Friday 21 December 2012

Shopping in Canary Wharf

I called in to the shopping mall under Canary Wharf in London Docklands today to buy some Christmas presents.  I didn't last long because the sight of my gray hair triggered a CCTV  intruder alarm.

Tuesday 11 December 2012

Who called your mobile phone?

Here's a useful service for tracking and recording unwanted calls to your mobile phone.

Monday 19 November 2012

The Shard London

I've just been reading the specifications for the Shard building and as a designer of trading areas, I groaned a quick "Oh no!"

This iconic skyscraper is built just a stone's throw from the edge of the City of London in the borough of Southwark at the [wrong] south end of London Bridge. It is five minutes walk from the financial services centre of the City of London. The tower also has almost instant access to the London Underground Jubilee line which gives fast (seven minutes) and easy access to the finance industry in Canary Wharf. The location is well served by resilient optical network backbone services. You'd think it to be ideal for the financial services industry.

However when you read the specifications on their public web site that picture changes. The raised floor gap is only 150 mm (6 inches). That can be really restrictive in delivering power and data cabling to a trading desk. With a 2.7 metre floor to ceiling height there is little scope for changing the raised floor height. Any full size equipment cabinets will be quite cramped.

The floor loading is only  3.5 kN/M2 (73 pounds/sq foot) with up to up to 7.5kN/M2 in parts. This essentially dictates only office furniture except in a few small strong points. I've yet to see the detailed floor plans but I'd guess the strong points are not conveniently located. From their web site I can see there's also two sets of internal structural building pillars within the office areas. The pillars will restrict office/trading floor layouts.

The Standby power generators are 2 x 2000 KVA; this represents only 70 KVA/M2 when both generators are running for the office space, provided there is no draw from the hotel sections of the building or the accommodation sections. As a rule of thumb I'd say about 15 - 20 Watts of available protected power per Sq Metre of office floor, ignoring any concentrated power draw of a technology/server room.

It strikes me the space is suitable for low intensity office occupation and not the technology intensive trading floors of the financial sector. The building specification must limit the attractiveness to banking clients currently on the north side of the River Thames. When will these property developers learn to provide accommodation with reasonable infrastructure services?

Wednesday 14 November 2012

New York Outages

New York and New Jersey suffered a lot of damage from Hurricane Sandy. Even now a couple of weeks after the impact of the damage is still effecting  businesses. Apart from the obvious damage to timber frame buildings, the main damage effects appears to be:

  • Loss of power
  • Lack of fuel
  • Results of flooding.
While I have enormous sympathy for those involved I can't but help think many of the after effects of this storm could have been mitigated or avoided if there had been some proper planning and infrastructure provision beforehand. Much of the problems seem to arise from trying to provide infrastructure on the cheap. Saving pennies has cost many more dollars. At some point in history an accountant may have shaved a few dollars from the budget for a company's infrastructure, but that same accountant will not stand up and proclaim "Mea Culpa" when the business is out of action for a couple of weeks as a consequence of those "savings". There are those who say this could be covered by business continuity insurance because it is only a 100 Year Storm. You'd trust the insurers to cover all costs and rebuild your company? Not!

Let's first look at fuel. In the case of Gasoline there were service stations which could not pump fuel because the pumps depended on mains electricity from the Utility Company. When the power went down the service stations became useless. There should be federal laws which require the provision of  weather protected standby generators at service stations to provide protected power to pump the fuel.

Then there was the problem of fuel shortages. The supply of gasoline in normal circumstances is a "just-in-time" facility where the service stations depend on regular delivery of fuel every few days by tanker trucks. If that delivery service is interrupted the service stations hold only a couple of days supply and quickly dry up. The situation is worsened by panic buying by consumers. When everyone decides to fill their gas tank to the top it places strains on the limited supplies causing them to run out more quickly. It leads to rationing such as this case. This whole situation could be mitigated if service stations were forced by law to hold a minimum of 15 days stock in their underground tanks. The 15 day buffer would make fuel supplies considerably more resilient to supply shock. The finance cost of holding 12 extra day's supply is not massive.

Your organisation may have standby power generators with a 24 hour fuel supply with a resupply contract with a local fuel depot. Have you actually checked whether the fuel supplier can honour those commitments in the event of a regional disaster such as Hurricane Sandy? Flooding might prevent despatch of the trucks. I recently saw a comment on a LinkedIn forum where an organisation fuel resupply company could not pump fuel into the delivery tankers because they also had lost mains power from the utility company. You should diversify suppliers and actually check their delivery arrangements in the event of a regional disaster, don't rely on contractual obligations. If you can extend your own fuel capacity to five days you'll give yourself more flexibility in the event of a disaster.

The regional loss of power is almost inevitable in the event of a disaster such as Hurricane Sandy. Power lines will collapse and the utility company will cut power to protect substations. You should ensure access to local power generation facilities. Perhaps your own standby generator or maybe a facility shared between local businesses. The standby facility should be regularly tested with total power down scenario's. You should walk through what would happen if the basement of your building becomes flooded. A co-location hosting centre discovered this to their cost. In some cases heroic efforts such as a human bucket chain may be necessary to keep the generators running.

Lloyds - the insurance people; have written a telling paper on energy security paper. It is worth visiting their site to download a copy to read.

Flooding can strike most businesses. You may be in a low lying coastal area, a river flood plain. You may have a major water main pipe running under the street outside of your front door with the risk of pipe burst. In storm conditions damaged roofing can allow rain water leakage. If there is risk you should look at: 
  • the arrangements you have to prevent the water from reaching your key machinery; this might include high sill doorways, sealing of leakage points, tanking of the room floor and walls.
  • the arrangements you have to quickly drain any flooding in your building. It might include pre-laid pipes, drainage sumps in equipment rooms and submersible drainage pumps. 
  • leak detection equipment and alerts;
  • overhead drip trays which can help prevent overhead water from leaking on to sensitive equipment. 
They plan for these type of flooding incidents in Australia. In New York a substantial about of phone/data cabling was destroyed by salt water flooding in a Verizon below ground level cable vault (90,000 cubic feet). The compressed air pumps, which help to keep the copper data cables dry, failed allowing salt water to saturate the insulation and destroy the data communications. Most of the old cable will have to be chopped away and replaced with optical fibre.

Saturday 20 October 2012

Blast mitigation

One development in physical security is the use of spray on polymer coatings to mitigate the effect of explosion blast on building walls. This can be a retrofit option is required. It helps to reduce penetration and spalling on the inside of the wall.

Here's a couple of suppliers Rhino LiningsDynasystems (Dynashield) and Line-X (Paxcon) and a video example. The polymer was originally developed as a coating to protect the the floor beds of pickup trucks from damage. It doesn't greatly strengthen the concrete/blocks/steel per se, but it does reduce the debris on the side of the wall opposite to the explosion. The reduces the risk to people within the building.

If you need some comparative research, there's a paper here.

If you are in the position to influence the construction of the building ,the blast resistance can be enhanced by adding polymer fibres and steel fibre (approx 7%) into the concrete during the mixing process. There is a great range of suppliers for these fibres. Such technology can also help protect concrete from the spalling effects of fire.

Friday 12 October 2012

Passive fire protection

This week I called in at the Facilities Management exhibition in Olympia Conference Centre in London UK.  Mostly I find such exhibitions a bit boring, but there are often useful nuggets of information to be discovered. 

One nuggets of these was the Promat stand, There I was shown products which can provide four hour fire protection and also limited blast protection too. If this topic interests you it is well worth visiting their website to browse for 30 minutes or so. They have some particularly useful handbooks available on-line free of charge. If you need to plan building firewalls or "fire stopping", this is the site to visit.

Tuesday 2 October 2012

Who are the software pirates?

I'm pretty sure that if a software manufacturer discovered a large organisation where software piracy is rampant there would soon be a court action and a claim for damages. For example if one in five of the users had an illegal copy of their software running on their desktop PCs the supplier would demand payment and compensation on top.

However several of the USA based suppliers are guilty of that same piracy when it comes to the differential between the USA price (USD) and the UK price (GBP) charged to customers. Take for example the latest release of the Adobe Acrobat product version XI. Its USA advertised price is $299, yet the UK price is the equivalent of $375 after the effects of sales tax (VAT) and exchange differences are taken into account.  It makes the same product bought in the UK (by download) 25% more expensive. There is no justification for this corporate greed by Adobe, it is in effect piracy by the suppliers.

I don't break the law by using pirated copies of Adobe software. I simply buy an alternative called Nitro PDF. It is just as good as Adobe Acrobat, yet it is one third of the price and they don't play games with the price in different currencies. Hence Adobe loses out. If they'd shown some integrity I might have done business with them. It could be worth checking what your own business policy is in regard to USA/UK pricing differential.

Wednesday 19 September 2012

Cheap Energy?

An article in Wired points to progress in cold fusion. Maybe it works or maybe it is like write only memory? It will be interesting to see. Here's another information source.

Friday 31 August 2012

Dubious business practices.

Over the years I've developed the view that common courtesy is an essential part of client relationships. Part of the process is to show respect to clients/prospects. Wasting the time of a client/prospect shows a lack of respect.  When I see those signs of discourtesy I tend to assign the supplier to the blacklist/scam pile and refuse to do business with them in the future.

Take for example Clifton Asset Management Plc based in Bristol UK. They have the practice of periodically sending me prospecting letters offering Overdraft and Loan facilities for my business. As it happens we do not run our business using the debt model and we have no interest in such offers. What really annoys me is that the signed letters from Adam Tavener Chairman of Clifton Asset Management Plc are marked "Private and Confidential" on the outside of the envelope. The contents are clearly neither private nor confidential. Mr Tavener has no prior business relationship with me or my company. The wrongful labelling of the envelope is a deceitful practice designed to help the letter bypass any secretarial filter.

Such deceit does not bode well for any future business relations with this company so they are now blacklisted. 

However their letters continue to arrive. There is a note at the foot of the letters saying please contact Clifton "should you wish to unsubscribe"; yet another clear indicator of unsolicited mail as we didn't subscribe in the first place.  We've contacted them on previous occasions to ask them to stop sending such letters to our organisation to no avail.  

Earlier this week I phoned Clifton Asset Management to speak with Mr Adam Tavener as invited to do so by his letter. The call was answered by a telephone clerk. I made it clear I  wanted to discuss the dubious marketing practices with Mr Tavener, but I was told he's a busy man and would not be available.  I left a message for him to call me. I'm not surprised to have not been called by Mr Tavener. Ignoring my complaint is a further sign of disrespect.

Since then I've undertaken some research into Clifton Asset Management,  the associated companies, the management team and their history. They are certainly not the type of company that I'd consider using in any shape or form in the future.  It just goes to show that a lack of courtesy is a reliable business indicator. I have no evidence that their services are a scam, indeed they are registered and regulated by the Financial Services Authority, but it would be a cold day in Hell before I'd consider using them.

It seems like I'm not alone in these thoughts, and here.

Edit: 14/11/2013 We just can't seem to get away from these people. We've moved to a new location and are receiving mail for the previous occupant including letters marked "private and confidential". The return address on those letters is  TMS PO Box 10044, Hucknall NG15 5DQ which appears to trace back to Clifton Asset Management. We've marked them Return to Sender, but have no great hopes about the letters ceasing.

Edit: 31/01/2014 It seems like the Clifton Asset Management is still up to its old tricks. Despite requests from us to cease and desist sending us their junk mail we had another letter from them dated 27th January 2014. Marked "Private and Confidential" which it is not, the letter puts forward schemes to raid our pension funds by borrowing against them and along the way paying large fees to Clifton Asset Management. It seems a rather parasitic use of funds which should be held in trust for the fund members. This time however their website has changed to There appears to be a new name in the frame of Neil Greenaway as the Managing Director. I wonder what it will take for Clifton Asset Management to leave us in peace? 

Wednesday 8 August 2012

Planning for power (2)

In the previous blog entry on planning for power in a data centre I considered the need for a power budget and the need for a good change control process for variation in the power budget.
Most new data centres are initially only partially filled with equipment with space, facilities and resources left free for expansion. The power budget for the equipment expansion areas needs to be estimated in the early planning stages of the main construction project. It might be possible to postpone both capital and operational expenditure by fitting only the power infrastructure required at the moment, but in this case engineering space needs to designated and protected for future expansion. Additionally consideration needs to be given to the route through which electrical and mechanical plant will be moved during upgrades and maintenance.

When the power load has been established it will be possible complete the planning for air conditioning requirement.  The design of the air conditioning system will depend on what facilities exist for chilled water and what additional equipment can be installed within the building.

Consideration will needed on how long standby power will need to endure during a mains outage. This decision is not simply how long it takes for a standby generator takes to start and provide power. Consideration needs to be given as to how long it takes to gracefully shut down a data centre should the standby generators fail to start. An uncontrolled power crash can have devastating effects on ICT processing and the dependent businesses. The endurance requirements of the business will dictate how much energy needs to be stored in batteries. Consideration also needs to be given on how cooling will be provided if the generators fail to start. The required battery capacity affects the design of the battery room, but I'll cover that in a later blog article.

A joint decision will have to be taken, with the electrical engineer, on the power transfer switching arrangements will handover power to the standby arrangements and consequently back when Utility mains power is restored. For a data centre this switching is usually automatic with manual over-ride capability.

When the power requirement has been agreed, the engineers should establish the space requirements and weight of the air conditioning plant, machinery, transformers, switching gear, UPSs, generators, fuel tanks, batteries, ducting, piping and cabling  required. Agreement should be sought on where all of this equipment will be housed and mounted. The electrical equipment and air conditioning plant will impose an additional weight load on the building structure. The positioning of this weight/vibration load should be reviewed by a structural engineer to ensure the load is within the capacity of the building. Machinery such as generators, compressors and air handling units will create vibration in operation. This vibration might affect the building. The plans for damping and handling such vibrations should be reviewed by the building structural engineer.

Throughout the planning process it is highly likely that there will be obstacles to providing the equipment necessary to provide the power and cooling. For example there may not be enough space or the equipment is too heavy. In the worst case these obstructions may limit the amount of technology which could be deployed within a building. In any event there are likely to be many cycles of the planning process before an agreed solution can be achieved. Depending on the complexity of the data centre this process may take a few months. The process may involve agreement from the owners of the building and also regulatory planning authorities. If the original assumptions on power requirements change after the design has been settled it can be an expensive problem to fix.

Part of the planning process is to provide a solution which is affordable to the business client. It is in effect a process of risk assessment with the client where the proposed capital and operational costs are weighed against the risks and what the business can afford.

Monday 23 July 2012

Planning for power (1)

I've been reading recently on forums questions about how much power should be allowed per cabinet/rack in the Data Centre. Making a mistake with this figure can be expensive. If you calculate too low a power supply it can be a very expensive process to remedy the mistake. Retro fitting power upgrades on a live data centre is an expensive and complex process. It usually comes at the time when a business is not expecting a major capital outlay, not a move popular with the Board. I've seen one case where "glowing bus bars" unable to handle any more power led to an international bank having to hurriedly construct a new data centre in a new location. There was still free space in the old data centre just not enough power for any new equipment for business expansion.

Over-provision for power can also prove expensive in terms of wasted initial capital expenditure, increased maintenance costs and wasted energy during the lifetime of the equipment. 

It would take a book to fully describe the steps needed to calculate and install the right power system. The actual design and installation must be undertaken by qualified electrical engineers, but the management or planners of the data centre have information which is vital to the power system design process. Key to the process is an energy budget and forecast process. The team must be able to measure the usage of existing equipment, predict the actual power usage of  equipment which destined to be installed and also facilitate the calculation of power needed for supporting systems such as air conditioning.

The power budget should be an actual document/spreadsheet/database recording the planned/actual usage of equipment. It should also document the physical location of the equipment so it is possible to calculate how much power should be delivered to a specific location within the data centre. It is normal for power to be delivered over three phases (circuits) of electric power. The engineers will need to know how the power demand will be distributed over the three electrical phases. Cable routes and switching arrangements can be affected by the decision to provide dual power routes to cabinets/rack and their physical location.

Power planning and the physical location of usage will affect the design of the earth bonding reference grid. The sensitive equipment in the technology racks will need electrically clean bonding connections. Those connections will need to be sized suitably for the power draw in that area. The electrical engineers should be asking questions about these matters as part of their design process. Data Centre management should ensure the engineers take bonding requirement into account.

The power budget must also have a change control process to ensure any variations in planned/actual usage are accurately investigated, where necessary authorised and scheduled for implementation. In some cases DCIM software (Data Centre Information Management) can assist the power usage/commitment documentation process.

Power demand planning should anticipate how power requirements will develop over the lime time of the data centre. It may be appropriate to increase power provision as data centre space utilisation increases. For example power equipment which can be expanded by adding further modules may be chosen.

From an early stage, the power budget should incorporate cost recovery/allocation mechanisms. The cost recovery process should be transparent and acceptable to the client base. The process of cost recovery, may require accurate power metering down to the level of individual servers. This will affect the design and investment of the infrastructure. The budgetary responsibility may be in different organisational silos such as ICT and facilities management. Part of the provision may affect the charges levied by the building owner/landlord. Carbon Credit recovery may also affect the budgetary process.

The power budget needs to be subdivided into:
  • Critical power
  • Protected power
  • Unprotected power
Critical power is that which should be continuously available without interruption. Facilities should be provided to provide back-up power sources without interruption when the primary source fails.

Protected power is that which should be continuously available, but can sustain short duration outages  while a standby power generator starts up if the primary source fails.

Unprotected power is routinely supplied by the utility company and may be subject to supply outages of a local or regional basis. Even with unprotected power it is sensible for an organisation to increase protection by attempting to provide dual sources from different sub-stations.

Each of these separate power categories will have different capital and ongoing operational costs associated with them. It is no longer really acceptable to lump all power costs into one financial pot.

The power budget should also separately document in power in-rush when equipment is switched on. This figure can be substantially larger than the steady state power draw. While is does not normally affect the power supply to the building it will affect the design of the circuit breakers in the power circuitry and may also affect the specification of electrical back up systems such as generators and UPS battery systems.

Those people calculating the power requirements should not use the "plated" rating on the outside of the unit, but should undertake careful research with the manufacturer's specification as to steady state usage. This should be backed up with actual measurement of power usage on equipment in use in a production environment.


Wednesday 18 July 2012

Demand controlled air ventilation

When building a trading floor and its associated data centre/technology room the designer will need to think about how much fresh air should be pumped into the room. Depending on external climate conditions the delivery of fresh air will need some energy to process it before introducing the air into the building. It may need heating/cooling/dehumidification and it will definitely require filtering.

Keeping the fresh air to a sensible minimum in an air conditioned building will reduce energy costs. If the building has a lot of people present you'll need more fresh air. If the building contains only machinery there will not be a great demand for fresh air. Some basic building ventilation systems will be designed to deliver a certain percentage of fresh air to ensure X air  changes per hour. However in practice many buildings are occupied only 60% of the day. Some rooms such as conference rooms are often empty with little need for fresh air, but when fully occupied the demand for fresh air increases.

Demand air ventilation systems incorporate carbon dioxide sensors to detect whe the people load is increasing the need for fresh air. Such systems are usually capable of air conditioning of the fresh air delivered, so only the specifically required fresh air will need energy. It reduces waste and improves the environment for the building occupants. Here is one such system. Here's a useful (12mb pdf) document. from Carrier Corporation

Thursday 5 July 2012

Reselling software licences in EU

In an article by Stewart Mitchell in PC Pro, and also Cnet (Don Reisinger) it was reported that a recent EU court judgement allows businesses to resell software licensing blocks. The implication is your company can sell software licences which it no longer needs. You can't split the licence blocks into smaller portions, but if you've purchased several blocks for the same software those can be individually resold.

As this ruling was in Germany under EU law you might want to seek legal advice first in other countries before following this route. It will be interesting to see how this affects supplier buyback of licences. No doubt the suppliers will attempt to repackage their licences as services. One company UsedSoft already offers a service in this respect.

Monday 18 June 2012

Power Outage $30K per second

An article in the Register caught my eye. It gives a really graphic view of the potential costs of power failure. In this case it was an LCD screen manufacturing plant.

Thursday 7 June 2012

Just in Time fuel supply - is it worth it?

In a BBC report we learn that Manchester (UK) airport ran out of jet fuel. They hold between 12 and 24 hours supply, relying on a single 30 mile pipeline from a refinery to provide replenishment. Considering planes can't fly without fuel I'd call the fuel storage and supply an essential asset. Without the fuel the airport would have to close or operate on a considerably reduced  traffic. Not keeping several day's supply in buffer tanks on site seems like a false economy to me.

The same problem can exist for the fuel supply of standby generators . Some organisations depend on fuel tankers arriving within a few hours of a major outage. If there is a regional power outage it is likely the demand for fuel will exceed the logistics capacity of the tanker firms to deliver fuel. Standby fuel storage capacity is a risk which must be formally assessed and the decision signed off by business management. The danger is that some bright accountant will think money can be saved by reducing fuel stocks and utilising a Just-In-Time delivery.

The pace of invention

When I started work in my first job as a Gov't sickness benefits officer life was very different from now. If we wanted to send a letter to a claimant, we'd write the letter in handwriting, pass the handwritten letter to a typist who'd return a typed version complete with a carbon copy a few hours later. After the letter had been reviewed by the supervisor it was put in an envelope  in an out-tray for collection by the post clerk. The carbon copy would be filed in the cardboard file jacket of the claimant. You'd get a rocket from the supervisor if your letter contained any spelling/grammatical errors or there any obvious typing corrections. Our only contact with a computer was that we had to punch a couple of circular holes in punched cards for each case file. There were no keyboards on our desks, just a pencil, ruler, black ink fountain pen and sometimes a government issue ball point pen

Lasers were in the early stages of development in scientific research labs. Primitive liquid crystal display (a four inch pane changing from dark to light) was about to be demonstrated on television in the Tomorrow's World programme. Over the years I've seen many developments in the field of technology and computing. I was begining to wonder what was now left for man to invent and exploit. Some basic browsing during last weekend has convinced me there are a lot of new discoveries awaiting.

Here's some examples of recent developments that excite me.:

Willow glass by Corning - a mass produced flexible glass.

Wednesday 6 June 2012

Corporate Theft

There's one thing a bit off the topic of technology, security or trading, but is in way related. I travel a lot and use hotel and airline loyalty schemes. I rarely claim the awards from the schemes but instead I prefer to allow the points to build up to maybe use for a big trip. As it happens I don't go out of my way to use those airlines or hotels. I travel mostly out of necessity to support my business. What really gets my goat (UK english colloq. "annoys me") is that once the points have been awarded the airline/hotel think they can snatch back the points if I don't use them within a given time frame. So far as I'm concerned that is theft by the supplier, I don't care what legalese grey fine print is tucked away in some obscure place. It is theft, plain and simple.

So when a supplier such as British Airways (BA) of the Hilton Hotels group (Hilton HHonors™ scheme) claws back points from me I always punish them by withdrawing business from them. The punishment always costs them more that the value of the "points" that they've stolen from me, and the business goes to a competitor.

Another extremely annoying business practice is selling my account to another business as though my business with the supplier is some kind of commodity they can package up and sell to another business. The banks and credit card companies are particularly prone to do this. When I do business with another business it is because I choose to do so. Once again I punish the business who "bought" me for assuming that I'll do business with them.  Sorry guys but you have to earn my trust, you have wasted whatever you paid for me and it will go to your competitor.  British Airways have played a similar game by selling my customer points holding to Avios, sorry guys it doesn't wash, you both lose.

If big business had the courtesy to ask me if I had any objections about being transferred to another supplier I might be more amenable, but they don't so they lose. Loyalty is a two way street.

Tuesday 5 June 2012

Wake on Wireless

Qualcomm announces Wake On Wireless for pocketbook computers and other devices. This could be a security nightmare as a command system for explosive devices and bugging devices. Essentially the system administrator will be able to wake up remote computing devices without a wired connection.

Be prepared to block unwanted WiFi signals to your business campus! Here a wallpaper which might help, but beware building a frequency selective Faraday cage is not always simple. Active WiFi and cell phone blocking is not legal in the UK, unless you are the security services. This is not a new technology QinetiQ dreamt this up a few years ago. There's other research on this, here, here and here.

Friday 1 June 2012

Destroy your hard drives

The UK Information Commissioner has fined a Hospital Trust £325,000 ($500,000) for not ensuring 1000 Hard Disk Drives were properly destroyed by one of its contractors (Sussex Health Informatics Service).  At least 252 of those drives went astray.

There's no excuse for this really. There are several good techniques to ensure the safe and complete destruction of data on hard drives. I'll happily advise any one who contacts me as to how this destruction can be done. From the BBC report it looks like one individual was not properly supervised and thought he could make a quick buck by selling rather than destroying the hard drives.

Businesses and other organisations have a legal responsibility to look after confidential data. If the personal information on those disks had leaked to the public domain it could done serious damage to those concerned. These types of cases will continue until the main Board of the organisation takes these privacy matters seriously and not treat it as an obscure technical matter. Break the rules and you can be fined up to £500,000. To be honest, fines will not fix the problem, the penalty should include a criminal record and a prison sentence of up to two years for negligent Directors.

Here's the 8 key principles by which the Information Commissioner can judge your organisation's processing of information, both electronic and on paper. They are based on the 1972 Younger report, it is not exactly a new concept.

Data must be Fairly and lawfully processed
• Processed for limited purposes
• Adequate, relevant and not excessive
• Accurate and up to date
• Not kept for longer than is necessary
• Processed in line with your rights
• Secure
• Not transferred to other countries without adequate protection

Monday 28 May 2012

New Data Centre built near disaster site?

I received a twitter message about a recently opened Data Centre run by Gyron and let by SEGRO. From a quick looks their security looks good. A perusal of their web site reveals a company address of HP2 7SU. If you look on Google Maps there's a satellite view of the area. I'd always do a google check before further considering whether to use a Data Centre site. Oddly enough when you look around the satellite picture you see another memorable name of "Buncefield Terminal" located at post code HP2 7HZ. People living in the UK will probably remember Buncefield as the place of a major oil terminal explosion and fire.  The Police National Computer operations managers definitely remember this site.The distance between the two post codes is about 300 metres according to Google maps. According to commercial property news the area is bouncing back after the explosion.

I'm not saying the new Gyron site is located near a high risk, but it is worth checking this type of thing before choosing a data centre site. I understand the Buncefield site is now much safer. At this point I'm not sure of the exact location of the Gyron Data Centre, the post code may just be an office address.

I wonder what the insurance situation is in terms of risk if you knowingly site your data centre near to a fuel terminal?

Update 29th May 2012
I've visited the site referenced by the Gyron Post code. There are signs of a large data centre with chiller units positioned in the back yard. There was no direct evidence of the site belonging to Gyron. However the Buncefield oil terminal is very much in live operation and it is about 300 metres distant, easily within the bounds of a cordon which the emergency services might impose in the event of another disaster. There are some intervening buildings.  I'm not picking on Gyron as a "bad" example. I'm merely saying Data Centre management need to perform risk assessment on the businesses in the area surrounding the location of their servers.

Update: 1st June 2012.
I must be wrong about the risk if NTT have taken a majority share in Gyron. Apparently supported by the UK Gov't Business Secretary Vince Cable see The Register report here.

Sunday 27 May 2012


Here's an interesting introduction to developments in NoSQL Databases. Any company choosing this route would need to investigate carefully before taking the step, chances are it could lead to an instant legacy system with extreme maintenance issues once the enthusiast has left the development cycle.

Wednesday 23 May 2012

Keeping humans keeping your Data Centre

I came across this link which lists many aspects of looking after the people who run your data centre. It is well worth a read.

Friday 11 May 2012

Another business scam

Received through the post this morning was a rather obvious scam "reminder" to renew one of our trademarks. It comes from the "Intellectual Property Agency Ltd" based in the Broadgate Tower in London. It has the typical grey fine print on the back in contractualese which boils down to them wanting to con you into paying an excessive fee for renewing a trademark. Note there are no refunds. Their contract for UK Trademarks is oddly based on Swedish law. Here's their web site. They do not appear on the UK Companies House web site.

Their form look official and has a name similar to the official government body who handle these things. It is clearly designed to fool a dumb clerk or a dumb manager.

Using the proper government agency is a much lower cost to renew the trademark. There is not much use in complaining to any government authorities though; they will just wring their hands and do nothing. This problem won't go away until a law is passed telling the scamsters that they'll go to prison for a minimum of 5 years.

Wednesday 9 May 2012

Hooray comes to its senses

I get regular emails from advertising whitepapers from the marketing departments of the likes of Oracle, Dell etc. Most of these I ignore, but occasionally I get some which are of interest and want to read. When I click through the link in the email I'm inevitably faced with a demand to provide either user name/password or to complete a boring questionnaire about my company and my role.

Personally I find this really annoying and usually abandon the research at that point. Techtarget already have this information and they are the ones who sent me the email. What is the need for user Id/password? We are not exactly accessing deep secrets, it is just a damn marketing whitepaper.

At long last this morning when I tried techtarget there was one simple question "Are you still working for the same company?" This time I followed through and read the marketing whitepaper. 

I wonder how many other business men gave up in frustration with Techtarget's previous policy of lazy link management. 

Update 10th May 2012
 Whoops, it looks like I spoke too soon. Just had an email from telling me that due to "inactivity" my account with TechTarget has been suspended. A rap on the knuckles or what?

Wednesday 2 May 2012

Project control - project creep

Senior managers and directors are generally great in the role of defining the broad strategy of their organisation. However, the same people tend to be not so good in the process of detailed specification for project requirements. A consequence is that when the budget funds for a new project is agreed and authorised it is usual that some features or services will be omitted from the authorised project. This omission  might have happened accidentally or intentionally.

As the the project progresses it is inevitable that senior managers, clients, team members will identify some feature or facility that "should have been included" within the project. It may be the proposed variation in the Project is a reasonable and sensible change. In those cases the changes can be incorporated by the variations route. In the cases where the Project Manager thinks the proposed change is stupid or would cost too much, or cause too much project delay he/she should reject the proposal via the variation process. No doubt the time will arrive when the project scope will have to be "frozen" in order to achieve a delivery date, but it always possible to parcel off some features to be delivered after the main implementation date.

The worst kind of project creep occurs when a senior manager/manager declares a related service feature must be included in the project even when it is nothing to do with the original project. The senior manager will even override the views of the project manager regardless of the impact on spend or the impact on project resources/time table. In those cases the project manager must be careful to document the impact in advance and then ensure the consequences are brought to the attention of the requesting manager/client. If there is no change in the senior manager's/client's demand for inclusion then the origin of the request and consequences should be included in the formal project progress documentation. The worst thing the project manager can do is to attempt to absorb the additional cost and resource time. A similar process should be followed where the senior manager/client instructs that a previously agreed element of the project should be removed.

A recent example I've seen of the latter is on a project to install a software based stock control system in the stock rooms for the consumable clinical equipment used in the operating theatres of a major hospital. Their existing manual systems are frequently out of date and the local operations managers have no idea of what is actually in stock. The project manager is rolling out this new system storage room by room. The project involves training staff in the new techniques and loading inventory details on the stock database. However during the project lifetime there was a major move from one hospital site to a new building. A substantial amount of stock, millions of pounds worth, was disposed of on the grounds of "hygiene", but in the rush of the move the stock disposals were not recorded in the manual records. A couple of months later a sample audit revealed the consequent discrepancies in the manual stock records and spreadsheets. 
A senior hospital manager diverted staff resources from the new stock control project team to undertake a full stock inventory exercise in order to be able to report to the Board the "reasons" for the difference. This "simple" command created an unauthorised extra £35,000 bill for consultancy and delayed the stock control system project by a month. Rather than use in-house clerical administrators to perform the inventory, expensive external software consultants were used to count scalpels. The senior manager subsequently tried to deny that she was the cause of overspend and delay to the project, but the project documentation showed otherwise.

Wednesday 18 April 2012

Project Control - Variances

A major part of the work of my company (Oaksys Tech Ltd) is project control. The process of fit out of a trading floor or data centre often involves close contact with a Main Contractor from a constructions firm. The Main Contractors have long experience in the project control of variances. It is inevitable that during the course of a project there will be changes to the original plan requested by the project team or the client management. Such changes will normally involve additional expenditure and/or a variation in the project time scale. There's nothing wrong in having variances provided the change to the project plan is properly authorised and agreed with the business or the part of the organisation who is ultimately paying for the project. The provision of funds for variations does not appear as if by magic, the agreement to pay should be documented. Similarly the acceptance of a variation in project time scale, if any, some be documented. Be sure if there are changes in a construction project the Main Contractor will document those and expect additional payment for those variations.

The project manager should track variances as an ongoing part of the project management process. He/she should avoid being surprised at the end of the project with a large list of project variances and a demand for supplemental payment. It is all too easy for the person requesting the original change to disappear from the project team prior to the end of the project leaving little evidence of why the change was requested and who agreed to fund the change/variance. Uncertainty over the reason for the change can lead to heated arguments and even expensive arbitration or legal fees.

The Main Contractor will sub-contract work to other contractors. The process is loaded with opportunity for mistakes, miscommunication and the use of out of date project specifications. In effect the project hasn't changed but the mistakes will have caused delay and additional expenditure. Delay or mistakes by one sub-contractor can affect other sub-contractors, leading to variations claims by them. The management team of the Main Contractor will have to explain to their owners why the cost of those mistakes will be deducted from the expected profits. In other cases the Main Contractor may not have delivered parts or all of their project on time leading to their client expecting compensation or withholding payments. Sometimes the Main Contractor will report those variations as having been caused by the client in an attempt to extract further payment from the client or to avoid paying compensation. It is important that the project manager challenges and rejects any such false claims of variation. Such challenges should occur as an ongoing process during the life of the project. Variations on the part of the contractor should be documented in an agreed process between the client project management team and the Main Contractor.

Ultimately the business rule of  "no payment without purchase order" should apply. In the case of a trading floor/ data centre project the Purchase Order might be substituted by a variation document. As with purchase orders, there must be agreement from the funds holder/owner that the variation in expenditure is permitted.  The big challenge to the Project Manager is ensuring there are no project delays caused by waiting for variation approval.

Monday 16 April 2012

Hidden security

A normal security provision in a data centre and trading floor is some form of CCTV monitoring and recording of key areas within and around the organisation's building. A visit to IFSEC at the  National Exhibition Centre in Birmingham UK will quickly convince you there are thousands of different CCTV security solutions. The art of placing the cameras is a balance between risk/cost. The costs involve installation, power, cabling, capital cost for equipment, operational cost for use and long term image storage. The more cameras you have the greater the capital cost and the greater the operational cost. 

These systems can have great deterrence effect and can help in the detection of offenders intent on crime.  However if the cameras are poor quality and not well placed they can be a waste of money. If the criminals are aware of the camera locations they can avoid their faces being recorded. Simple techniques such as looking away and "hoodie" type clothing can destroy the effectiveness of the CCTV systems. The careful positioning of some hidden or covert cameras as part of the overall security system can help to detect the faces of criminals who would otherwise attempt to defeat the visible or overt cameras. Hidden cameras can be positioned at eye level to record facial images whereas overt cameras may have to be positioned at a high level to prevent damage by criminals. 

I'll relate a true life example which happened in a major international bank. The bank in question had a good CCTV security system protecting access to its data centres. The visible camera systems were of high quality, were well positioned, and the CCTV recording system provided good quality playback. One of the Bank's data centres was subjected to a raid by well organised thieves during the evening after the day time computer operators had left. The entrance to the building was protected by 24 hour security guards. A thief "tailgated" one of the operations staff during the day to gain access to the main technology area and hid somewhere in the room. This was a suspected, but never proved "inside job".  Later in the evening the hidden man opened the door to the data centre to allow access to other thieves who were disguised as cleaners. They wheeled in a large high sided trolley into the technology area and set to work pulling Sun servers from the computer racks/cabinets and loading them into the trolley. It was a crude process as they were pulling live operational servers from the cabinet without first powering down the machines. In less than five minutes they'd filled the trolley with expensive Sun servers. They covered the trolley and wheeled it out of the building under the watchful(!) gaze of the security guard who assumed they were departing cleaners. The stolen equipment value was in excess of £200,000 but the greater cost was the recovery of the systems the following day.

Throughout the whole process the gang of thieves kept their faces shrouded and hidden from the from the visible CCTV security cameras. The police believe that they'd been told of the position of the cameras by their "inside" accomplice.  However unknown to the gang of thieves there were also a set of covert/hidden security cameras which provided evidence quality video of the faces of the thieves. They were apprehended within a few days and are now serving time in prison. The covert cameras had been positioned to catch the faces of people hiding from the visible cameras.

Friday 30 March 2012

Queuing for fuel at UK filling stations.

One of the UK trades unions (Unite) has received member approval from the fuel tanker drivers to call a strike in respect of deliveries of fuel to filling stations for Diesel/DERV and/or petrol/gas. By law the union has to give five days advance notice of the actual strike dates, but so far they've not been announced.

Some ill judged comments by a couple of government ministers has prompted the public to rush out and top up their vehicle fuel tanks prior to the strike. In some cases people are filling fuel cans with additional spare fuel, in the case of petrol if it is badly stored it can increase risk to the family home. On the whole I think that keeping the fuel tank well topped up is a good precaution in advance of a potential strike.

Unfortunately the public commercial fuel distribution system for petrol (gas) and diesel is a fragile model based on the "just-in-time-delivery" system which may reduce supplier costs slightly, but which has little resilience to a surge in demand or a sudden interruption of delivery. The filling stations only hold 2-3 day's supply of fuel in their storage tanks and depend on regular timely delivery by tanker lorries. The public know this and want to avoid the inconvenience of not having fuel for their vehicles so they take the next opportunity to top up with fuel which exacerbates the limited fuel supply problem. In this morning's news it is reported demand for petrol is 172% of the norm on 29th March 2012.

The filling station managers attempt to reduce the flow of fuel from their limited local storage tanks by shutting off most of the filling pumps on their forecourts. As anyone with a rudimentary knowledge of queuing theory knows reducing the number of service points for a single queue tends to lengthen the queues. As a consequence the whole thing spirals into a worsening situation, though in the longer term the overall amount of fuel consumed by motorists doesn't actually alter.

The whole situation would be considerably eased if the owners of filling stations were forced by legislation to routinely hold ten times the average daily fuel sales in their local storage tanks. In effect increasing the capacity of the system to withstand short interruptions to the supply chain. The cost would be a few pence per vehicle driver. In effect the interest payable on seven days fuel, even less if the tax system was altered to be charged when fuel is dispensed rather than when it is delivered. The suppliers blame the motorists for demand variation when in fact it is the suppliers who create the problem in the supply system.

Regular readers may be asking how this affects the operation of a trading floor/data centre which the normal focus of this blog.  Firstly there is the risk to the top up supplies of diesel to the standby generators. If tanker drivers are either on strike or are servicing public filling stations in preference it is a risk to your business. Secondly the fuel situation could affect vehicles of key personnel or delivery vehicles. Maybe the assumptions of availability of replacement fuel for your business are not valid, regardless of any contractual commitment by your suppliers. In the absence of a resilient national infrastructure for fuel supply it may be worth considering alternatives such as increased or dispersed fuel storage, perhaps even providing fuel to key employees vehicle during a national fuel supply crisis.

Here's another list of useful points to consider.

Wednesday 28 March 2012

Cheaper Photovoltaic Power?

3D stack of photovoltaic cells. Increase output by 20x

Cheaper production of silicon photovoltaics.

Edit 30th March:
Amazing! I posted the above two links really as a memo to myself to do more research into these technologies Yet I've had loads off people reading this page. Must be hot stuff!

Wednesday 14 March 2012

Efficient combined cooling - Turbomiser

As you can see from the recent blogs, a couple of weeks ago I visited the Data Centre World exhibition. I've suffered a few days outage thanks a cold caught from a stupid woman who sat next to me on the train home, but finally I've been able to complete research into one of the devices I saw at the exhibition.
The device was the Turbomiser which was presented by Klima-Therm. The Turbomiser is a compact self contained chiller unit which weighs in at a relatively light two tonnes. It makes good use of recent developments in efficient chiller technology to gain from cheaper methods of cooling when climate conditions allow. The core of the unit is an intelligent controller (Geoclima Microsmart) which coordinates the mechanics to use the most efficient method of cooling. When free air cooling is available the Turbomiser will make maximum use of that. Where external air conditions permit, evaporative cooling is automatically used to reduce the cooling load on the chiller compressor. The compressor built into the unit is the highly efficient Turbocor twin rotor unit.  The heat exchange panels are constructed using micro channel aluminium which give good efficiency while reducing the amount of refrigerant required.

The combined technologies lead to greatly reduced operating costs with payback period of 2-3 years. The unit looks to be a good option if you are in the position to install new or refit an existing installation.

The evaporative cooling does not use aerosols sprays to cool, but moistens evaporative panels. Air flowing over the panels is cooled by evaporation of the water. Avoiding the use of sprays and the lack of standing water avoids the risk of Legionella  infections.

Friday 9 March 2012

Resilient power housed in a shipping container

One more power item from the Data Centre World show that piqued my interest was the Advanced Diesel Engineering Ltd "plus one" continuous power unit. They've squeezed a pair of Volvo (2 x 410KVA) diesel generators, a pair of fuel day tanks into a "shipping container". They have a combined output of a shade over 800 KVA or 400 KVA in a resilient ( 1+1) configuration. They have a five hour endurance on the just the day tanks.

From my casual inspection it looks well designed and well constructed and ADEL have been in this business for some time.
The total self contained unit weighs about 13 tonnes wet weight and has a footprint of 6 x 2.4 metres. The standard silencing yields 75 dB(A) at 1 metre when both engines are on full load.  The unit is essentially plug and play.  If you have the need for an easy installation for 400 KVA of back up power and can handle the site placement of a shipping container, then this may be the solution for you. Do lock it down well though, they have a very similar unit on special offer at £140K at the moment.

Riello do a similar unit which they call "PowerBox", but this has a single diesel motor coupled with a rotary UPS.

Monday 5 March 2012

Air Battery UPS

An interesting technology at Data Centre World was the Air Battery. It is a system designed to provide 30 seconds of uninterrupted power to a Data Centre electrical power UPS system. The technology uses stored compressed air to drive fast starting in-cabinet expanding scroll rotors. Ultra capacitors are used to provide energy during the brief start up time of the scroll rotors. The  endurance of the Air Battery can be extended by use of additional stored compressed air tanks.

The advertised power ranges are from 3KW right through to 3MVA. The technology has been deployed in National Grid environments, so it is no lightweight technology struggling to find a market. The units depend on having a fast start generator to take up the power load as the air supply depletes. To facilitate fast generator start up, heat from the air battery can be used to pre-heat the diesel motor. Such diesel generators need to be kept at 35 degrees Celsius to aid fast starting. When the air battery is triggered and producing power there is a by-product of cold exhaust air which can be used to contribute to the cooling of a data centre while the air conditioning await the power from the standby generator..

The manufacturers claim lower operating costs, lower maintenance and greater reliability over a stored energy rotor system because there are no moving parts until the scroll rotors are activated. (excepting recharge compression of the air storage tanks.  Bearing friction, air friction and even the centrifugal effect of the earth's rotation can affect the storage efficiency of a rotor system.

Similarly this system is claimed to have a a lower ongoing operational and maintenance cost when compared with a lead acid battery power stage system. Batteries need a trickle charge to maintain power over long periods. In the case of gel acid lead batteries there is a cost a relatively short life. If you keep the gel acid battery cool it should give a five year lifetime. Wet acid batteries can survive 10 - 15 years depending on their construction and storage conditions. If the ambient temperature in the battery room is allowed to rise over 20 degree Celsius (about 70 Fahrenheit) the battery life will be shortened. So there is also an implied air conditioning cost unless free cooling can be used.Wet acid batteries generally need a specially constructed secure room with good ventilation, temperature conditioning and an acid proof tanked floor.

We managed one trading floor data centre implementation where the lead acid batteries had to be located away from the UPS power unit by reason of the weakness of the floor in the Data Centre Room. We had to install a very thick copper cable to take the low voltage power from the batteries to avoid too much voltage drop before it reached the UPS unit. Going from the basement to the fourth floor meant we were working at the absolute low voltage power cable length recommended by the manufacturers of the UPS. If we'd been able to use a compressed air battery we'd have been able to use much longer compressed air pipes.

Friday 2 March 2012

Standby generator fuel contamination

I paid my annual visit to Data Centre World in Olympia London yesterday. Several items caught my attention. One of those was the IPU Fuel conditioning stand. I'd read about the "diesel bug" problem a couple of months ago but not really paid much attention, but it could be an important issue for Data Centres/businesses which depend on diesel powered standby generator for their electricity backup. This may be an issue which IT managers think is purely a responsibility for the facilities manager, but at the very least they should ensure that their FM colleagues have considered the issue. Similarly if your business operates in a multi-tenanted building it may be appropriate to ask the Landlord how they approach this matter.

Essentially the diesel bug is contamination in diesel fuel arising from bacterial, fungal, yeast growths which can grow in water contaminated diesel. The diesel bug can grow in to a mat which floats between the interface of diesel/water layers. The mat can also form as a plaque on the inner surface of the diesel tank, pipework and filters. If that mat is disturbed chunks can float off and be sucked into the fuel filters of a diesel motor, such as a standby generator. In a way it is a bit like the diesel engine is having a heart attack as essential feeds are blocked. The by-products of the diesel bug can also create acidic conditions which can corrode metal tanks and fittings.

If the fuel filter of  diesel motor becomes clogged with contaminants the motor will be starved of fuel and cut out. This is not a good thing if there is no mains power and your data centre is depending on the standby generator. Even in N+1 resilient generator configurations the risk is high if the fuel is contaminated.

Diesel fuel in the UK can contain up to 7% bio-diesel. Bio-diesel is hygroscopic, which means it is prone to absorb water vapour from the air. In the diesel storage tank this absorbed water can separate out in to the base of the tank, potentially leading to the diesel bug. If the water reaches the injectors of the diesel engine it can cause serious damage to those injectors.

IPU were offering fuel polishing units which can be installed, either permanently or as a mobile unit, to remove contaminants, such as water, from the stored diesel in the fuel tanks of a standby generator. They also provide assistance in testing the fuel to detect contamination and bacterial growth. The stagnant conditions of a standby generator fuel tank could mean that the problem of the diesel bug in your fuel does not become apparent until you most need the generator for an extended period of operation. Short monthly test runs might only create only minor disturbance in the fuel tank, but a longer term run in a live regional outage could see your generator failing and sustaining serious damage.

The fuel polishing technique could also be used to avoid the potential problem of contaminated fuel being delivered by tanker to your main tank during a period of extended mains power outage, such as a regional power outage. One solution may be to deliver the fuel to a holding tank, polish it and test it before releasing it to the main fuel tank(s). The equipment required to polish the diesel is not terribly expensive, it is a technology which has been around for some time in the maritime and transport industry.

I'll certainly be adding this topic to the check list of questions I use when auditing a data centre site for physical security. It is a risk factor that can be avoided.

There is a short video of the IPU system in operation:

Here's another useful blog site on generator issues.