 |
|
Carnegie Mellon researchers save electricity with low-power processors and flash memory
October 15, 2009PITTSBURGH-Researchers at Carnegie Mellon University and Intel Labs Pittsburgh (ILP) have combined low-power, embedded processors typically used in netbooks with flash memory to create a server architecture that is fast, but far more energy efficient for data-intensive applications than the systems now used by major Internet services.
An experimental computing cluster based on this so-called Fast Array of Wimpy Nodes (FAWN) architecture was able to handle 10 to 100 times as many queries for the same amount of energy as a conventional, disk-based cluster. The FAWN cluster had 21 nodes, each with a low-cost, low-power off-the-shelf processor and a four-gigabyte compact flash card. At peak utilization, the cluster operates on less energy than a 100-watt light bulb.
The research team, led by David Andersen, Carnegie Mellon assistant professor of computer science, and Michael Kaminsky, senior research scientist at ILP, received a best paper award for its report on FAWN at the Association for Computing Machinery's annual Symposium on Operating Systems Principles Oct. 12 in Big Sky, Mont.
A next-generation FAWN cluster is being built with nodes that include Intel's Atom processor, which is used in netbooks and other mobile or low-power applications.
Developing energy-efficient server architectures has become a priority for datacenters, where the cost of electricity now equals or surpasses the cost of the computing machines themselves over their typical service life. Datacenters being built today require their own electrical substations and future datacenters may require as much as 200 megawatts of power.
"FAWN systems can't replace all of the servers in a datacenter, but they work really well for key-value storage systems, which need to access relatively small bits of information quickly," Andersen said. Key-value storage systems are growing in both size and importance, he added, as ever larger social networks and shopping Web sites keep track of customers' shopping carts, thumbnail photos of friends and a slew of message postings.
Flash memory is significantly faster than hard disks and far cheaper than dynamic random access memory (DRAM) chips, while consuming less power than either. Though low-power processors aren't the fastest available, the FAWN architecture can use them efficiently by balancing their performance with input/output bandwidth. In conventional systems, the gap between processor speed and bandwidth has continually grown for decades, resulting in memory bottlenecks that keep fast processors from operating at full capacity even as the processors continue to draw a disproportionate amount of power.
"FAWN will probably never be a good option for challenging real-time applications such as high-end gaming," Kaminsky said. "But we've shown it is a cost-effective, energy efficient approach to designing key-value storage systems and we are now working to extend the approach to applications such as large-scale data analysis."
Carnegie Mellon University
A next-generation FAWN cluster is being built with nodes that include Intel's Atom processor, which is used in netbooks and other mobile or low-power applications.
Developing energy-efficient server architectures has become a priority for datacenters, where the cost of electricity now equals or surpasses the cost of the computing machines themselves over their typical service life. Datacenters being built today require their own electrical substations and future datacenters may require as much as 200 megawatts of power.
"FAWN systems can't replace all of the servers in a datacenter, but they work really well for key-value storage systems, which need to access relatively small bits of information quickly," Andersen said. Key-value storage systems are growing in both size and importance, he added, as ever larger social networks and shopping Web sites keep track of customers' shopping carts, thumbnail photos of friends and a slew of message postings.
Flash memory is significantly faster than hard disks and far cheaper than dynamic random access memory (DRAM) chips, while consuming less power than either. Though low-power processors aren't the fastest available, the FAWN architecture can use them efficiently by balancing their performance with input/output bandwidth. In conventional systems, the gap between processor speed and bandwidth has continually grown for decades, resulting in memory bottlenecks that keep fast processors from operating at full capacity even as the processors continue to draw a disproportionate amount of power.
"FAWN will probably never be a good option for challenging real-time applications such as high-end gaming," Kaminsky said. "But we've shown it is a cost-effective, energy efficient approach to designing key-value storage systems and we are now working to extend the approach to applications such as large-scale data analysis."
Carnegie Mellon University
More Low-power Processors Current Events and Low-power Processors News Articles
 |
Intel Core 2 Quad Q9550S 2.83GHz 12M L2 Cache 1333MHz LGA775 Low Power (65W) Processor by Intel Leaders of the pack seeking monster performance, look no further. With four execution cores, the Intel Core 2 Quad processor blows through processor-intensive tasks in demanding multitasking environments and makes the most of highly threaded applications. Whether you're creating multimedia, annihilating your gaming enemies, or running compute-intensive applications at one time, new quad-core processing will change the way you do everything. Pioneer the new world of quad-core and unleash the power of multithreading. |
 |
Low-Power Processors and Systems on Chips by Christian Piguet (Editor) The power consumption of microprocessors is one of the most important challenges of high-performance chips and portable devices. In chapters drawn from Piguet's recently published Low-Power Electronics Design, this volume addresses the design of low-power microprocessors in deep submicron technologies. It provides a focused reference for specialists involved in systems-on-chips, from low-power microprocessors to DSP cores, reconfigurable processors, memories, ad-hoc networks, and embedded software. Low-Power Processors and Systems on Chips is organized into three broad sections for convenient access. The first section examines the design of digital signal processors for embedded applications and techniques for reducing dynamic and static power at the electrical and system levels.... |
 |
Processor - 1 x AMD Opteron 2427 / 2200 MHz ( 2400 MHz ) Low-Power - Socket F (1207) - L3 6 MB - PIB/WOF by Advanced Micro Devices The AMD Opteron processor, enabling simultaneous 32- and 64-bit computing, represents the landmark introduction of the AMD64 architecture. The AMD Opteron processor is designed to run existing 32-bit applications with outstanding performance and offers customers a simplified migration path to 64-bit computing. This evolutionary processor provides a dramatic leap forward in compatibility, performance, investment protection, and reduced total cost of ownership (TCO).The AMD Opteron processor provides a highly scalable architecture that delivers next-generation performance as well as a flexible upgrade path from 32- to 64-bit computing. With a single architecture designed to meet current and future business needs, the AMD Opteron processor can help to minimize the integration complexities... |
|
|
Intel Core 2 Quad Q9400S 2.66GHz 6M L2 Cache 1333MHz LGA775 Low Power (65W) Processor by Intel Leaders of the pack seeking monster performance, look no further. With four execution cores, the Intel Core 2 Quad processor blows through processor-intensive tasks in demanding multitasking environments and makes the most of highly threaded applications. Whether you're creating multimedia, annihilating your gaming enemies, or running compute-intensive applications at one time, new quad-core processing will change the way you do everything. Pioneer the new world of quad-core and unleash the power of multithreading. |
 |
Ultra-Low Power Processor Design Using Subthreshold Design Techniques by Bo Zhai (Author), David Blaauw (Author), Dennis Sylvester (Author) Power consumption is becoming worse with every technology generation. While there has been much research in recent years proposing design methods addressing this issue, one of the most efficient approaches is to reduce supply voltage. In our research, we found the optimal voltage (Vmin) for energy efficiency in CMOS technology. We analyzed the different factors affecting Vmin and find that Vmin usually lies in subthreshold voltage regime. To verify the high energy efficiency of subthreshold operation, we designed and fabricated two subthreshold processors in 0.13um technology, specifically, the Subliminal 1 and Subliminal 2 processors. Measurements confirm 2.60pJ per instruction efficiency for the Subliminal 1. Also, we designed and fabricated the first sub-200mV compact 6-T SRAM in a... |
 |
Processor - 1 x AMD Opteron 2435 / 2600 MHz ( 2400 MHz ) Low-Power - Socket F (1207) - L3 6 MB - PIB/WOF by Advanced Micro Devices The AMD Opteron processor, enabling simultaneous 32- and 64-bit computing, represents the landmark introduction of the AMD64 architecture. The AMD Opteron processor is designed to run existing 32-bit applications with outstanding performance and offers customers a simplified migration path to 64-bit computing. This evolutionary processor provides a dramatic leap forward in compatibility, performance, investment protection, and reduced total cost of ownership (TCO).The AMD Opteron processor provides a highly scalable architecture that delivers next-generation performance as well as a flexible upgrade path from 32- to 64-bit computing. With a single architecture designed to meet current and future business needs, the AMD Opteron processor can help to minimize the integration complexities... |
 |
Advanced Memory Optimization Techniques for Low-Power Embedded Processors by Manish Verma (Author), Peter Marwedel (Author) The design of embedded systems warrants a new perspective because of the following two reasons: Firstly, slow and energy inefficient memory hierarchies have already become the bottleneck of the embedded systems. It is documented in the literature as the memory wall problem. Secondly, the software running on the contemporary embedded devices is becoming increasingly complex. It is also well understood that no silver bullet exists to solve the memory wall problem. Therefore, this book explores a collaborative approach by proposing novel memory hierarchies and software optimization techniques for the optimal utilization of these memory hierarchies. Linking memory architecture design with memory-architecture aware compilation results in fast, energy-efficient and timing predictable memory... |
 |
HP 407435-B21 Opteron 2214 2.2GHz HE Processor for DL385 G2 by Hewlett Packard With advances like high-performing, low-power DDR2 memory and hardware-assisted AMD Virtualization, Second-Generation AMD Opteron processors with Direct Connect Architecture are designed to advance AMD's performance-per-watt capabilities and extend the leadership in x86 virtualization, while maintaining the same thermal infrastructure and helping you leverage existing investments.structure and helping you leverage existing investments.AMD Opteron processors feature a Common Core Architecture that is consistent across 1-way, 2-way, 4-way, and 8-way systems and is also consistent with previous AMD Opteron processors, helping minimize the cost of transition and maximize past investments in software optimization. |
 |
Designing Embedded Processors: A Low Power Perspective by J. Henkel (Editor), S. Parameswaran (Editor) As we embrace the world of personal, portable, and perplexingly complex digital systems, it has befallen upon the bewildered designer to take advantage of the available transistors to produce a system which is small, fast, cheap and correct, yet possesses increased functionality. Increasingly, these systems have to consume little energy. Designers are increasingly turning towards small processors, which are low power, and customize these processors both in software and hardware to achieve their objectives of a low power system, which is verified, and has short design turnaround times. Designing Embedded Processors examines the many ways in which processor based systems are designed to allow low power devices. It looks at processor design methods, memory optimization,... |
 |
HP 408840-B21 Opteron 2216 2.4GHz HE Processor for DL385 G2 by Hewlett Packard With advances like high-performing, low-power DDR2 memory and hardware-assisted AMD Virtualization, Second-Generation AMD Opteron processors with Direct Connect Architecture are designed to advance AMD's performance-per-watt capabilities and extend the leadership in x86 virtualization, while maintaining the same thermal infrastructure and helping you leverage existing investments.structure and helping you leverage existing investments.AMD Opteron processors feature a Common Core Architecture that is consistent across 1-way, 2-way, 4-way, and 8-way systems and is also consistent with previous AMD Opteron processors, helping minimize the cost of transition and maximize past investments in software optimization. |
| © 2009 BrightSurf.com |