ANU Green ICT Seminar Series 2009

This is to invite nominations for presentations in the ANU Green ICT Seminar Series. A six month program of weekly seminars is to be held on Wednesdays 2 to 3pm at the Australian National University in Canberra, as part of the Masters course "Green ICT Strategies" (COMP7310) . The weekly topics for the seminars have been chosen to match what the students are learning that week. Members of the ICT and related professions are invited to nominate to give a presentation, or simply to attend and discuss the issues with the students. There is no charge for attendance at the seminars, however seating priority will be given to the students enrolled in the course.

22 July: Understanding climate science

Understand environmental, social and business context for sustainability, and overview of background, boundaries.

29 July: The Global ICT Footprint

Estimate the embodied carbon and the footprint from use of telecommunications, data centres and desktop PCs.

5 August: Enabling ICT

Investigate how ICT systems can reduce energy and materials use by improving the efficiency of business systems by replacing the movement of goods with information (dematerialisation), improve the efficiency of machines (smart motor systems), logistics, buildings and grids.

12 August: Energy saving - Data Centres and Client Equipment

Computers and telecommunications equipment contributes about 2% to greenhouse gas emissions. Look at how data centres and client equipment can be made more efficient.

19 August: Materials Use

Energy reduction is only part of making a Green ICT system, there is also the issue of use of materials and hazardous substances.

26 August: Methods and tools

Ensure that appropriate methods and tools for the planning, development, operation, management and maintenance of systems are adopted and used effectively throughout the organisation.

2 September: Business process improvement

Recommend alternative solutions which reduce environmental impact, assesses feasibility, and recommends new approaches. ICT has the potential to provide significant environmental improvements, by replacing energy and materials consuming processes with more efficient ICT ones. How do you analyse business processes to identify alternative solutions which reduce environmental impact, assesses feasibility, and recommends new approaches?

9 September: Improving Data Centre Energy Efficiency

Investigate how to implement and assess data centre efficiency.

16 September: Enterprise Architecture

The business of business is business, so any environmental goals have to fit into the systems capability strategy which meets the strategic requirements of the business. How do you incorporate Green ICT into the models and plans to drive forward the strategy, taking advantage of opportunities to improve business performance, as well as environmental benefits?

23 September: Procurement

Much of the environmental benefits come about by selecting the right products and services. How do you write ICT requirement documents to ensure that your hardware, software and services suppliers provide green products?


14 October: Energy Star Program and Quality Management

Ensure that the processes for producing a product or service sustainability will do so to a consistent standard. Investigate the fit between US EPA's Energy Star Program and those of Quality Management Systems.

21 October: Compliance audit

Assess the conformity of corporate systems to environmental standards, such as ISO 14000 series of environment management standards, Electronic Product Environmental Assessment Tool (EPEAT), and Energy Star Program.

Green Data Centre In Canberra

Yesterday I had a tour of Canberra Data Centres new facility in Canberra. This is a commercial data center set up to host government and non-government computer systems. It is located in a converted warehouse in an industrial park. Apart from the emphasis on green computing and security, what is most impressive about this facility is that it is open for business and is servicing clients already, unlike proposals such as that for Canberra Technology City (CTC) .

The facility uses APC’s "Hot Aisle" system. Two rows of computer racks are placed back to back with a polycarbonate roof and doors at either end enclosing the hot air. Coolers are placed at intervals in the racks, drawing the hot air, cooling it and supply it to the front of the racks. The coolers are supplied with chilled water from a central plant. The result is that the cooling is supplied to where it is needed, making the system more efficient and more flexible.

The APC system has all power, data and the cooling supplied from above. There is no need for a false floor. Pods can be devoted to a particular client and even isolated with a wall where security requires. Smaller clients rent racks in a shared pod.

The central chiller plant has multiple units and an insulated tank to hold a supply of cold water. This allows the load on the chillers to be balanced and a backup supply of chilled water if the units have to be shut down (or mains power is lost).

The APC pods have battery backup, to keep servers running until the multiple diesel generators start to supply power. But this full backup power is expensive. I suggested to CDC that there would be scope for using the resiliency features of the web and the power saving of modern servers to provide clients with a lower cost option. Web servers could be programmed to reduce their power use during a mains failure, by lowering their serving rate. The customers using this option could be charged a lower rental rate, as they would be making less use of the backup power. Those customers with an alternate server at another location could rely on that server taking the load. Otherwise, a well designed web application would automatically provide the essential information (such as the text) and delay delivering non-essential information (such as graphics).

The CDC facility provides an alternative to companies and government agencies building their own facilities. All but the largest agencies would have difficulty meeting the stringent requirements for such facilities and the increasingly stringent additional environmental requirements. The power to the CDC's pods is separately metered, allowing the customers to each be charged for electricity used (as well as for cooling). This would assist with carbon emission reporting and also to show the power savings.

Government agencies can rationalise their computing equipment by placing it in such a facility. But it would be unfortunate of they simply took a lot of old inefficient equipment and put it in a new centre. Agencies need to look at rationalising the number of servers they use and the efficiency of their existing equipment.

A more difficult task would then be for the Government to rationalise IT use between agencies. There seems no good reason why dozens of agencies run their own web servers, records management, financial and human resource systems. As these systems become server applications with web interfaces and with web user interfaces, there is increased scope for rationalisation. If government systems were rationalised in this way only a few data centres the size of the CDC facility would be needed to service all of the federal government's requirements.

Grow tomatoes at Canberra Data Centre?

A greenhouse is using the exhaust from an electricity generator to help grow tomatoes and reduce carbon emissions. Perhaps this could be used at the proposed data centre in Canberra to reduce emissions and objections. There may be some objection to eating plants grown in engine exhaust, but it could be used for a plant crop which is not eaten:

... Last month, two of GE’s 4MW, natural gas-fuelled J624 GS engines began powering Royal Pride Holland’s new greenhouse cogeneration plant. ...

exhaust ... gas to be recycled in the greenhouse as a special fertiliser to help boost crop production instead of venting the gas into the atmosphere. In addition to supporting the greenhouse’s operations, surplus electricity from the cogeneration plant is being delivered to the local grid...

From: Netherlands Commercial Tomato Greenhouse Boosts Production with Cogeneration Plant Powered by World’s First 24-Cylinder Gas Engines, GE, 02 October 2008

Data centre and solar power station for Canberra

A consortium of Sanctuary Energy, Wizard Power and Springfield Land Corporation has proposed a $220 million data centre and solar thermal power station for Canberra. This is separate to the Canberra Technology City (CTC) consortium of ActewAGL, Technical Real Estate, Galileo Connect and CB Richard Ellis proposal for a data centre and gas fired power station.

Previously I criticised the CTC proposal (mostly for the way it was presented) and suggested a data center with roof mounted solar power station. The new consortium didn't tkae up my idea of putting the power station on the roof (they propose to use a separate site), but are proposing to use ANU's thermal solar technology.

There is also a suggestion to site a data centre at Cooma, near Canberra. Of course, provided the site is in Australia, so that Australian law applies to it, and it has access top a fibre optic backbone, it does not matter how far the site is from Canberra. The new Mach Technology Data Centre just opened at Noosa in Queensland would do just as well (it has a fibre optic link to Brisbane).

One a handful of staff is needed at a data centre. The users of the computers in the centre never need visit it (in almost three decades in the ICT industry I have only been into a large computer room a handful of times). It would be preferable for security reasons to ban visitors and keep the exact location secret.

Combined data center and solar power station for Canberra

A study commissioned by the ACT Government and ActewAGL has found that a large-scale solar power plant is feasible in Canberra: "Solar Power Plant Pre-feasibility Study". If re-elected, the government will seek expressions of interest to build the plant. As suggested to the Chinese government in 2003, I propose the plant be built on the roofs of new data centres and warehouses in industrial parks. This will allow the land under the plant to be put to use, provide power for the data center and industrial buildings. It would also overcome environmental problems with the previously proposed data centre for Canberra.

Building Design

The complex would use a standard industrial building design, with a roof strengthened to hold the solar collectors. Most buildings would be initially used as low cost warehouses and only fited out as data centers when required. The data centers would use a proven, simplified modular design for equipment, which would be placed on a simple sealed concrete floor. The height of the warehouse building would be used to reduce air conditioning requirements with proven low cost techniques.

Data Center Customers

Scientific customers, such as ANU's supercomputer center and government agencies, could be large scale tenants to anchor the development. Some ancillary office space could also be provided.

Twinning

To allow for backup data supplies and exploit the daily solar cycle, the center could be twinned with centers in other time zones. While concentrator solar power stations can store energy, there is still a peak power supply during the day. This peak power could be most efficiently utilised by the data centre under the collectors. A center in another time zone, such as near Perth, would have a peak at a different time of day and could take over some of the processing load from Canberra, as well as providing a backup.

Sighting

The feasibility study raised the issue of the visual intrusiveness of a large field of solar collectors. If placed on top of a warehouse, in industrial parks, the collectors would be less visible. The industrial park could be designed to high environmental standards, retaining natural vegetation around the buildings to soften the visual impact, with the buildings coloured to blend in. In addition the opportunity could be taken to collect high quality water from the complex, for use in Canberra.

Solar Power Plant Pre-feasibility Study
Executive summary

This Solar Power Plant Pre-feasibility Study was undertaken for ActewAGL and the ACT Government (the joint parties) by PB. Its purpose was to investigate solar power generation technologies, identify an appropriate solar technology for the ACT, and establish the economic viability of a solar power facility.

Technology for producing electricity from solar energy is technically proven for both PV and solar thermal technologies. 354 MW solar thermal plants, using trough technology, have been operating in the USA since the 1980s and new plants of this type (between 50 MW and 70 MW) are now coming into service in the USA and Europe. Other solar thermal technologies that are not yet in commercial use are power towers, paraboidal dishes and Fresnel systems. Large multi-megawatt PV plants, to approximately 50 MW, are now in operation. Solar technology is expensive, and significant financial assistance from government is available to the developers and operators of new plants. There is significant local community and market support for solar power generation.

This study identifies a 22 MW project that uses solar thermal trough technology, similar to new overseas plants, as the best option for the ACT. This technology has been chosen because of its substantial operational record (more than 20 years), lower cost compared to other solar technologies, and use in new commercial plants in the USA and Europe.

The plant will produce enough electricity for approximately 10,000 Canberra homes and the project cost, before government assistance, is estimated at $141 million (including land and infrastructure). A site of 120 ha will be required and if engineering, planning and environmental work commenced immediately, it is envisaged that a plant could be
commissioned by 2012.

An alternative option is a large PV cell-based plant. To produce the same amount of electricity (that is, to service 10,000 homes), 75 ha of land would be required and the plant would have an electrical capacity
of 57 MW. This would be one of the largest PV plants in the world but the risks would be lower than the solar thermal plant, reflecting the more mature status of PV technology, its predictable performance and
cost. However, the total project cost of $424 million is high.
It is recommended that this pre-feasibility study be followed by a feasibility study that includes engineering studies, ongoing commercial evaluation, financial modelling and environmental and planning studies.

A staged study, extending over eighteen months, could be conducted and lead directly into procurement and construction. However, trough technology is not cost effective for a staged development at the size
of the proposed ACT plant. Even though the solar field is modular, the balance of the plant is not suitable for staged development without incurring significant additional costs. A financial evaluation of the solar thermal project, assuming 100% equity funding, a 9.5% Weighted
Average Cost of Capital (WACC) and a 20-year project life was undertaken, Key results were:

• a levelised electricity cost of $106/MWh for a net project cost of $47 million. This is for a plant cost of $2,500/kW, which is forecast for the technology in Australa, and allows grant funding of 50% of the project capital cost;
• the relatively high cost of generation is due to the high capital cost of plant itself, the high proportion of infrastructure and land (38% of project cost) and the relatively low productivity (measured by the 42% capacity factor).
• larger plant size would significantly improve the economics by spreading the infrastructure costs over a larger productive plant and capturing economies of scale of the production plant itself. For example, doubling the plant to 44 MW would lower electricity cost by about 25%;
• 57% grant funding was required to reduce the levelised electricity cost to $95/MWh which is the expected Power Purchase Agreement (PPA) electricity selling price;
• higher solar radiation levels such as at Mildura would lower levelised electricity cost by about $50/MWh, or 17% (before rebates); and

Government grants and subsidies have been fundamental to the facilitation of the growth of solar energy generation around the world. The requirement for government support also applies to this project. This
project would appear to fit well with current Australian and ACT Government policies (such as the move toward zero/low carbon emissions and renewable generation) and it supports ActewAGL regulatory
requirements for renewable energy.

The Sun is a reliable but intermittent and diffuse source of energy. There is strong daily and seasonal variation and availability, and it may be limited by cloud cover. To extend power generation beyond periods of sunlight and to allow a steady supply of heat, two approaches to solar thermal plant energy storage were proposed:

• storage of heat at the plant and use of this heat when direct sunlight is not available. This would give an extra four to six hours operation without the Sun shining; and
• use of natural gas as an auxiliary fuel to supply heat as an alternative. If this is supplied by the waste heat from a cogeneration plant, an additional 47 MW could be generated by a gas turbine. The use of gas auxiliary fuel does not affect the eligibility of solar generation as renewable or green energy under the current regulatory arrangements, but may have some impact on community perceptions.

The solar thermal plant would occupy a significant area and unless it is well-shielded, it is likely to be a prominent visual feature. It would combine the physical features of the large solar field with a small
thermal power station, possibly with a gas boiler or small gas turbine for back-up. While the solar technology itself is considered to be relatively benign, it is likely to require consideration environmental issues, that are similar to those raised by a small gas-fired power station with the additional issues raised by the large land area and visual amenity.
Formal evaluations of potential sites for the solar facility will occur only if the project is found to be viable and progresses to a more detailed study, at which time such sites would undergo a rigorous environmental and planning assessment.

From: Executive Summary, "Solar Power Plant Pre-feasibility Study", Parsons Brinckerhoff Australia, (PB 2158583A-RPT001-Qpc, 2 September 2008

Opposition to gas fired power station in Canberra

Efforts to build a data center with co-generation in Canberra have met with local opposition. The Garnaut Review interim report proposes a carbon trading scheme. The effect of this will be to increase the cost of electricity compared to that from gas (which is less polluting). But this is not as easy as it looks. As an example the proposal for a gas fired co-generation plant for the "Canberra Technolofy City" data center in Canberra is meeting community opposition. The plant will be much cleaner than a coal fired one, but the residents don't want it within a few tens of kilometers of them (the coal fired plants are hundred of kilometers away).

The opponents remapped the data provided by the plan's proposers to emphasize how far pollution would spread. This is a sensitive issue as Canberra is subject to an inversion layer in winter, trapping pollution. However, a new gas fired power station is going to produce very little pollution. Perhaps its output should measured in FWDE: "four wheel drive equivalents". ;-)

The proposers of the plant brought some of the problems on themselves by calling what they were building a "power station" and emphasized its capacity. If this was described as just a way to supplement the data center's power it is unlikely anyone would have noticed.

The community consultation documents show the data center "ground view-from Macarthur", of a starkly visible white building with four towers which look like smoke stacks, rising out of the countryside. If the building was painted grey-green and tan to blend into the landscape it would be far less objectionable. This would not increase the cost of the plant, the steel panels of the building could simply use the standard palate of Colorbond colours developed to developed to blend into the Australian landscape. For increased blending, the rectangular building panels could be of subtlety different colors, forming digital comflage and making the building difficult to see from more than a few hundred metres away.

If the stacks were in low visibility colors and below the ridge line of the surrounding hills, it would not look so much like a power station. Gas fired generators are very small units, and the one for the data center probably occupies about the space of a double garage. The buildings show in the plans will be the offices and the cooling plant for the data center, not the power station.

Canberra Technology City Proposed

Yesterday a "next generation data centre campus" was proposed for Canberra. This is an initiative of the local energy company ActewAGL. They previously set up the Transact fibre optic network around the city and so have a track record in large scale IT projects.

... The Canberra Technology City (CTC) development would revolutionise the way that computer systems and data is secured and supported and provide a comprehensive long-term solution for Australian and international organisations.

A consortium, including ActewAGL, Technical Real Estate (developer and owner of the data centres), Galileo Connect a UK-based world leader in engineering and designer of future proofed data centres and CB Richard Ellis (global advisory property services), has combined their capabilities to present CTC. ...

The CTC project would provide for data centres over two separate sites in the ACT. The first site at Hume is intended to service primary production demand suitable as a regional data centre ‘hub’ for large global companies operating within the Asia-Pacific region. The second site at Belconnen is intended to service secondary production demands, as well as to provide backup and recovery and business continuity for the primary site tenants.

The CTC development would be an environmentally friendly data centre campus with onsite power generation. In recognition of C02 emissions and global warming the CTC data centres will lead the industry in reducing the carbon footprint of such facilities.

Invitations to CTC briefings have been distributed to 300 companies across Australia and internationally. The Canberra briefing will be held on 15 October 2007. ...

From: ActewAGL investigates next generation data centre development, Media Release, ActewAGL, 2 October 2007

ActewAGL is emphasizing the power requirements of data centers. But the data center will still use a lot of power.

... The consortium, which includes ActewAGL and hi-tech property developer Technical Real Estate, plans to build about 130,000sqm of data centre space at two sites in the suburbs of Hume and Belconnen.

The company said it expected to complete the first data centre by September 2008 with the remaining sites to become operational over the next five to 10 years.

On completion the combined facility will comprise about 30 data centre units.

TRE director Bruce McEwen said research by power equipment suppliers showed that the world, and particularly the Asia-Pacific region, would soon face a shortage of data centres with adequate power to serve current computing technology.

... "Customers are buying technology that's calling for 15kW and 20kW per rack and you only have to go back five years ago, to find they were 1.5 to 2kW per rack," ...

Utility provider ActewAGL has also agreed to build facilities to power the site using gas piped from Bass Strait if power loads at the facility reach 40MW.

Using gas is expected to reduce its carbon emissions by a further 10 to 15 per cent.

From: Canberra a data hub target, Andrew Colley, Australian IT, October 02, 2007

Power consumption of 20kW per rack works out to be about 6kW per square metre. So the 130,000sqm data center would use 780 MW of power and cause about 300,000 tonnes of carbon dioxide greenhouse gas per year. But there are advantages to dedicated data parks, as I pointed out to the Chinese government in 2003. Google have placed solar cells on the roof of their Mountain View campus.

But before leasing a large new data center, organizations should consider making their systems more efficient. As an example, one government agency recently sent me a mail message which was 100 times larger than it needed to be. If that is reflected in overall government ICT use, they could reduce a data center from the size of a football field (about 6,000 square metres) to the size of a cricket pitch (about 60 square meters) by more efficient implementation. This would also save about 35 MW of power for such a center.

I might see if I can put a bit more in my course on e-documents for public servants. More efficient documents might save a few hundred thousand tones of pollution each year.