Skip to main content

Toolset for development of high performance and energy-efficient software, realising the potential of ubiquitous computing and the Internet of Things

Periodic Reporting for period 1 - GreenSoft (Toolset for development of high performance and energy-efficient software, realising the potential of ubiquitous computing and the Internet of Things)

Reporting period: 2017-05-01 to 2017-10-31

The huge, emerging market that is referred to by terms such as ‘ubiquitous computing’ and the ‘Internet of Things’ involves not mere hundreds of millions, but tens to hundreds of billions of processors. When planning to manufacture ten to the tenth or eleventh power of items, cost
rules; when planning to deploy them, energy efficiency and maintenance cost rule. Conventional processors, running legacy operating systems and legacy software, were never designed to address these three crucial considerations.

We believe that the wisest way to ameliorate this problem is to abandon the legacy software that has been driving conventional computer architectures and address these demanding problems in a fresh, simple way. The GreenSoft project targets commercialisation of a novel,
highly productive and energy-efficient programming framework and software toolset to synergetically support use of state of the art low-energy processors, thereby realising the potential for over 1,000-fold reduction in energy consumption.

Although the GreenSoft software development methodology and toolset can be applied for development of improved software for use with any processor, it is proposed that the first application will be orientated around GreenArrays multi-core processors.

The GreenArrays GA144 is a recent example of a low-power spatial processor, composed of many small, simple, identical cores. Among the many challenges of programming the GA144, programmes must be meticulously partitioned and laid out onto the physical cores. A spatial
architecture is an architecture for which the user or the compiler must assign data, computations, and communication primitives explicitly to its specific hardware resources such as computing units, storage, and an interconnect network. Future low-power processors will likely be spatial with simple interconnects between resources or cores, and have radically different instruction set architectures (ISAs) from what we commonly use today. They will likely be minimalistic, as for the GA144, providing little programmability support and so placing a greater burden on programmers and compilers.

GreenArrays chips offer the possibility for unprecedented control over power consumption in both quantity (many small computers, each consuming power only when necessary) and in time (ability to stop and start consuming power in picoseconds at the level of a
single core).

These chips are state of the art in respect of performance vs. energy consumption yet require special techniques for programming to realise the benefits, indeed to enable their practical usage. Consequently, they represent an ideal focus for first application of the developed toolset.

Within this SME Instrument Phase 1 project, a feasibility study has been undertaken to evaluate and analyse the potential of the innovation. This has been based on extensive investigation to support the process of decision making and define the business plan for reaching the targeted goals.

The overall conclusion from this study is that it is highly worthwhile to bring GreenSoft to market. The next step in this process will be to submit an SME Instrument programme Phase 2 application, which will be based on the positive results of the feasibility analysis and
business plan resulting from this study.
The work of this SME Instrument Phase 1 project has involved eight tasks, the results being briefly summarised as follows.

Task 1 of the study involved development of a detailed business model (BM) for exploitation of the GreenSoft results based on the approach proposed by Osterwalder. This approach encompasses work of previous authors by identifying nine basic model building blocks that,
considered together, can adequately describe the main characteristics of a business model.

These identify
• Key partners,
• Key activities,
• The Value Proposition,
• Customer Relationships,
• Customer Segments,
• Key Resources,
• Channels,
• Cost Structure, and
• Revenue Streams.

From this starting point, a SWOT analysis was then undertaken to identify
• Strengths (characteristics of the business or project that give it an advantage over others),
• Weaknesses (characteristics that place the business or project at a disadvantage relative to others),
• Opportunities (elements that the business or project can exploit to its advantage), and
• Threats (elements in the environment that could cause trouble for the business or project).

Within Task 2 of the study, market opportunities and user requirements for GreenSoft have been evaluated in detail, based on literature and patent searches, personal interviews and questionnaires involving a broad range of stakeholders. These have included companies involved
in the design of electronic systems, together with manufacturers of electronic products covering a broad spectrum of applications.

Using the identified user requirements from Task 2 of the study as an input, the work of Task 3 has identified requirements and specifications for development of the partitioning, layout and routing, code separation, and code generation tools. The feasibility of implementation of
each aspect has been considered in relation to the associated effort and cost to adapt results of previous work and develop the overall commercialised solution. These activities have resulted in definition of the detailed programme of work and estimated budget for the Phase 2 project.

Within Task 4 of the study, a large number of organisations have been consulted, both in relation to the market study and with the objective of identifying the most appropriate partners for collaboration in the Phase 2 project as the system is enhanced and brought to market.

Within Task 5 of the study, identified revenue streams, together with associated projected costs have been integrated. This process has enabled financial ratios and investment indicators to be quantified. The results have been used as an input to Tasks 7 and 8 to define the
comprehensive business plan including feasibility analysis.

Based on consideration of alternative protection mechanisms in Task 6 of the study, a strategy has been developed for protection of the GREENSOFT Intellectual Property (IP), including that expected to result from the Phase 2 project.

Integration of the preceding tasks within tasks 7 and 8 of the study has resulted in a feasibility report detailing the methodology used, the evaluation criteria, the study findings and recommendations. The market, technical, business, and economic and financial model viabilities
have each been critically appraised in the study process.
Progress beyond the state of the art
It should be appreciated that this is a feasibility study, not an R&D or technology development project per se. Consequently, the current study, in itself, did not have the primary objective of advancing the state of the art. Nevertheless, aspects of the work have related to advancement of the state of the art. As part of Task 3, the programme for enhancement and refinement of the GreenSoft toolset was developed, with work being undertaken to establish the technical feasibility.

Socio-economic impacts
Also referred to as pervasive computing, Internet of Things, etc., such computing is generally characterised by small size, low cost and low power. For example, ‘smart dust’/ networks of sensors for environmental monitoring, etc., generally involve energy harvesting and have
critical requirements for low power consumption. GreenSoft provides an effective solution to address these requirements and realise the latent potential.