Talking at the 2016 ARM Research Summit

Last week there was an inaugural ARM Research Summit.

https://developer.arm.com/research/summit

I gave a talk on Power & Compute Codesign for “Little Digital” Electronics.

Here are the slides of this talk:

https://www.staff.ncl.ac.uk/alex.yakovlev/home.formal/Power-and-Compute-Talk

Here is the abstract of my talk:

Power and Compute Codesign for “Little Digital” Electronics

Alex Yakovlev, Newcastle University

alex.yakovlev@ncl.ac.uk

The discipline of electronics and computing system design has traditionally separated power management (regulation, delivery, distribution) from data-processing (computation, storage, communication, user interface). Power control has always been a prerogative of power engineers who designed power supplies for loads that were typically defined in a relatively crude way.

 

In this talk, we take a different stance and address upcoming electronics systems (e.g. Internet of Things nodes) more holistically. Such systems are miniaturised to the level that both power management and data-processing are virtually inseparable in terms of their functionality and resources, and the latter are getting scarce. Increasingly, both elements share the same die, and the control of power supply, or what we call here a “little digital” organ, also shares the same silicon fabric as the power supply. At present, there are no systematic methods or tools for designing “little digital” that could ensure that it performs its duties correctly and efficiently.  The talk will explore the main issues involved in formulating the problem of and automating the design of little digital circuits, such as models of control circuits and the controlled plants, definition and description of control laws and optimisation criteria, characterisation of correctness and efficiency, and applications such as biomedical implants, IoT ‘things’ and WSN nodes.

 

Our particular focus in this talk will be on power-data convergence and ways of designing energy-modulated systems [1].  In such systems, the incoming flow of energy will largely determine the levels of switching activity, including data processing – this is fundamentally different from the conventional forms where the energy aspect simply acts as a cost function for optimal design or run-time performance.

 

We will soon be asking ourselves questions like these: For a given silicon area and given data processing functions, what is the best way to allocate silicon to power and computational elements? More specifically, for a given energy supply rate and given computation demands, which of the following system designs would be better? One that involves a capacitor network for storing energy, and investing energy into charging and discharging flying capacitors through computational electronics which would be able to sustain high fluctuations of the Vcc (e.g. built using self-timed circuit). The other one that involves a switched capacitor converter to supply power as a reasonably stable Vcc (could be a set of levels). In this latter case, it would be necessary also to invest some energy into powering control for the voltage regulator. In order to decide between these two organisations, one would need to carefully model both designs and characterise them in terms of energy utilisation and delivery of performance for the given computation demands. At present, there are no good ways for co-optimising power and computational electronics.

 

Research in this direction is in its infancy and this is only a tip of the iceberg. This talk will shed some light on how we are approaching the problem of power-data co-design at Newcastle, in a series of research projects producing novel types of sensors, ADCs, asynchronous controllers for power regulation, and software tools for designing “little digital” electronics.

[1] A. Yakovlev. Energy modulated computing. Proceedings of DATE, 2011, Grenoble,  doi: 10.1109/DATE.2011.5763216

My vision of Bio-inspired Electronic Design

I took part in a Panel on Bio-inspired Electronic Design Principles at the

Here are my slides

The quick summary of these ideas is here:

 

Summary of ideas for discussion from Alex Yakovlev, Newcastle University

 

With my 30 years of experience in designing and automating the design of self-timed (aka asynchronous) systems, I have been involved in studying and exploiting in practice the following characteristics of electronic systems:  inherent concurrency, event-driven and causality-based processing, parametric variation resilience, close-loop timing error avoidance and correction, energy-proportionality, digital and mixed-signal interfaces. More recently, I have been looking at new bio-inspired paradigms such as energy-modulated and power-adaptive computing, significance-driven approximate computing, real-power (to match real-time!) computing, computing with survival instincts, computing with central and peripheral powering and timing, power layering in systems architecting, exploiting burstiness and regularity of processing etc.

In most of these the central role belongs to the notion of energy flow as a key driving force in the new generation of microelectronics. I will therefore be approaching most of the Questions raised for the Panel from the energy flow perspective. The other strong aspect I want to address that acts as a drive for innovation in electronics is a combination of technological and economic factors, which is closely related to survival, both in the sense of longevity of a particular system as well as survival of design patterns and IPs as a longevity of the system as a kind or as a system design process.

My main tenets in this discussion are:

  • Compute where energy naturally flows.
  • Evolve (IPs, Designs) where biology (or nature as a whole) would evolve its parts (DNA, cells, cellular networks, organs).

I will also pose as one of the biggest challenges for semiconductor system the challenge of massive informational connectivity of parts at all levels of hierarchy, this is something that I hypothesize can only be addressed in hybrid cell-microelectronic systems. Information (and hence, data processing) flows should be commensurate to energy flows, only then we will be close to thermodynamic limits.

Alex Yakovlev

11.08.2016

 

My Keynote “Putting Computing on a Strict Diet with Energy-Proportionality”

I gave a keynote talk on “Putting Computing on a Strict Diet with Energy-Proportionality” at  the XXIX Conference on Design of Circuits and Integrated Systems, held in Madrid on 26-28th November 2014.

The abstract of the talk can be found in the conference programme:

http://www.cei.upm.es/dcis/wp-content/uploads/2014/10/DCIS_2014_program.pdf

The slides of the talk can be found here:

http://async.org.uk/Alex.Yakovlev/Yakovlev-DCIS2014-Keynote-final.pdf

 

Eliminating “competitors” by not giving them enough energy

One of possible strategies for differentiating some types of electronics from other types is to stage a “power-modulated competition” between them, by gradually tuning power source in different ways, for example in terms of power levels, either through voltage level or/and current level, also in dynamic sense as well. The circuits that require stable and sufficiently high level of voltage will be gradually eliminated from the race … Only those who can survive through the power dynamic range context will pass through the natural selection!

Building such a test bed is an interesting challenge by itself!

Technical report about survival instincts in electronic systems

 

http://async.org.uk/tech-reports/NCL-EEE-MSD-TR-2013-181.pdf

I wrote this article as a chapter to Peter Cheung’s 60th birthday Festschrift.

Here is the abstract:

The writing of this paper has been inspired by the motivating ideas of
incorporating self-awareness into systems that have been studied by
Prof Cheung in connection to dealing with variability and ageing in
nano-scale electronics. We attempt here to exploit the opportunities for
making systems self-aware, and taking it further, see them in a
biological perspective of survival under harsh operating conditions.
Survivability is developed here in the context of the availability of
energy and power, where the notion of power-modulation will navigate
us towards the incorporation into system design of the mechanisms
analogous to instincts in human brain. These mechanisms are
considered here through a set of novel techniques for reference-free
sensing and elastic memory for data retention. This is only a beginning
in the exploration of system design for survival, and many other
developments such as design of self-aware communication fabric are
further on the way.

Asynchronous Static RAM demo video

It is now possible to see the video of our demo of the Self-timed SRAM, as it works (Write and Read) under a wide range of power supply conditions:

(1) stable levels of Vdd in the range from 1.2V down to 0.4V; and

(2) with a run-time varying supply from our Capacitor-Bank power supply (second box in the setup).

 

towards survival instincts in computing systems

I have recently talked about developing survival instincts in computing systems. This opens up an interesting paradigm for designing autonomous systems for applications that require them to be on earth, underwater and in space. The conditions for operation of such systems are often harsh, unpredictable and it seems most natural to look for analogies to envisage the ways of their design in the nature, in animals and humans, particularly looking at the nervous systems. Another important pathway to such systems would be to look how energy affects their behaviour and how power levels activate various layers of instinct mechanisms …

These were the ideas that I discussed in my keynote talk at NoCArc’12 in Vancouver  (http://www.unikore.it/nocarc/index.html).

Here are the slides http://www.unikore.it/nocarc/slides/yakovlev.pdfand and video http://www.youtube.com/watch?v=lgcugX44EIg&feature=youtu.befrom