust their power consumption based on real-time power prices. The most promising mechanism for data center participation is RS.

Independent system operators (ISOs) manage demand response programs like RS. Each ISO must balance the power supply with the demand, or load, on the power grid in the region it governs. RS program participants provide necessary reserves when demand is high or consume more energy when demand is lower than the supply. The ISO communicates this need by transmitting a regulation signal, which the participant must follow with minimal error. In return, ISOs provide monetary incentives to the participants.

This essay appears in Circuit Cellar #293 (December 2014).

Data centers are ideal participants for demand response programs. A single data center requires a significant amount of power from the power grid. For example, the Massachusetts Green High-Performance Computing Center (MGHPCC), which opened in 2012, has power capacity of 10 MW, which is equivalent to as many as 10,000 homes ( Additionally, some workload types are flexible; jobs can be delayed or sped up within the given SLA.

Data centers have the ability to vary power consumption based on the ISO regulation signal. Server sleep states and dynamic voltage and frequency scaling (DVFS) are power modulation techniques. When the regulation signal requests lower power consumption from participants, data centers can put idle servers to sleep. This successfully reduces power consumption but is not instantaneous. DVFS performs finer power variations; power in an individual server can be quickly reduced in exchange for slower processing speeds. Demand response algorithms for data centers coordinate server state changes and DVFS tuning given the ISO regulation signal.

Accessing data from real data centers is a challenge. Demand response algorithms are tested via simulations of simplified data center models. Before data centers can participate in RS, algorithms must account for the complexity in real data centers.

Data collection within data center infrastructure enables more detailed models. Monitoring aids performance evaluation, model design, and operational changes to data centers. As part of my work, I analyze power, load, and cooling data collected from the MGHPCC. Sensor integration for data collection is essential to the future of data center power and cost management.

The power grid also benefits from data center participation in demand response programs. Renewable energy sources, such as wind and solar, are more environmentally friendly than traditional fossil fuel plants. However, the intermittent nature of such renewables creates a challenge for ISOs to balance the supply and load. Data center participation makes larger scale incorporation of renewables into the smart grid possible.

The future of data centers requires the management of power consumption in order to control costs. Currently, RS provides the best opportunities for existing data centers. According to preliminary results, successful participation in demand response programs could yield monetary savings around 50% for data centers.[2]

[1] J. Koomey, “Growth in Data Center Electricity Use 2005 to 2010,” Analytics Press, Oakland, August, 1, 2010,

[2] H. Chen, M. Caramanis, and A. K. Coskun, “The Data Center as a Grid Load Stabilizer,” Proceedings of the Asia and South Pacific Design Automation Conference (ASP-DAC), p. 105–112, January 2014.

LaneTTF Annie Lane studies computer engineering at Boston University, where she performs research as part of the Performance and Energy-Aware Computing Lab ( She received the Clare Boothe Luce Scholar Award in 2014. Annie received additional funding from the Undergraduate Research Opportunity Program (UROP) and Summer Term Alumni Research Scholars (STARS). Her research focuses strategies power and cost optimization strategies in data centers.


Remote Control and Monitoring of Household Devices

Raul Alvarez, a freelance electronic engineer from Bolivia, has long been interested in wireless device-to-device communication.

“So when the idea of the Internet of Things (IoT) came around, it was like rediscovering the Internet,” he says.

I’m guessing that his dual fascinations with wireless and the IoT inspired his Home Energy Gateway project, which won second place in the 2012 DesignSpark chipKIT challenge administered by Circuit Cellar.

“The system enables users to remotely monitor their home’s power consumption and control household devices (e.g., fans, lights, coffee machines, etc.),” Alvarez says. “The main system consists of an embedded gateway/web server that, aside from its ability to communicate over the Internet, is also capable of local communications over a home area wireless network.”

Alvarez catered to his interests by creating his own wireless communication protocol for the system.

“As a learning exercise, I specifically developed the communication protocol I used in the home area wireless network from scratch,” he says. “I used low-cost RF transceivers to implement the protocol. It is simple and provides just the core functionality necessary for the application.”

Figure1: The Home Energy Gateway includes a Hope Microelectronics RFM12B transceiver, a Digilent chipKIT Max32 board, and a Microchip Technology ENC28J60 Ethernet controller chip.

Figure 1: The Home Energy Gateway includes a Hope Microelectronics RFM12B transceiver, a Digilent chipKIT Max32 board, and a Microchip Technology ENC28J60 Ethernet controller chip.

Alvarez writes about his project in the February issue of Circuit Cellar. His article concentrates on the project’s TCI/IP communications aspects and explains how they interface.

Here is his article’s overview of how the system functions and its primary hardware components:

Figure 1 shows the system’s block diagram and functional configuration. The smart meter collects the entire house’s power consumption information and sends that data every time it is requested by the gateway. In turn, the smart plugs receive commands from the gateway to turn on/off the household devices attached to them. This happens every time the user turns on/off the controls in the web control panel.

Photo 1: These are the three smart node hardware prototypes: upper left,  smart plug;  upper right, a second smart plug in a breadboard; and at bottom,  the smart meter.

Photo 1: These are the three smart node hardware prototypes: upper left, smart plug; upper right, a second smart plug in a breadboard; and at bottom, the smart meter.

I used the simple wireless protocol (SWP) I developed for this project for all of the home area wireless network’s wireless communications. I used low-cost Hope Microelectronics 433-/868-/915-MHz RFM12B transceivers to implement the smart nodes. (see Photo 1)
The wireless network is configured to work in a star topology. The gateway assumes the role of a central coordinator or master node and the smart devices act as end devices or slave nodes that react to requests sent by the master node.

The gateway/server is implemented in hardware around a Digilent chipKIT Max32 board (see Photo 2). It uses an RFM12B transceiver to connect to the home area wireless network and a Microchip Technology ENC28J60 chip module to connect to the LAN using Ethernet.

As the name implies, the gateway makes it possible to access the home area wireless network over the LAN or even remotely over the Internet. So, the smart devices are easily accessible from a PC, tablet, or smartphone using just a web browser. To achieve this, the gateway implements the SWP for wireless communications and simultaneously uses Microchip Technology’s TCP/IP Stack to work as a web server.

Photo 2: The Home Energy Gateway’s hardware includes a Digilent chipKIT Max32 board and a custom shield board.

Photo 2: The Home Energy Gateway’s hardware includes a Digilent chipKIT Max32 board and a custom shield board.

Thus, the Home Energy Gateway generates and serves the control panel web page over HTTP (this page contains the individual controls to turn on/off each smart plug and at the same time shows the power consumption in the house in real-time). It also uses the wireless network to pass control data from the user to the smart plugs and to read power consumption data from the smart meter.

The hardware module includes three main submodules: The chipKIT Max 32 board, the RFM12B wireless transceiver, and the ENC28J60 Ethernet module. The smart meter hardware module has an RFM12B transceiver for wireless communications and uses an 8-bit Microchip Technology PIC16F628A microcontroller as a main processor. The smart plug hardware module shows the smart plugs’ main hardware components and has the same microcontroller and radio transceiver as the smart meter. But the smart plugs also have a Sharp Microelectronics S212S01F solid-state relay to turn on/off the household devices.

On the software side, the gateway firmware is written in C for the Microchip Technology C32 Compiler. The smart meter’s PIC16F628A code is written in C for the Hi-TECH C compiler. The smart plug software is very similar.

Alvarez says DIY home-automation enthusiasts will find his prototype inexpensive and capable. He would like to add several features to the system, including the ability to e-mail notifications and reports to users.

For more details, check out the February issue now available for download by members or single-issue purchase.