Archive for martin
Green IT is the way forward in many ways
Sun’s Workplace Innovation Initiatives and Energy Efficient Technologies and Services Lower Carbon Footprint Significantly for Company and Its Customers
SANTA CLARA, CALIF. October 8, 2008 Sun Microsystems, Inc. (NASDAQ: JAVA) today announced the reduction of its greenhouse gas (GHG) emissions from US operations by 23 percent, surpassing its goal five years early. Originally committed to a 20 percent reduction over 2002 levels by 2012, Sun’s combination of workplace innovation and use of energy efficient technologies and services, including its Sun SPARC Enterprise servers with CoolThreads technology, helped the company surpass its original goal in 2007. As part of its continued commitment to Eco Innovation and the goals of the U.S. Environmental Protection Agency’s (EPA) Climate Leaders program, Sun has pledged an additional 20 percent GHG emissions reduction from worldwide operations over 2007 levels by 2015. To hear more about Sun’s achievement, listen to the following segment on Blog Talk Radio: http://www.blogtalkradio.com/stations/sunradio/SunNews/2008/10/02/Oct-2-2008-459PM
The more we talk about Green IT, both in terms of corporate social responsibility, discussing not only the costs savings, but the operational and end user empowerment of doing so, the more we can illustrate Green IT as an enabler to your business. Not just the cost of doing business. That by consolidation and virtualization of the infrastructure on to newer hardware, not only do I save money on operational costs, energy, support contracts etc, I can abstract the user from the infrastructure allowing me to scale the infrastructure more in line with the business needs. Need more disk space or memory/virtual cpu’s for that batch run on Friday, not a problem. That’s empowerment through Green IT. More so than the financials, though they are not to be under-sold.
Talking about HP
Last Friday, not only did I get a first look at HP’s containerized data center, but I was given a tour of its factory floors in Houston, where HP makes high-value custom servers for clients. One factory makes the servers, while the other assembles the computers into racks and complete systems. The systems can contain any combination of gear that a client could want, including servers from rival hardware vendors. (photos after the jump)
The factories are roughly 140,000 square feet each and employ about 1,600 people working around the clock, five days a week. HP keeps enough material on hand to keep the plant going for 12-16 hours without new inputs from vendors with warehouses in the Houston area. Each day they use about 3,000 chips worth some $3 million. The chips arrive twice a day in an armored lock box accompanied by guards. The photos below document all the steps involved, from gathering the boards and chips to the shipping of individual servers, with a quick shot of an HP system being built. I wanted to get a shot of the lock box of chips, but was denied.
An interesting post showing the HP factories and discussing their production lines and container data center, do check it out!
Using IT as an enabler
In recent years, Vienna Insurance Group has grown beyond Austrian borders and has become a leading international insurance group in Central and Eastern Europe (CEE). To consolidate their powerful IT growth and create resources for further growth, the company reinforced its partnership with IBM.
Günter Geyer, CEO, Vienna Insurance Group, said, “With this step, we can concentrate on our further growth and our core business (main tasks). Now we are prepared to further develop our role as a forerunner in the insurance sector. Trendsetting developments in the insurance sector can be driven quicker in the future.”
Check out this article talking about how this organization is using IBM to allow it to focus on its core business and have IBM concentrate on delivering IT services on demand, in line with the business needs. It’s something that is set to continue, particularly as we are able to abstract the user from the application and underlying infrastructure - that I can provision and request only what I need to meet our business SLA’s, our application requirements can save money and improve our ability to earn revenue.
DataSynapse continues FabricServer innovation
DataSynapse Inc., a leader in dynamic application service management software for the next-generation data center, today announced integration of its FabricServer® software platform with Cisco’s VFrame data center provisioning and infrastructure orchestration product. As a result of this technology integration, enterprise data center architects can now dynamically scale a complete application service stack as virtualized application and hardware infrastructure objects. As a result, IT organizations can reduce capital and operating costs while delivering significant improvement in time to market and service levels for critical business applications.
FabricServer and VFrame treat the application and infrastructure as service components, respectively, and can dynamically apply or remove these services in unison. Customers benefit from right sized hosted infrastructures and enhanced capacity with real time utilization policies.
An article talking about innovations of FabricServer, I’m off to read up more, do check it out. Anything the vendors can do to aid infrastructure provisioning, application workload management - of on demand computing has to be a good thing for the end user in terms of efficiency and delivering capacity in line with the business need.
IDataPlex is the way forward
Merrill - which is being acquired by Bank of America - will use the iDataPlex servers to help build and evaluate new risk analysis programs, says IBM.
The new server model is part of IBM’s $100 million “Blue Cloud” initiative and is designed to achieve “stateless computing that effectively turns many separate computers into a pool of shared resources”.
IBM says this paves the way for mass-scale, Internet-style computing in a compact, energy-saving package. IDataPlex more than doubles the amount of systems that can run in a single IBM rack - packing five times the compute power of typical systems, claims the vendor.
It’s great to read up about IBM’s IDataPlex, and see how Merrill is benefiting from the platform. I remain a fan of the concept and can see how IDataPlex would be great in a web farm or grid/hpc solution. I’m off to read more.
Solid state disks in laptops rule
I’m totally finished with spinning drives on my ThinkPads. At least if I have anything to say about it.
I was fortunate enough to be able to swap to a solid state hard disk drive in my ThinkPad this week. I had always derided the 64GB capacity as too small, but when the opportunity presented itself, it was too good to pass up.
I have always been a fan of 7200 rpm HDDs in notebooks, long ago having rejected 5400 rpm as being way too slow for use in any system that does more than surf the Web. Thus, a 7200 rpm HDD has been a basic requirement for me in my last three notebooks.
Let’s look at the requirements of a PC running three years ago vs. today. Both may be running Windows XP, but today’s PC has to contend with more personal firewalls, security scanners, management agents, and system utilities all running constantly in the background. Over the next 1 -2 years, plan on adding virtualization to this list. This is before you have even launched your first application. Worse, they all require care and feeding (i.e. processor cycles) in order to keep themselves up to date.
In short, though your application load hasn’t changed much in the last several years, your background computing load most certainly has – and not for the better. All of this activity is heavily disk bound, and I didn’t realize just how much until I switched two days ago.
An interesting article illustrating the benefits of a solid state disk, not only is it set to be more power efficient they should be more reliable long term - though we’ll have to see. Do check it out, if you’ve been wondering about the solid state disk in your laptop.
Training is the key to success - can the vendors help me?
A National Skills Academy for IT, announced yesterday and due to open in 2009, will bring together an unprecedented, sector-wide collaboration of employers, educators and stakeholders to meet the skills needs of the IT workforce.
Information technology (IT) skills are critical to the growth of the UK economy. There are currently over a million IT professionals in the UK, with 141,000 new recruits needed every year. The IT industry is predicted to grow at five times the rate of the workforce as a whole and recent research suggests that optimising the ICT capabilities of the UK economy is worth as much as £35 billion a year.
Karen Price, Chief Executive of e-skills UK said: “e-skills UK is delighted to be working with employers across the sector to get the skills academy off the ground. It offers a unique opportunity for employers to take collective responsibility for the skills and accreditation of the IT workforce, with innovative development programmes and qualifications that are valued by the sector. I believe this will play a major role in helping the UK become a world leader in IT in the coming years.”
The more training we can provide to our end user community and our IT teams, the more we can improve the services we offer, align what we can deliver with the business need, to add value and illustrate IT’s enablement of the end user community.
I wonder if we could not see more from vendors?
Improvement in not only the training they can provide, but the product knowledge and accessibility of the end user to their support, their guidelines and best practice.
That I can easily access the information I need about a product, what firmware should be used with my products - “what happens when content….”. That I can type http://vendorname.com/support, type in the product name and be given instant information, drivers, the manuals etc.
We need to change vendors support sites, from one of designer layout, of structure and corporate branding to one around the end user need based on the way the user looks for information.
Could I one day have an engineers blog talking just about DL380s? That blog would issue firmware updates, recalls/announcments - why I want a DL380G4 not a Dl380G3 etc. An IBM guy talking about what he’s doing on a daily basis with his x3550 with links to content/articles, best practice, “when swapping a drive do this…”?
That on the Proliant 2500 blog, there might be an entry for the engineer that had to visit to fix a Compaq Proliant 2500 which had blue screened and all he did was upgrade the array controller firmware to fix it (Windows NT hotfix was the cause) . On the DL380 blog, that the back-plain had failed on another clients’ DL380 and that he could tell this because….(several hard drives all reported as failed, but when re-seated recovered, then started failing again). You ask any windows server guy who’s played with a vendor server platform, say a model name and hardware issue and I’m sure they’ll know a few quick steps to try - oh ASR, check the firmware, drivers and think about re-seating the memory….
True empowerment to the end user community. All realistically at low cost to the vendor (the engineers are dealing with these issues on a daily basis) but of great effect to the end user. 9 times out of 10 I have a five minute question, the kind of thing that could be blogged, could be answered and pre-written/responded to without needing to trouble an engineer directly, or providing product support information - serial number, model number etc.
Understanding that 70% of the users accessing vendor.com/drivers want drivers, displaying that information quickly and easily is key, everything else is secondary - thus the use of search engines to get the information I want. I can go to ‘vendor.com’, click support, click drivers, click product classifcation, click product line, click model, then click drivers again…
That information is displayed in product structure not platform, I know I have a DL380G3 or an IBM X3550, but as a new user, I might not know that comes under X series servers, or Proliant servers. The URL address, the experience with the site in terms of downloading the drivers and accessing the information can be just as significant to the end user experience as the equipment working. Could I not type http://www.ibm.com/x3550/drivers and be re-directed internally in the IBM site? Or http://www.hp.com/dl380g4?
The more we empower the right users with the right information, the more we can empower end users with experience of the platform, with empowerment to get their technology working for them, to avoid those “I bought blades from… and they were rubbish because….”.
VMWare on Apple the powerful combination
VMware reports that its Fusion software for Macintoshes is being used by scientists at CERN, the European Organization for Nuclear Research, creators of the Large Hadron Collider.
VMware Fusion enables Intel-based Macs to run non-Mac OS X based operating systems without having to reboot first. CERN scientists are using Fusion to share Linux-based computer code on Fusion “virtual machines” running on Macs. The software links the computers to the LHC Computing Grid — a network of about 40,000 CPUs.
I was reading up about CERN during the week, it’s great to see how the technology is being used, and pleased to read about their use of VMWare on Intel Mac’s as well as grid technology. Do check it out.
IT Managers to reduce IT carbon footprint
The majority of IT managers are feeling pressure to reduce their carbon footprints, according to a survey commissioned by Zycko, a value-added distributor ofconvergent IT solutions. 61% feel pressure to cut down on their energy usage and 60% plan to reduce their carbon footprint within the next 18 months. Encouragingly, most (70%) already have some green measures in place such as remote working and video conferencing facilities to reduce the need for employees to travel.
The independent OmniBoss research, conducted by Vanson Bourne, provides a detailed overview of the green IT issues currently faced by medium to large UK businesses.
The survey found that:
• 60 % of UK businesses plan to reduce carbon footprint in the next 18 months
• 61% feel pressure to reduce carbon emissions
• 67% of IT managers do not know how much power their data centre draws
• Only 45% enforce shutting down computers at night
Check out this interesting article highlighting the interest in Green IT, as well as the adoption of ‘Green IT’ in terms of shutting down computers at night. The rising cost of energy is probably going to be a driver for adoption of green practices as much as those of corporate social responsibility. Recognizing the financial as well as the operational benefits of good systems configuration/green IT can improve service to the user experience, to your support costs, your energy and air conditioning costs.
By switching the pcs off overnight, you can reduce the power of the air conditioning, you should reduce the number of faults reported to the helpdesk - the computers should be less prone to memory leaks, explorer hanging, to one process going bad causing the desktop to freeze. End user experience improvements from an IT perspective might be just as important if not more so than saving money.
Are blade servers for everyone?
This note describes a conversation I’ve had multiple times with data center owners and concludes that blade servers frequently don’t help and they sometimes hurt, easy data center power utilization improvements are available independent of the blade server premium, and enterprise data center owners have a tendency to buy gadgets from the big suppliers rather than think through overall data center design. We’ll dig into each.
In talking to data center owners, I’ve learned a lot but every once in a while I come across a point that just doesn’t make sense. My favorite example is server density. I’ve talked to many DC owners (and I’ll bet I’ll hear from many after this note) that have just purchased blades servers. The direction of conversation is always the same. “We just went with blades and now have 25+kW racks”. I ask if their data center has open floor and it almost always does. We’ll come back to that. Hmmm, I’m thinking. They now have much higher power density racks at higher purchase cost in order to get more computing per square foot but the data center already has open floor space (since almost all well designed centers are power and cooling bound rather than floor space bound). Why?
Earlier, we observed that most well designed data centers are power and cooling bound rather than space bound. Why is that anyway? There is actually very little choice. Here’s the math: Power and Cooling make up roughly 70% of the cost of the data center while the shell (the building) is just over 10%. As a designer, you need to design a data center to lasts for 15 years. Who has a clue of the needed power density (usually expressed in W/sq ft) 15 years from today? It depends upon the server technology, the storage ratio, and many other factors. The only thing we know for sure is we don’t know and almost any choice will inevitably be wrong. So a designer is going to have too much power and cooling or too much floor space. One or the other will be wasted no matter what. Wasting floor space is a 10% mistake whereas stranding power and cooling is a 70% mistake. This 10% number applies to large scale data centers of over 10MW not in the center of New York – we’ll come back to that. Any designer that strands power and cooling by running out of floor space should have been fired years ago. Most avoid this by providing more floor space than needed in any reasonable usage and that’s why most data centers have vast open spaces. Its insurance against the expensive mistake of stranding power.
Check out this great post with some relevant and valuable comments. Blade servers are by no means the answer to all your issues (business and IT related). You need to think about how you’re implementing the technology, how it coupled with your business processes, the way you and your IT works, will work in providing the IT you want for your business. Implementing a virtual infrastructure for example can bring real benefits in delivery and availability, however your billing, your support and issues of ownership need to be discussed, resolved and processes created as you implement the technology - the what if scenarios.
Hello everyone ! Can anyone tell me what is the estimated value of this green datacenter complex ? And when ...
Yes, outsourcing can bring challenges to us, especially for the owners. Even before there are lots of challenges in outsourcing. 
Internet protection is a must for business security and data integrity. Electronic Banking 
We are in the process of doing much the same; decommissioning our relatively old installations at Docklands in favour of ...
I always use `hponcfg -g` from the commandline of the Linux server. 
Martin, It's great to see that you have picked up this article, many articles report large numbers and tempting savings ...
If you have problems, email me and I'll send you a few config scripts. 
To answer your question:
The six data centre buildings will be powered from multiple redundant sources, principally from the adjacent 38MW ...
Given the depth of some of the Piles Driven into the ground for such buildings or companies with large car ...