Newsletter

Power semiconductors key processing density improvements

Advanced power stages and dynamic power monitoring deliver data-center efficiency improvements

Page 1 of 3

Power Management DesignLine

From e-mail to online buying and selling, social networking to gaming, and VoiP telephony to video on demand, recent years have seen exponential growth in e-commerce, internet communication, and online or downloadable entertainment. This growth has fuelled ever-increasing demand for the electronic systems " many of them in secure data centers - that are needed to store, process and transfer the underlying electronic data 24 hours a day, 365 days a year. And there are very real economic, legislative and environmental pressures to make these systems as efficient as possible. Increasing the amount of processing power per m2 is key and ultimately depends on the efficiency of the power system architecture.

A typical data center will include servers, storage area networks, routers and switches. In the past, these centers have often been measured on their performance density (MIPS/m2). However, this is changing for a number of reasons.

Firstly, lifetime operation costs for the equipment in the data center are now estimated at about three times initial capital outlay. Secondly, an increased concern on the environmental impact of electricity usage is driving legislation to maximise efficiency. And, thirdly, there is genuine concern that, without efficiency improvements, many data centers are going to be unable to handle the growth in processing power and supporting infrastructure needed to meet demand. Recent research from ARI, for example, found that 38% of data centers are already running at dangerous levels of power consumption and are at risk of failing due to a power shortfall.

As a result, the key Figure of Merit (FOM) metric as we move forward is performance efficiency, or MIPS/W. And, by improving both equipment efficiency and the efficiency of the supporting infrastructure, the latest power semiconductor technologies are playing a significant role in driving up this FOM.

Power consumption in the data center

We can identify two general sources of power consumption in the data center. The first is the processing, storage, switching and routing equipment itself. The second is the infrastructure required to cool and protect these servers, storage networks, switchers and routers. The energy usage from each is about equal and they are directly related.

Equipment energy consumption consists of three main elements. The electronic loads such as microprocessors and memory banks consume 60 to 70 percent of energy, while power supplies consume 25 to 30 percent, and fans use five percent.

While there have been significant advances in reducing the load's power profile (for example, the introduction of multi-core efficient processors and virtualization technology), there are other opportunities for engineers to significantly reduce the consumption of all three of these major energy consumers. New smart-power management systems, for example, include the co-design of several critical components of the power supplies that are integrated into the platform. Based on advanced power semiconductor technologies, the key elements of these power systems are highly-efficient and dense power stages, advanced highly responsive power controllers, digital interfaces for programmability and diagnostics, accurate power monitors, system controllers, and sequencing.



Page 2: Advanced Power Stages  

Page 1 | 2 | 3







Related Content

TECH PAPER
1. Simple Calibration Circuit Maximizes Accuracy in Li-Ion Battery Management Systems

WEBINAR
2. Energy Efficient "Green" Design Test Tips - Part I of II

TECH PAPER
3. Intelligent Digital Power Management

WEBINAR
4. How to Get the Most from Your Low Current Measurement Instruments

 


 Featured Jobs
Ascension Health seeking Solutions Development Analyst in St. Louis, MO

National Semiconductor seeking Principal IC Design Engineer in Santa Clara, CA

Taylor Guitars seeking Sr. Web Designer in El Cajon, CA

Covidien seeking Hardware Manager in Boulder, CO

Sierra Nevada seeking Software Engineer in Hagerstown, MD

More jobs on EETimesCareers
 Sponsor
 CAREER CENTER
Ready to take that job and shove it?
SEARCH JOBS:

 SPONSOR

 RECENT JOB POSTINGS
For more great jobs, career related news, features and services, please visit EETimes' Career Center.