The future of corporate computing will be thin client connecting to a virtual data center (VDC). The following is from a Yankee Group white paper (emphasis below mine - paper not available online. See here as well for a 2008 interview with Zeus Kerravala, SVP of Enterprise Research at Yankee Group)
Yankee Group’s vision of an Anywhere Enterprise™ is an organization whose employees, customers, assets and partners connect to applications, information and services when and where they need them. Yankee Group’s definition of the Anywhere Enterprise is built on the following five key pillars that will drive data center transformation:
Consumer technology will continue to lead the way in delivering a personalized IT experience.
Content will be king as information becomes available everywhere but it needs to be stored and secured in the data center.
Client devices will continue to evolve, creating an environment where more data will need to be stored centrally in data centers and the end device will become a thin terminal.
Connectivity becomes seamless, creating a connectivity fabric enabling users to work from anywhere.
Collaboration will be a key focus as companies look to make key decisions faster by harnessing information from across the extended enterprise.
The shift to an Anywhere Enterprise will bring forth a new era in data center computing—the virtual data center
Just as it is hard for many of us baby boomers to really grasp and use the technology our grandkids are using and developing today, how it will evolve and change the world around us (see Robert Cringely’s excellent article on this), so too is it for many in the data center server business to see that the future of data centers is complete virtualization. Nevertheless this is the future. It is a future that all the big software and hardware providers, such as Microsoft, Cisco, IBM, VMWare, Exanet and many more are planning for. It is also where RackForce is headed.
There are many reasons for this path, not the least of which are Total Cost of Ownership (TCO) benefits and reduced GHG emissions, a factor which is becoming ever more important for corporations as the public, governments and shareholders start demanding better environmental practices. People are becoming ever more connected. As cell phone technology evolves mobile phones have become minicomputers and their capabilities just keep growing. They are becoming thin clients to online web services. This has allowed for an even greater distributed workforce as employees are no longer constrained to working in their office’s desktop environment. This movement puts tremendous pressure on corporate data centers as the work is offloaded from the PC to the server. Enter the virtual data center.
Consider a typical data center as it stands today. While the servers may be networked so that they can communicate with each other they each have their own separate hardware resources, RAM, CPU, storage, and those are dedicated solely to that piece of hardware. This leads to both under utilized resources (typical hardware utilization in a traditional data center is between 25% and 35% - source: Yankee Group) as well as maxed out resources and increases the number of physical servers a company must have. This increases a whole range of costs: network infrastructure, cooling, power, hardware, personnel, and so on. The VDC solves these problems.
Earlier I said that the big software and hardware companies are planning for a future of VDC. Nowhere is this planning more important than on the network side. Not only does the network infrastructure have to carry out it’s traditional roles, but in the VDC it faces a new challenge; connect all the diverse pools of resources available on the network and create an environment where any resource (RAM/CPU/storage) can be accessed as needed by any device. Obviously this creates a whole separate set of challenges for the underlaying network hardware. To begin with the network must be:
Ubiquitous: data center components can be located anywhere with only the network itself connecting them
Scalable: It must be able to grow with demand and it must do so without interruption to services. An example of an interesting Network Access Storage system is Exanet’s Exastore system
Neutral: It must be transparent and invisible to the devices and endpoints residing on the network
To achieve those goals requires a whole new breed of network hardware and software. Systems that can be upgraded with no interruption to service. That last point is very critical to the successful operation of a VDC. Cisco, for one, is working on it. See their NX-OS, Nexus 7000 switches and their Data Center Network Manager as examples.
IBM is moving forward with their Big Blue project which is morphing into Big Green and 3-D Data centers. The 3-D virtualized concept is quite an intriguing way of managing IT/Data Center resources
IBM is giving new meaning to the phrase “virtual data center.” And it looks a lot more like Second Life than VMware.
Rather than build a virtual world for online gaming or to give users an alternative reality, Big Blue made a virtual world where IT executives can examine and manipulate hardware running in their very real data centers. The IBM project — called 3-D Data Center — gives IT shops a 3-dimensional, real-time virtual view of their data centre resources, even if they are spread across the globe.
“It’s a new way to look at systems and interact with them,” says IBM researcher Michael Osias, the man behind this new idea. “Objects aren’t just visualizations. You can think of them as little machines.”
So instead of battling wizards and warriors, data centre administrators get to play with their servers and storage ( compare storage products ). And it does look something like a game, even if it is not one, Osias notes. IBM contends its new technology will help businesses identify underutilized machines that can be eliminated, distribute workload among data centers, monitor power and cooling, and move processing to cooler sites depending on the weather.
Using avatars, IT operations executives move through their virtual data centres, viewing “a tailored 3-D replica of servers, racks, networking, power and cooling equipment.”
For a different look at how this is being implemented see this article on Ugo Trade. For the same reason that online 3-D virtual reality games, such as Second Life, became much more popular than the older text based RPGs, adding that 3-D level of abstraction makes managing IT resources much easier. It makes it more intuitive for starters.
So you can see the overall sense of the virtual data center, save energy, save labor, save hardware costs, increase file storage/retrieval efficiency, access from anywhere at any time via any device, this is the gist of cloud computing. It is the future fast approaching. Even Google has now opened up its cloud to developers when they announced this month that application developers could beta their new application (if written in Python) using Google’s internal resources, their massive server cluster.
When our new GigaCenter is completed later this year, RackForce will be incorporating the very latest in virtual data center technologies. We have led the way in offering virtual services to date with our Dynamic Dedicated Server (DDS) server strategy (easy, worry free, work free upgrades as you grow), the first to offer Windows virtual servers using Microsoft’s server virtualization software (VS2005r2) and RackForce promises you we will continue to be a leader in the future as well . We see virtualization as the future of computing and we will be pushing forward strongly towards complete virtualization.
RackForce’s staff and principals take the state of the environment very seriously. British Columbia, Canada, is truly one of the most beautiful regions in the world. From the Pacific coast through to the Rocky Mountains, BC is the place for outdoor adventure and many of RackForce’s staff are avid outdoor adventurers. We are grateful for the opportunity to live and work here and thus are committed to preserving what we have.
British Columbia has been a crucible of environmental activism and change for many decades now. Indeed one could even make the case that the whole modern environmental movement really got its start here in 19711 when Greenpeace was formed to protest the Amchitka underground nuclear tests in the Aleutian Islands off of Alaska. Add to that the many protests to save virgin coastal rain forests such as the Clayoquot Wilderness Area and Haida Gwaii, and you can understand why we take global climate change seriously here.
The above is to underscore why we at RackForce take the environment, and what we can, and should, do for it seriously (see the previous two posts on Gigacentre and datacenter green technology for a background.) Following is some of what is happening in BC recently on the carbon front and how it applies to RackForce.
The BC provincial government just this past month brought in a revenue neutral carbon tax on fuels that will gradually ramp up over the next 4 years. Last year BC Premier Gordon Campbell was the only Canadian representation at the ICAP summit in Lisbon where BC joined many European countries and US states in signing on to the International Carbon Action Partnership agreement.
(The Vancouver Sun) The carbon tax will apply to virtually all fossil fuels, including gasoline, diesel, natural gas, coal, propane, and home heating fuel. B.C.’s carbon tax, the provincial government claims, will be the most comprehensive in the world.
[snip]
the new carbon tax will begin July 1, starting at a rate that will have drivers paying about an extra 2.4 cents per litre of gasoline at the pumps.
[snip]
The tax will then increase each year after that until 2012, reaching a final price of about 7.2 cents per litre at the pumps.
This is certainly a start in the right direction. However, if it is to make a substantial difference in the province’s GHG emissions this needs to be applied across all industries and products, not just fuel products. RackForce, because it uses zero carbon hydroelectric power, will be minimally affected by this tax. As carbon taxes become more prevalent across North America and Europe data centers that rely on non-renewable energy sources are going to find themselves at a tremendous cost disadvantage. As this is an industry with very low margins it will not be surprising to see some of the commercial datacenters go out of business and as mentioned in a previous article it will make ever more sense for corporations running their own in house datacenters to seek out companies zero carbon datacenters, like RackForce, to do their hosting and/or server co-location for them so as to not only avoid those tax penalties but in many cases be able to claim green carbon credits for doing so (if they are involved in cap and trade as well.)
Carbon taxes, because they are not open to the types of abuse and cheating (at the corporate level) that cap and trade is (see video below), are ultimately better than cap and trade systems, as long as they are mostly revenue neutral (and there’s the rub) for the government. That portion of that tax revenue that is not returned to the general public, in the form of other lowered taxes, must be allocated to support environmental research whose goal is to develop technology that further lowers GHG (Green House Gas) emissions. Doing anything else with those taxes would most likely be viewed by the voters as just a tax grab. By making GHG emitting products more expensive and then rebating that increased revenue stream through separate tax reductions, the overall cost to most members of society will remain approximately the same while making the perceived cost of using inefficient GHG producers much higher. This will cause consumers to look for ever lower GHG producing products in order to reduce their tax burden. That search for lower GHG producing products (which therefor carry a lower tax burden) combined with government investment, of some of the carbon tax generated, into GHG technology research companies, is a great economic development incentive for new and existing companies in that area of research, within the political boundaries of the tax collecting entity. The net effect of carbon taxes (when done with appropriate tax penalties) is the reduction of GHG whereas that is by no means certain with cap and trade.
Here’s an interesting look at the relative values of Cap and Trade vs. Carbon Tax
Or, as two former senior BC provincial civil servants (Bruce McRae Assistant Deputy Minister Forests and Energy ministries, and Don Wright Deputy Minister Forest and Education ministries and Secretary to the Treasury) , said last year in a report .
Carbon taxes are the most effective approach because they can apply equally to consumers, businesses and industries, and serve as incentives for all those groups to reduce their energy consumption”The market will go to work. The cost of energy will rise, which will provide businesses and households with an incentive to consume less and demand more fuel-efficient vehicles, equipment, appliances and buildings. Businesses will pursue technologies that result in less greenhouse gas emissions.” [ed. emphasis mine]
that is exactly what RackForce is doing, pursuing a course of business using technologies that reduce our carbon footprint to as close to zero carbon as possible. Join us in helping our planet!
1 well it really got its start with the publication in 1962 of Rachel Carson’s seminal work Silent Spring but the movement truly took off when Greenpeace was formed and generated international publicity through its confrontation with authorities and corporations over bad environmental practices.
In the last post (Gigacentre: Where we are going) I talked a bit about green technologies for data centers. The fact is everybody is talking “green” but very few are actually being proactive about it. The importance of Data Centers going green can not be overstated, and not just for the obvious positive environmental effects . As more and more companies begin to report their greenhouse gas emissions as part of their carbon footprint focus will be drawn to the large role their IT infrastructure plays in that reporting. Offloading corporate IT infrastructure to green, zero carbon data centers will be suddenly be very attractive to corporate Accounting, PR departments and CTOs (far less physical infrastructure to manage.)
In a February 21, 2008 interview with a major finance magazine (not yet published), Tim Dufour, RackForce’s company president, said:
the old ways of IT, especially for the mid-market enterprise (less than 1000 employees), are changing. In the mid-market you often see multiple branch offices with small numbers of inefficient servers operating in back-room closets with inadequate cooling and obsolete UPS electrical systems. By contrast, today you see more and more mid-market companies centralizing IT infrastructure and many are outsourcing to larger more efficient data centers, which might be the best way to make their IT green. The customer needs to be savvy though, as we see most data center infrastructure providers claiming to be green, even when they are using coal-fired generation as their power source, their data centers are 10 to 20 years old with inefficient cooling/UPS and simply are not designed for today’s server density.
Here’s the reality check for you the datacenter customer. If your current data center is operating like the ones Tim just described then they really aren’t green. What they are doing is using some form of carbon offset trading to claim green status but as I explained in the last post that is not the same thing as zero carbon. To be green a data center really needs to be zero carbon, not carbon neutral 1.
If it were carbon neutral then no matter how much efficiency we achieve in lowering power usage per server, the resulting cost savings, when passed on to the consumer, will result in even more servers being used and thus more GHG being produced.
For datacenters zero carbon is greener because it causes an overall reduction in GHG through the centralization, virtualization and concentration of server usage and resources in one place, rather than in multiple inefficient data centers, thereby allowing those inefficient datacenters to be closed or scaled down.
Enter Gigacentre, the future of green datacenters; designed from the ground up with the latest in green technologies:
modular and scalable
efficient
free cooling
chilled water to the rack
enclosed hot and cold aisle design
Virtualization
The single biggest factor in costs for a data center is the cost of power and the greatest use of that power is in cooling the tremendous amount of heat the servers produce. Cost for powering the servers and HVAC is nearing, or exceeding, the cost of the servers and hardware themselves.
IDC estimates for every $1.00 spent on new data center hardware, an additional $0.50 is spent on power and cooling, more than double the amount of five years ago. According to Gartner, 70 percent of CIO’s are reporting that power and/or cooling issues are now their single largest problem in the data center. Gartner estimates that 50 percent of data centers in 2008 will have insufficient power and cooling capacity to meet demand with 48 percent of the data center budget being spent on energy, up from 8 percent a few years ago.(More at blade.org)
Here’s how that power gets used in a traditional datacenter. As you can see the majority is not spent on actually running the actual servers but on cooling them and providing back up power and power conditioning. That 45% (Chiller, CRAC, Humidifier) spent on cooling is where the biggest opportunity exists to save electricity.
Let’s take a closer look at some of the technologies I mentioned earlier and see how they can assist in lowering power usage, both at the rack and at the HVAC level.
Green Data Center Technologies
Modular and Scalable
One problem that confronts datacenters is what do you do with all the open space, the room you have left to grow into? It presents heating/cooling/airflow problems for the parts of the room you are currently using.
Efficient
Efficiency takes on many forms in a data center, from automatically turning overhead lights off when no one is present to running cabling through proper channels to ensure air flow is not interrupted.
While we all understand the importance of a quality infrastructure to support and carry data traffic, there are other areas in the data center where cabling may be hurting your environment. In particular, your cooling capabilities and the degradation of connections over time. The first is rapidly becoming a cost drain. Older cooling units and even the latest and greatest cooling units will suffer if they can not move air into the desired locations. The effect of abandoned cable under a data center floor is an air damn [sic].
Efficiency is a function of proper design of the physical plant which in turn influences work flow and work habits. Gigacentre is being designed with all of this firmly in mind.
“Free Cooling“
While cooling is never free there are ways to dramatically lower, the cost of it by upwards of 50%, and additionally reduce significant amounts of usage. This is accomplished by using IBM’s revolutionary Stored Cooling System
This stored cooling solution is designed to optimize the efficiency of equipment that is often overprovisioned and running at low utilization and efficiency. It provides a turnkey solution that is designed to be maintenance-free, requiring only an “oil change” to replace the phase change material once every 25 years. By shifting energy consumption for cooling to off-peak hours when utility rates are lower, it helps reduce energy cost.
During the night, when power rates are down, the system refrigerates a large mass of a cooling gel which is used during the day, when power rates are high, to provide the cold air for the chillers, in essence it is a massive cold battery. This allows data centers to maximize the efficiency of their existing chillers thereby reducing the need to overbuild chiller capacity to meet intermittent peak usage times, which is not only a saving in infrastructure investment but also an additional lowering of potential electrical usage and thus an environmental benefit as it lowers the load on the power grid.
Blade Servers
The bigger the physical footprint the bigger the carbon footprint. Datacenters striving for zero carbon need to use every tool at their disposal for reducing that physical footprint. One such tool is blade servers. For example, from an April 7, 2007 press release by IBM
Microprocessors can account for a sizable portion of the power used by a server. The new systems introduced by IBM today are based on low-voltage industry standard processors that provide the same application performance as their higher wattage cousins, but in some cases consume less power.To put this in perspective, consider that for every kilowatt of electricity consumed, on average over a pound of CO2 is released into the environment. For example, with the new low-voltage, quad-core Intel-based blades introduced today, businesses can save up to 60 watts of energy per two-socket blade server and in an enterprise environment with 1000 blade servers can prevent the release of nearly 20,000 pounds of CO2 into the atmosphere over a year. That is the equivalent of the amount of CO2 produced by an air-traveler flying in a passenger jet round-trip from New York to London seven times.
Gigacentre will be using IBM X series blade servers.
Blades take up much less room than traditional servers. In a standard 42u server rack you can get up to a 45 percent density improvement when compared to standard rack mounted and non rack mounted stand alone servers.
Virtualization
Virtualization is the big one. Everything else helps but this technology, especially with the advent of Hyper-V, really reduces the need for space and hardware. By unifying multiple servers through virtualization infrastructure is maximized. The use of multiple virtual environments on the same hardware node allows for easier mounting, operation and maintenance of multiple OS as each environment is separately bootable. because each physical server can now do the work of several servers the need for hardware and the use of electricity is lowered. As you can see, virtualization combined with blade servers leads to a substantial positive impact on the environment.
Chilled Water to the Rack
Pre-cooling server racks with chilled water is more effective than with air. Water is a much better conductor of heat than air is. The resulting hot water can be cycled through heat exchangers or used to heat parts of the building that do not contain server racks. See IBM’s IBM Rear Door Heat eXchanger which can remove 55% of the heat (55,000BTU/15,000Watts) coming off the rack. That’s heat that won’t require additional A/C chillers to cool down. Removing heat at the source is far more effective than letting it escape then trying to remove it from a much larger mass of air via A/C chillers.
Enclosed Hot and Cold Aisle Design
Crucial to the efficiency, and thus the cost and carbon footprint, of a datacenter’s HVAC systems is the layout of its floor space. The use of hot and cold aisles allows an operator to separate the hot air and cold air flows for greater HVAC efficiency and, as mentioned earlier, by running your cabling through proper cable channels, only under (and/or above) the hot aisles you decrease cold air flow interruptions and thus improve cooling efficiency even further. There is an excellent article on this here.
These are just some of the technologies that will go into making Gigacentre one of the greenest datacenters in the world. Stay tuned for more news
Add the RackForce blog to your favorite news aggregator today - click here
1 one of the major problems with carbon offsets is that there is no way to accurately show that it really is being offset. There has been a lot of fraud going on with carbon trading where the money etc does not go into real renewable energy projects but instead into false front operations that appear to be doing renewable energy projects. Other problems arise when companies claim carbon credits for carbon taxes they pay. The whole carbon trading scheme is so full of holes that it can not be relied upon. See
Well the cat is out of the bag it seems with this article in CIO magazine and this one in the Kelowna Capital News. Yup, RackForce, along with some yet to be disclosed partners, (while RackForce is an IBM Business Partner, and IBM is involved in the planning and design of Gigacentre, the articles were wrong in identifying IBM as an investment partner in this venture) is building Gigacentre , a state-of-the-art super green data center in Kelowna, BC, Canada. When completed Gigacentre will be one of the largest, data centers in North America and we are working to make it the greenest data center that current and cutting edge technologies will permit. While the articles linked to above state that the storage capacity will be 35 Petabytes (35,000TB) and the square footage to be 80,000 that is very misleading as what’s described:
is only the first phase
implies that this center is only for data storage, when it is also for Disaster Recovery, Dedicated Servers, Virtual Servers, Remote Backup Services, and much more.
With the advent of global climate change the efficient operation of data centers has become much more critical. Researchers are starting to become aware that ICT (Information, Computer, and Telecommunications) technologies are quickly becoming major sources of global Green House Gas (GHG) emissions. When you consider the entire production cycle for those technologies they could overtake most other industries in GHG production in the next few decades (see Bill St. Arnaud’s excellent discussion document at http://docs.google.com/View?docid=dgbgjrct_2767dxpbdvcf.)
Due to the massive amounts of electricity needed to operate them Data Centers are becoming major sources of GHG. NANOG (North American Network Operators Group) recently did a study (25 page .pdf) that discusses data center power trends. Here’s a factoid from that study, a 1Megawatt data center uses 17.7 million Kilowatt hours of energy per year and that is a very small data center. Recently, EYP Mission Critical analyzed a series of datacenters and found that 50% of power usage was consumed by the IT equipment (or the critical load), while the other 50% was used by cooling, electrical system inefficiencies and trivial requirements such as lighting. For example, a small datacenter with a 4MW feed would actually consume 2MW of that power before any IT equipment is accounted for. Every step a data center can take to reduce both critical and non-critical load will save on GHG and money.
Focusing on power consumption is only the first step in making a green data center as carbon neutral (GHG generated offset by GHG saved) and zero carbon (no GHG generated) as possible. It is the most important and cost efficient vector to begin with. Gigacentre will primarily use hydroelectric power. Hydro does not generate any GHG in its production so the vast majority of Gigacentre’s power consumption will be zero carbon. This is important because as mentioned in the Bill St. Arnaud article, the Khazzoom-Brookes postulate shows that energy efficiency at the micro level leads to greater energy consumption at the macro level by bringing the cost of using that energy down and thereby encouraging greater use of it.
You can see the importance of Gigacentre being primarily zero carbon when you consider the implications of the Khazzoom-Brookes postulate. If it were carbon neutral then no matter how much efficiency we achieve in lowering power usage per server, the resulting cost savings, when passed on to the consumer, will result in even more servers being used and thus more GHG being produced. Of course Gigacentre will also use the very latest in server hardware technologies to reduce the amount of energy required as energy efficiency is important for other reasons, specifically reducing the need for more hydroelectric projects (and the concomitant environmental damage that creates) to meet the increased demand. More importantly virtualization will be used extensively to achieve much of this energy savings at the server level. I will be expanding more on further technologies such as:
Totally self enclosed, pre-engineered modules
Enclosed hot and cold aisle design
Stored cooling program to reduce active chiller time
Deployment of IBM’s Stored Cooling can reduce energy costs associated with the cooling plant by up to 40%.
Cable optimization to reduce impact on hot and cold air flow
Thermal analysisof target location to ensure proper placement of equipment for maximum thermal efficiency
that are going to be used in Gigacentre to achieve zero carbon status shortly in a follow up post.
Virtualization is key to energy efficiency. It reduces the total number of servers needed to accomplish the same tasks. To quote from St. Arnaud’s paper again
virtualization combined with zero carbon data centers and optical networks are enabling the possibility of the industry to adopt a zero carbon strategy. In a recent New Scientist article ” Can we stop the Internet destroying our planet?” brings this all together by pointing out that virtualization is now seen as the low-hanging fruit in data centres’ green transition.”[13] Virtualization, as the word implies, allows physical servers to be removed and replaced by virtual servers in data centers.
Relocating physical servers to virtual facilities in data centers is the first step. This will at least, minimize the carbon footprint of the many millions of servers in our corporations and businesses. Next we need to convert the data centers to use 100% renewable energy and become zero carbon data centers.
Companies that shift their server needs off their premises and into energy zero carbon data centers will see a reduction in their IT costs and be able to claim carbon credits with their government carbon tax programs for the carbon savings they make by doing so. Win/Win/Win
Over the years I have had many customers ask me why Kelowna? Well there are excellent reasons for that.
Kelowna has an abundance of green hydro electric energy and two providers, BC Hydro and Fortis, of that energy.
Kelowna has a burgeoning hi tech industry that has attracted a lot of exceptional talent to the city and to the Okanagan, the surrounding area.
Kelowna airport has been expanded this year to accept international long haul flights from outside of North America and the 3 hour drive to Vancouver, BC’s major city on the coast , is via a 4 lane highway.
Wine, did I mention wine? The Okanagan is not known as the Silicon Vineyard for nothing. The area abounds with local wineries, from farmgate operations to major producers and everything in between. If you haven’t tried BC Wines you are missing out on something very special. Just like the California high tech industry is situated around the Napa Valley and other wine regions so too is BC’s high tech industry situated around BC’s wine growing regions.
So now you know. RackForce is growing, expanding its scope of operation and about to bring you some very new products (see last article on Hyper-V as an example) in an exciting new data center. Stay tuned - much more great news coming
RackForce is considered by many, including Microsoft, to be “the experts” in hosted virtualized server technologies. We launched the world’s first hosted Microsoft Virtual Server in the Fall of 2005 and have been perfecting the virtualized server model through our unique Dynamic Dedicated Server (dds) strategy since 2003. In 2008 we see more and more in our industry, including Rackspace that are realizing the power of virtualization and are beginning to launch their own virtualized server strategies.”
Virtualization is the future of server technology. There are a lot of reasons for this, among them are that it makes economic sense to consolidate multiple underutilized hardware resources (servers) onto a single server running multiple virtual environments. It also make environmental sense as the newer hardware uses less power and creates less heat than any single one of the old servers did. By fully utilizing the resources on each server there are big savings on both counts. See what one company achieved by consolidation and virtualization.
Virtualization in its original form created multiple virtual environments on the same piece of hardware. While it was a great step forward the next step is truly exciting as it transcends the physical limitations of any single hardware platform. Imagine a world where when you rent your “dedicated” server what you are really renting is access to “virtually” unlimited resources. A world where, at the click of a mouse, you can add more storage, RAM, or processing power. This is the world that Microsoft’s new Hyper-V hypervisor technology is taking us to. Already out in beta release Hyper-V is generating a lot of excitement around the industry and it fits very well with the holistic vision of the virtualized datacenter that we have here at RackForce. Let’s have a peek at what it brings to the table:
Virtualize multiple operating systems—Windows, Linux, and others—on a single server and fully leverage the power of x64 computing. With virtualization built into the operating system and with simpler, more flexible licensing policies, it’s now easier than ever to take advantage of all the benefits and cost savings of virtualization. Windows Server 2008 provides you with the flexibility to create an agile and dynamic datacenter to meet your changing business needs.
Hyper-V enables four core scenarios: Server consolidation, business continuity, test and development, and dynamic datacenter. Combined with Microsoft System Center management suite, you can now have a complete and integrated server management solution that works with virtual machines and physical servers and helps extend the platform capabilities of Hyper-V.
Server consolidation - One of the leading drivers for adoption of virtualization technology is server consolidation. Businesses are under pressure to ease management and reduce costs while retaining and enhancing competitive advantages, such as flexibility, reliability, scalability, and security. The fundamental use of virtualization to help consolidate many servers on a single system while maintaining isolation helps address these demands. One of the main benefits of server consolidation is lower total cost of ownership (TCO), not just from lowering hardware requirements but also from power and cooling costs and management costs. Another benefit is from optimizing infrastructure, both from an asset utilization standpoint as well as the ability to balance workloads across different resources. A softer benefit of server consolidation is improved flexibility of the overall environment and the ability to freely integrate 32-bit and 64-bit workloads in the same environment.
Business continuity and disaster recovery - Business continuity is the ability to minimize both scheduled and unscheduled downtime. That includes time lost to routine functions, such as maintenance and backup, as well as unanticipated outages. Hyper-V includes powerful business continuity features such as live backup and quick migration, enabling businesses to meet stringent uptime and response metrics.Disaster recovery is a key component of business continuity. Natural disasters, malicious attacks, and even simple configuration problems like software conflicts can cripple services and applications until administrators resolve the problems and restore any backed up data. Leveraging the clustering capabilities of Windows Server 2008, Hyper-V now provides support for disaster recovery (DR) within IT environments and across data centers using geographically dispersed clustering capabilities. Rapid and reliable disaster and business recovery helps ensure minimal data loss and powerful remote management capabilities.
Test and development - Test and development are frequently the first business functions to take advantage of virtualization technology. Using virtual machines, development staffs can create and test a wide variety of scenarios in a safe, self-contained environment that accurately approximates the operation of physical servers and clients. Hyper-V maximizes utilization of test hardware, reducing costs, improving life cycle management, and improving test coverage. With extensive guest OS support and checkpoint features, Hyper-V provides a great platform for your test and development environments.
Dynamic data center - Hyper-V, together with your existing system management solutions, such as Microsoft System Center, helps realize the dynamic data center vision of providing self-managing dynamic systems and operational agility. With features like flexible resource control and quick migration, you can create a dynamic IT environment that leverages virtualization to not only respond to problems, but to anticipate increased demands.
Of course it is the last point that I was referring to earlier.
New and improved architecture: New 64-bit micro-kernelized hypervisor architecture enables Hyper-V to provide a broad array of device support and improved performance and security.
Broad OS support: Broad support for simultaneously running different types of operating systems, including 32-bit and 64-bit systems across different server platforms, such as Windows and Linux.
SMP support: Ability to support up to 4 multiple processors (SMP) in a virtual machine environment to enable you to take full advantage of multi-threaded applications in a virtual machine.
Memory support: Support for large memory allocation per virtual machine enables you to now virtualize most workloads, making Hyper-V an ideal platform for both enterprises and small to medium businesses alike
Improved storage access: With pass-through disk access and broad support for SAN and internal disk access, Hyper-V provides greater flexibility in configuring and utilizing storage environments optimally.
Network load balancing: With the new virtual service provider/virtual service client (VSP/VSC) architecture, Hyper-V provides improved performance and utilization of core resources such as disk, networking, video, etc.
New hardware sharing architecture: With the new virtual service provider/virtual service client (VSP/VSC) architecture, Hyper-V provides improved access and utilization of core resources such as disk, networking, video, etc.
Quick migration: Hyper-V enables you to rapidly migrate a running virtual machine from one physical host system to another with minimal downtime, leveraging familiar high-availability capabilities of Windows Server and System Center management tools.
Linux integration components: Beta Linux integration components are now available for SUSE Linux Enterprise Server 10 SP1 x86 and x64 Editions. These integration components enable Xen-enabled Linux to take advantage of the VSP/VSC architecture and provide improved performance. Beta Linux Integration components are available for immediate download through connect.microsoft.com.
Virtual machine snapshot: Hyper-V provides the ability to take snapshots of a running virtual machine so you can easily revert to a previous state and improve the overall backup and recoverability solution.
Scalability: With support for multiple processors and cores at the host level and improved memory access within virtual machines, you can now vertically scale your virtualization environment to support a large number of virtual machines within a given host and continue to leverage quick migration for scalability across multiple hosts.
Extensible: Standards-based Windows Management Instrumentation (WMI) interfaces and APIs in Hyper-V enable independent software vendors and developers to quickly build custom tools, utilities, and enhancements for the virtualization platform.
Now if that doesn’t get you excited check your pulse
We have been working behind the scenes with Hyper-V as a member of the Microsoft Technology Adoption Program (TAP), and we plan to be one of the first in the world will be launching Hyper-V here as soon as possible.to bring you this exciting platform. Stay tuned!
Points to Consider When Choosing a Data Center For Your Business
Virtualization
Virtual servers are often more powerful and they do not limit your scalability. Virtual servers offer you a way to minimize your costs while maximizing your flexibility. Does your data center provider offer virtual technology and what is their expertise in it?
Backup
Does your business depend on having your data backed up? Most hosts do not have a backup strategy yet online companies are completely dependent on their e-commerce data for their survival. Does your host have a backup strategy? If they do do you have any control over it? Is it an Enterprise class system? How important is the ability to restore individual files to you? Is the ability to have a dozen or more different copies of the same file important to you so that you can restore from a specific point in time instead of being forced to restore from just one point?
Bandwidth
Does your host guarantee you will be able to do a specific amount of data transfer per month (bandwidth) or do they just throw a high number at you and tell you if you go over that you’ll get charged extra? If the latter then you are definitely on a shared and over subscribed connection and your chances of doing that 1000GB or 2000GB per month are really quite slim, but it sure looks good in the ads doesn’t it? Make certain, wherever possible, your host offers you a dedicated rate of connection, such as 10Mbps. When your connection rate is guaranteed you always know what you can rely on.
24×7 Support
Who does support for your host? Do they outsource their support to someone not on site? Do they own their own facilities and thus can lay hands directly on the hardware if needed? How knowledgeable are the support staff? Can you reach them by phone if necessary? How long is the average response time to a trouble ticket?
Server Quality
Old servers are risky. What brand names do they have? Are they real servers or desktop computers re-purposed as a server (if it wasn’t designed as a server it shouldn’t be used as one - server usage puts more strain on the hardware than desktop usage does?) Is the hardware technology old (P4, Celeron etc?) Do the servers have hot swappable hard drives? Are they using unreliable software RAID instead of hardware RAID? How long has the hardware been in use?
Data Center Management Systems
How do they manage so many variables
Future Upgrade and Migration Paths
When you get bigger what do they offer? Can you be easily moved to bigger and better hardware requirements? Can your data center take you to where you need to go? Do they offer multiple server clusters with load balancing for example?
Emergency Growth Plan
What if you need to grow right now? What can they do for you? Do they have a set of plans in place that allow you to move, with ease and minimal downtime, from the smallest VPS to the largest Dedicated Server?
Green
It is becoming ever more necessary for businesses to show that not only are they green but that their suppliers, such as data centers, are also green”. How Green is your data center provider? Where do they get their power from? Is it from Hydro-electric which is green or is it from coal fired plants? Do they invest in power saving technology? Are they promoting server virtualization which is very green as it it fully utilizes the servers resources?
Sales Staff
Are there any? Do they know what they are talking about? If they don’t know something do they offer to find out for you or do they just tell you what they think you want to hear?
There has been substantial concern in Canada from consumers who are paying 15 to 40% more, across the board, than their US counterparts even though the Canadian dollar has been hovering close to par for several months now. Canadian businesses need to see and react to these forces before they are forced to. For example, just last month some Canadian consumers launched a class action lawsuit against Canadian auto manufacturers because of the tremendous price differential between buying a car in Canada and in the US and the blocks put in place by those manufacturers US counterparts to prevent Canadian consumers purchasing cars in the US. There will likely be many more cases like this in the near future, particularly if the US dollar continues its downward momentum.
One way that Canadian companies can compete is to lower their day to day operating costs by making substantial investments in cutting edge technology that gives them both a technological edge over their competition (thereby making their services more attractive) and at the same time lowers their day to day operating costs through greater efficiency.
In the data center industry this is achieved by moving to more energy efficient servers like IBM’s Quad core servers, and also moving customers towards virtualization. Virtualization allows both customers and providers to lower their costs by making more efficient use of server hardware, power and HVAC requirements. As a bonus this makes for a greener business model as well.
This is the direction that RackForce has been going in for quite some time now. The future of server technology and utilization lays in virtual services. On the Windows side we were selected by Microsoft to work with them to bring their virtual server technology (VS2005r2) to the public as a data center application, resulting in our Windows VPS servers being a stunning success. On the Linux side we are a Platinum Partner with SW-Soft, and were one of the very first companies to offer their industry leading Virtuozzo virtualization product.
RackForce moved its pricing to parity between the US and Canadian dollar on Thursday September 25th.
If you own a business there are at least two plans you must complete if you want that business to be successful, and to last. The first is your business plan, which lays out your product(s), your expected market, your risks, and your projected costs and revenues. Without such a plan your chances of success are slim and your chances of attracting financing are none. The second plan you should have in place is a Business Continuity Plan (BCP).
A BCP is needed so that your business can recover and continue operations in the event of a disaster, be it natural or man made. Consider what happened on 9/11 when the twin towers went down. Those towers contained a lot of businesses in the financial services sector (banks, brokerages, insurance companies) and in some cases they were able to restart operations in a very short time because they had a well thought out plan. In other cases they simply went out of business or suffered a sever impact on their business because they didn’t because key people and data were lost and they had no plan in place for dealing with that.
Floods
Hurricanes
Global Warming
Water Shortage
Pandemic
Tornadoes
Earthquakes
Brownouts or Blackouts
Sudden loss of key personnel
Terrorism
Loss of a key supplier
Labor Disputes
All of the above and many more are factors that can severely affect your business’ ability to carry on.
Marsh Canada (company provides risk and insurance services, investment management and consulting services) recently did a survey of Fortune 1000 companies in regards top risk assessment and planning. The results are quite interesting as you can see in this graph
(click on image for a larger image)
Risk assessment for the purpose of business continuity must, as you see, take into account a lot of factors many of which are beyond your ability to control and many you probably haven’t considered.
One of the things you can control, however, is your IT recovery strategy. When Hurricane Katrina hit the gulf coast in August 2005 many companies were caught unprepared. Many had their electronic data stored locally and lost all of it. Other companies not located in the gulf region had their data backed up to servers in that region and they too suffered as the power went out and they could not back up their data anymore. Some are still trying to recover from that. Planning to manage all emergencies is essential for business continuity.
Here are a few links to help you with continuity planning:
One of the key factors for any Internet enabled business is how to safeguard its data. Key to that is a well thought out data backup plan. The first thing to remember in this strategy is to back up to a separate location and choose that location well. Wherever it is, it should be safe from natural disasters and, unfortunately these days, potential terrorist activity.Choose your data center well is a key operative concept. Where you place your backups, and your servers if you are outsourcing your infrastructure, is of critical importance. Is you data center subject to brownouts or blackouts? Does it sit in a hurricane zone? Is it located on the west coast inside the San Andreas fault zone? If you answered yes to any of those you should reconsider where you have your data located. When disaster strikes you need to be able to start your business again, if not seamlessly then as quickly as possible. You can’t wait on your data center to recover so that you can recover.
Also key is the type of backups you do. Do them often and choose a company that has an Enterprise class backup system, such as Tivoli Storage Manager from IBM, as anything else is playing dice with your company’s data. RackForce now offers Tivoli Storage Manager.
RackForce operates Tier 3 data centers in Kelowna BC, Canada that are secure from the natural disasters and their consequences as mentioned above. If you are interested in making this, and/or remotely hosted servers, part of your Business Continuity Plan please contact our sales staff and we will be glad to discuss your needs with you.
Tivoli Storage Manager from IBM is now available from RackForce on all Windows servers, VPS and Dedicated, and fully dedicated Linux servers (5000, 5100, 5300 and 5350). To check it out click here
RackForce is one of the only server infrastructure providers in North America to provide an enterprise level backup system like IBM’s TSM. IBM Tivoli® Storage Manager is insurance for your data. If a disk drive fails, Storage Manager restores your data.
RackForce is joining the select crowd of service providers who offer an enterprise backup solution to meet its customers enterprise backup storage needs. We chose the IBM Tivoli Storage Manager on our IBM storage area network (SAN) using an N5200 SAN controller with 4700 disk arrays capable of over 50 Terabytes of storage as the best solution for all of our customers. IBM TSM will allow our customers to backup their Virtual Private Servers, Dedicated Servers, and Dynamic Dedicated Servers to tape or disk and retrieve those backups when they need them, from a complete backup restore to individual file restores. This next step allows RackForce to provide enterprise backup abilities and to be a disaster recovery center.
The following is info from IBM’s site about their Tivoli Storage Manager.
The 24×7 availability of mission-critical data and applications is no longer a goal—it’s a business necessity. The IBM Tivoli Storage Manager is designed to provide centralized, automated data protection that can help reduce the risks associated with data loss while also helping to reduce complexity and address compliance with regulatory data retention requirements.
IBM Tivoli Storage Manager enables you to protect your organization’s data from failures and other errors by storing backup, archive, space management and bare-metal restore data, as well as compliance and disaster-recovery data in a hierarchy of offline storage. Because it is highly scalable, Tivoli Storage Manager can help protect computers running a variety of different operating systems, connected together through storage area networks (SANs). It uses Web-based management, intelligent data move-and-store techniques and comprehensive policy-based automation.
IBM Tivoli Storage Manager’s core functions include:
~ Backup and Recovery Management and Archive Management ~ Disaster Preparation Planning and Recovery ~ NDMP Backup for Network Attached Storage ~ Small and Large Tape Libraries ~ Compliance with the most stringent regulatory requirements ~ Invisible, real-time file replication protection
Attributes that set Tivoli Storage Manager apart:
~ Easy Management of Multiple Types of Inactive Data in a Hierarchical Repository ~ Lower Storage Cost through Intelligent Hierarchy of Storage ~ Centralized, Comprehensive Management ~ Reduced Network Bandwidth through Intelligent Data Movement ~ Policy-Based Automation
For more on TSM features, please contact RackForce directly and our customer experience team will be happy to assist you.
Phone 1-866-468-1158 or +1-250-717-2340 www.rackforce.com
Blog Home
Get RSS Feed
