Abstract
With energy efficiency one of the main challenges on the way towards ultrascale systems, there is a great need for access to high-quality energy consumption data. Such data would enable researchers and designers to pinpoint energy inefficiencies at all levels of the computing stack, from whole nodes down to critical regions of code. However, measurement capabilities are often missing, and significantly differ between platforms where they exist. A standard is yet to be established. To that end, this paper attempts an extensive survey of energy measurement tools currently available at both the hardware and software level, comparing their features with respect to energy monitoring.
Highlights
Energy sustainability is a significant concern for high-performance computing, with cost of operation due to power draw being one of the main limiting factors for the design of new systems
A low level of precision and temporal resolution can be sufficient for administration and maintenance purposes, but many interesting applications, such as application energy efficiency analysis or energy-aware dynamic scheduling, require finer-grained measurements
External custom designs have been developed for previous work such as PowerScope [13], which used a digital multimeter with a trigger input connected to profiling software, or the Energy Endoscope [43], where an application-specific integrated circuit (ASIC) dedicated to real-time energy monitoring was built
Summary
Energy sustainability is a significant concern for high-performance computing, with cost of operation due to power draw being one of the main limiting factors for the design of new systems. This need to improve the energy efficiency of computation is compounded by the growth to ultrascale infrastructure. Current data center management standards (such as the Intelligent Platform Management Interface (IPMI) [21] or the Data Center Manageability Interface (DCMI) [20]) cannot leverage most sensors, providing instead low-resolution data from supported motherboards For these reasons, we believe there is a justification for standardization of energy data acquisition techniques. The final section concludes the paper and outlines future work
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
