Difference between revisions of "FOSDEM 2014 Energy-efficient Computing devroom"

Latest revision as of 17:37, 7 January 2014

The MAGEEC project will be hosting the Energy-efficient Computing Developer Room at FOSDEM 2014. The focus for this will be wider than the MAGEEC project, but will remain at the level of energy-aware computing across the system stack.

The CFP is included in full below for circulation.

The schedule will be added here as it develops.

Call for participation (now closed)

We're delighted to announce the call for participation for the Energy-efficient Computing developer room at FOSDEM 2014.

FOSDEM is held annually at the Université Libre de Bruxelles and takes place over the first weekend in February. The developer room will run on Sunday 2nd February only.

While energy-efficient computing is a broad domain that extends well beyond computing devices, the developer room focus is on energy-aware computing at the different levels within the computer system stack. Topics include, but are not limited to, techniques to measure the energy consumption of applications, energy-efficient compilation, energy-performance trade-offs, energy-aware operating systems, benchmarks, hardware that supports energy-efficient system design.

Proposals are solicited in the area of energy-efficient computing under the following categories: a) 10-minute lightning talks; b) 30-minute technical presentations; c) workshop contributions.

Lightning talks should aim to give the audience a brief introduction to a new topic or innovation, with the expectation that technical presentations will provide a more in-depth analysis of more developed concepts.

A workshop is planned at which participants will have the opportunity to use instrumented hardware to measure the energy consumption of applications. Platforms available will include BeagleBone, STM32F4DISCOVERY and the Arduino-compatible, Shrimp. Contributions to the workshop may include applications which will have their energy consumption measured and additional hardware platforms to be instrumented.

All contributions should be aimed at a technical audience and related to F/OSS and/or, where applicable, open source hardware.

To propose a lightning talk, presentation or workshop contribution visit:

 https://penta.fosdem.org/submission/FOSDEM14

And please ensure that you select the *Energy-efficient computing devroom* track.

The deadline for submissions is Monday 25th November (extended from 18/11).

The developer room is being organised by the MAGEEC project and any questions should be directed to the mailing list:

 http://mageec.org/cgi-bin/mailman/listinfo/mageec

Schedule

09:00-09:30 — Energy scavenging, battery life and should we build more power stations, Dr Jeremy Bennett (Embecosm)

This introductory talk will set the context for the day. It will take a look at how energy efficiency is *the* major challenge for systems developers, and will then provide an overview of a number of open source projects that demonstrate how the energy efficiency of the entire system can be significantly improved.

Energy efficiency of systems - hardware and software - matters. For the smallest energy scavenging systems it is a matter of eking out the picowatts. For handheld consumer electronics it is battery life that is a key product differentiator. Even for mains powered consumer devices, energy efficiency affects utility bills. And for datacenters run by the Googles and Facebooks of this world, more efficient systems mean fewer new power stations need to be built.

This dev room is dedicated to the whole subject of energy efficiency in computer systems. This introductory talk provides an overview of all the approaches being taken to address this issue, particularly looking at how free and open source hardware and software is taking a leading role. It will provide a guide to the remaining sessions of the day, including the hands-on workshop where participants will have the opportunity to work with energy measurement hardware for themselves.

09:30-10:15 — Measuring energy consumption in embedded systems, Simon Hollis (University of Bristol)

In this talk, I will introduce the need for energy measurements for embedded devices and show how they may be performed accurately and for very low cost using a combination of off-the shelf parts and a wide range of target embedded systems.

I will cover the basic physics of energy measurement and go on to display designs for energy measurement kits, including the power sensing boards recently developed as part of the MAGEEC research project.

10:15-10:45 — An approach for energy consumption analysis of programs using LLVM, Kyriakos Georgiou & Neville Grech (University of Bristol)

Energy models can be constructed by characterizing the energy consumed by executing each instruction in a processor's instruction set. This can be used to determine how much energy is required to execute a sequence of assembly instructions. However, statically analysing low level program structures is hard, and the gap between the high-level program structure and the low-level energy models needs to be bridged. We have developed a tool for performing a static analysis on the intermediate compiler representations of a program. Specifically, we target LLVM IR, a representation used by most modern compilers including Clang.

One of the challenges in doing so is that of determining an energy cost of executing LLVM IR program segments, for which we have developed a mapping tool. This tool associates LLVM IR program segments with assembly program segments, producing a mapping. Mapping information is useful when performing an analysis at one layer using energy models defined at a lower layer. Essentially, this propagates the energy model defined at the instruction set level up to the LLVM IR level, at which the analysis is performed. When this is used with our analysis tool, we are able to infer energy formulae that characterize the energy consumption for a particular program. This approach can be applied to any languages targeting the LLVM toolchain or architectures supported by LLVM.

Static energy estimation has applications in program optimization, and enables energy-aware software development.

10:45-11:45 — spEEDO: Energy Efficiency through Debug suppOrt, Dr David J Greaves (University of Cambridge Computer Laboratory)

The spEEDO project aims to augment existing debug APIs (such as GNU's RSP and ARM's Coresight) with a power component for reporting and tracing energy use in multicore systems-on-chip. Energy is logged per IP block and per application thread and reports are made available to the operating system, to applications programs and over the debug interface. The aim is facilitate optimizations for energy-efficiency at all stages of software and silicon development. Dr Greaves will report on the current state of the API and solicit adoption by and improvements from the SoC community.

11:45-12:15 — Open Energy Measurement Hardware, James Pallister (University of Bristol)

I will discuss how to measure energy consumption and show off the University of Bristol-designed energy monitor. This board can sample energy up to 6 million samples per second and the designs are open. This will allow fine grain measurements of energy consumption, and power profiling of applications to find the energy hot-spots of a program.

12:15-12:30 — Open Low Power Devices, Emilio Monti (mbed)

mbed is an open platform for developing ARM-based low power embedded systems (with a focus on IoT devices).

This talk will provide an overview about:

• why you might want to base your next low power device on the mbed platform
• how to start developing only using the Free GNU GCC Toolchain and the open mbed SDK (Apache v2)
• the measuring of the energy consumption of an mbed

12:30-15:30 — Workshop: Measuring application energy consumption with instrumented hardware, MAGEEC Team

Bring along your applications and have their energy consumption measured on a pre-instrumented Arduino, Raspberry Pi or BeagleBone. Alternatively, bring along your own design on a breadboard and we'll hook up a PowerSense shield to measure the energy usage.

15:30-16:15 — MAchine Guided Energy-Efficient Compilation, Simon Cook (Embecosm)

MAGEEC, a collaboration between the open source software house, Embecosm, and Bristol University's microcomputer group, aims to use machine learning to improve the energy efficiency of compiled code. This entirely open source project is funded by the UK government through the Technology Strategy Board, and aims to provide working systems based on LLVM and GCC by the end of 2014.

16:15-16:45 — EACOF, The Energy-Aware COmputing Framework, Hayden Field & James Pedlingham

EACOF, an Energy Aware COmputing Framework, is a modular framework that provides a layer of abstraction between sources of energy data and the applications that exploit them. It replaces platform specific instrumentation with two APIs, to input and output data from the framework. This design allows developers to profile their code for energy consumption in a simple and portable manner.

This talk will provide an overview of the structure and implementation of EACOF. It will also demonstrate how the framework can be integrated into real code to provide useful information about software energy consumption.