Archive for September, 2007
September 30, 2007 at 11:01 pm · Filed under virtualization, vmware
http://blogs.zdnet.com/virtualization/?p=245
When I’ve had a chance to speak with executives of suppliers of virtual machine software, they often point out that one of the major benefits of deploying application environments inside of virtual machines is simplicity. That is, they suggest that using virtual machine software reduces complexity in the environment.
While that might be true from the perspective of that single application, it may not be true of the whole environment. Installation, migration from one physical machine to another and even some application management tasks can be facilitated by the use of virtual machine technology.
Installation may change from an arduous task to merely copying a file. Migration of an application environment from one physical system to another also become a file transfer.
Check out this interesting article. Does virtualization reduce the complexity of the infrastructure? Well not always, but then in some cases there are only so many changes to the infrastructure, not to mention the billing and the way we do business that can be achieved without examining the actual business, the business model.
Virtualization as with any technical (or non-technical) solution will only work as well as your teams, and your business can implement and adopt to it. If you’re billing model isn’t set up in the right way I might have more virtual machines but have the ESX hosts tied to a department or an application team, making management more complex, but equally I might find that the way I can roll back changes, or allocate more disk, more cpu or more memory to a session without going through the normal evaluation, project initiation and sign off stages, by just rebooting my virtual machine could be a real enabler. Work out how your business works with your IT, what the show stoppers are, and implement a virtualization platform which can either work around these constraints, the goal is not to make things less complex, the goal is to make provisioning, support and the lifecycle maybe not easier, but more manageable, more aligned to the business need.
September 30, 2007 at 10:50 pm · Filed under Other things, blades
http://sixty94956.blogspot.com/2007/09/blade-servers-and-how-they-effect-your.html
Welcome to Blade Server 101, where we will teach you the ins and outs of what a blade server is and how its use can lower the total cost of ownership (TCO) for any business.
First of all, what is a blade server? The best way to answer that question is to compare it to a traditional low-profile dual-processor 1U (one rack unit) rack-mount server. An Intel blade server chassis houses up to 14 dual Xeon servers or 7 Quad Xeon Servers in just 7U of data center rack space.
Check out this interesting post which is talking about blade servers, its got some useful content and it’s always great to read how people explain their view on the technology. Blade servers can be a great solution for your business when deployed with the right tools to deliver an adaptive infrastructure, but as with anything, whether blade servers are right for you is going to depend on you and your business needs.
September 30, 2007 at 10:46 pm · Filed under DataSynapse, Grid
http://www.itpro.co.uk/news/126590/ubs-virtualises-application-infrastructure.html
Financial services firm, UBS has added flexibility and reduced the cost of running complex, computing intensive applications using application virtualisation.
The company dynamically allocates processing resource across its global IT infrastructure to support its bespoke risk management applications.
Andrew Morgan, UBS application architect and development manager said the company’s focus on complex financial services products that place high demand on the IT infrastructure led it to use application virtualisation products from DataSynapse.
Very cool, check out this article which is talking about how UBS has used DataSynapse as a vehicle for change in its application infrastructure, using grid and application virtualization to be be a business enabler, to reduce costs and improve flexibility, very cool.
September 27, 2007 at 8:48 pm · Filed under Other things
http://www.theregister.co.uk/2007/09/26/sun_x86_play/
The fresh Xeon-based servers from Sun mark the company’s most muscular play to date in the x86 realm. The four-socket 2U X4450, in particular, has Sun pushing the equivalent of a midrange SMP down into the heart of the so-called “commodity†market. This move plays well with Sun’s historic strengths in the SMP (symmetric multi-processing) arena and supports Sun’s “R&D will win†mantra.
Since entering the x86 market, Sun has more or less gone through two generations of hardware. Initially, it relied on systems designed by then Opteron-based server start-up Newisys. Sun tried to learn from its mistakes with this somewhat generic gear by later rolling out a fresh line of Opteron-based systems.
This interesting article from theregister is talking about Sun’s development of their servers in the x86 markets, and it seems quite positive. I agree, the more choice Sun offers in its server range, the better chance that a user will find a system or configuration that meets their business needs. It’s also helped with competition in these markets, continuing innovation has to be a good thing for the market and the end user, as well as bring new opportunities like IBM and Solaris.
September 27, 2007 at 8:45 pm · Filed under environment
http://www.vnunet.com/vnunet/news/2199788/dell-sets-carbon-neutral-plans
Dell has announced plans to neutralise the carbon impact of its worldwide operations.
“Never before in the history of business have we seen such a critical need to build a worldwide community dedicated to improving the environment,” said chairman Michael Dell at a policy forum organised by the Centre for Strategic and International Studies.
“Leadership starts at home, which is why we are going carbon neutral, but this should only be the beginning of building long-term partnerships with customers, stakeholders and suppliers to make a difference for the Earth we all share.”
Very cool, I wonder if Dell’s competitors will start down this path as well, but do we want to be heading for carbon neutrality as a business, or do we want to be owning the supply chain, ensuring that from commission to decommission (the life-cycle of the server), we do everything we can to minimize any impact on the environment through good design, effective configuration and implementation until recycling? We’ll see, well done to Dell for moving to be more environmentally considerate.
September 26, 2007 at 11:02 pm · Filed under Abn Amro merger
http://www.ft.com/cms/s/0/e710e112-6c6f-11dc-a0cf-0000779fd2ac.html
Royal Bank of Scotland has cleared one of the final remaining hurdles in its break-up bid for ABN Amro by raising more than €6bn (£4.2bn) through an issue of preference shares.
But in spite of strong interest from investors, the pricing of the issue, believed to be the largest ever by a bank, underlines how the cost of financing the deal has risen as a result of the turmoil in the capital markets.
Very interesting, we’ll need to see how this develops, and what this will mean to the deal, the shareholders and the employees of the banks’ involved, an interesting article from the FT, do check it out, if you’ve been following the deal.
September 26, 2007 at 10:52 pm · Filed under How IT works, Other things
One of my friends who works for one of the banks was having a chat with me, he was telling me about work, what was going well with their virtualization project, what wasn’t, what the key issues technical and non-technical, it was very interesting. Over dinner he then asked me what I thought about these kind of projects, where I thought IT got it wrong and it resulted in a very interesting conversation, the following comments/post you might say result from his question, “So in an ideal world, what is it you want?”
I want an IT that delivers. That goes the extra mile, that invests not because it has to but because it wants to add value to be the best. I want a desktop service where I can log into any pc, laptop, Wyse terminal in any location and have everything just work without a guy having to do something.
I want an IT where I can request a server and get it the same day, where the server is configured to my application with the right level of access permissions configured, which is all tracked from an asset/support level so that everyone knows what’s going on.
I want to be able to access the monitoring system to see how my application is doing, how the IT infrastructure is performing all online, I want my blackberry to give me an overview of what’s going on, to know what issues I’ve had overnight before I get to the office, to have an IT infrastructure that’s independent of the data center, that I can fail systems over to another office if the performance isn’t good enough, or if my NYC branch goes down to have it run from London.
I want an IT infrastructure where I can seamlessly request new storage or provision new systems by logging a call and get it that day - where I’m not asked questions about the type of storage, or which machine will run my virtual server.
I want a grid infrastructure that scales up and down on demand; that when my analytics batch kicks off, individual tasks can be sent to other application servers or desktops automatically, with resilience and on demand capacity in it’s true form, to avoid my batch running late at the end of the month due to volume of deals.
I want IT to “own the infrastructure”, to be testing the latest firmware or drivers to be saying to me, “HP have released 15/09/2004 firmware for the DL580G2’s, we’ve tested it and need to upgrade these servers..”, to have an asset overview, to know what systems are allocated to which business lines and are located in which data centers - a world where I as a business sponsor or an IT person can say “I want to know how many DL580G2’s we have, and what operating systems they run and who owns them” without waiting.
I want IT to have follow the sun support, that Mike doesn’t get called at 3am to reboot a server because Charlie is already on site, has the right level of access, can raise a break fix change and fully support the servers. That we work with the time zones not react to them - “Mike can you apply service pack 2 to the following systems after 10pm - 2pm your time?”
Most of all I want an IT team that understands the applications, that knows what effect “ping not responding on gaps983747″ will have on the derivatives team and can help resolve it or work our way around it, where they understand the application, the business criticality, what it means for the server to be down from a user perception, from an application performance and an infrastructure level.
But at the same time:
I want an engaged business team, one that understands you need to spend money to make (or save) money, that wants to continually evolve the application and the platform to get the best performance or functionality/reliability as well as to control our costs.
I want everything as much as possible abstracted from the platform through Web, through Citrix, where the client is just that client, where the application code has no specific requirements other than Windows or Java runtime, where the application can be moved from system to system or region to region. I want a lot of things, whether IT can do all of this, elements or components of this will depend on your business, how you achieve it will be dependent on your business, your constraints, be those technical or non-technical.
Any views?
September 26, 2007 at 10:19 pm · Filed under Other things, virtualization
http://home.businesswire.com/portal/site/google/index.jsp?ndmViewId=news_view&newsId=20070926005274&newsLang=en
CUPERTINO, Calif.–(BUSINESS WIRE)–Blue Lane Technologies today announced the October 5th general availability of VirtualShield 4.0, the first VM security solution to deliver real-time network flow visibility and security by protocol, operating system, application and VM cluster. The latest version of the award-winning VirtualShield features VirtualFlowâ„¢ Center console management, integration with the VMware VirtualCenter, template-based deployment of VM shields, and data center, cluster, host and VM-level vulnerability protection, policy and visibility into network flows.
Existing VirtualShield features also included in VirtualShield 4.0 include: dynamic cluster/VM risk reports; VM vulnerability detection/correction; VM operating system, application and port-based policy administration and dynamic content updates. Blue Lane is a VMware Technology Alliance Partner.
Great news! The article mentions the new features in version 4.0, securing the virtual infrastructure remains part of the due diligence process, at the same time introducing a tool that can help with the administration of the virtual machines from a security patching/exploit arena has to be a good thing, particularly as you scale your virtualization project to include those isolated networks/applications.
September 26, 2007 at 10:16 pm · Filed under Grid
http://www.macresearch.org/computing_grid_in_a_box_with_an_8_core_mac_pro
Of course an 8-core Mac Pro is nice for “mine is bigger than yours” taunts, however if you’re going to spend the money to buy a machine with 8-cores and 16 GB of RAM you better figure out how you are going to use all that horsepower. In a previous article I demonstrated how to use R + LAM/MPI to turn an 8-core Mac Pro (a personal machine I’ve dubbed the MacZilla) into a statistics supercomputer. In an effort to further justify the purchase of the MacZilla to the PI of our lab, I experimented with turning an 8-core Mac Pro into an all-in-one computing grid. To accomplish this I employed the VMWare Fusion virtualization software and Sun Grid Engine. I chose VMWare Fusion over Parallels since VMWare Fusion can fully utilize 16GB of RAM and plays very nice with 64-bit Operating Systems.
Very cool, check out this article, it’s an interesting post about grid using a Mac Pro. Interestingly, one of the thoughts I’d had within the grid space was about VMWare and for example DataSynapse.
Could I take that DL585 (or the equivalent), with my virtual sessions on it, (then this is where I’ll need to explain myself), put DataSynapse grid service on the ESX server and each of the virtual instances. Is this more efficient than putting it on a DL580G2 with Windows/Linux?
Probably not, but am I not getting more efficient use of the infrastructure? - You could argue I suppose that the amount of processing that each instance could do is limited, or that you’re paying a high price for it - I suppose it’s one of those things best thought off in concept and not practice.
September 26, 2007 at 10:11 pm · Filed under Consolidation, Other things, datacenter
http://www.infoworld.com/article/07/09/26/Ghost-servers-that-haunt-your-bottom-line_1.html
The problem may not rival the movies Poltergeist or The Amityville Horror for sheer terror, but CIOs and datacenter managers are still well advised to deal decisively with so-called ghost servers. Like celluloid zombies, these forgotten pieces of equipment are dead when it comes to improving the bottom line, but they are very much alive when it comes to eating up IT budgets.
The unproductive — and usually undocumented — servers take up valuable real estate, consume increasingly expensive electricity and, in some cases, absorb ongoing maintenance and lease payments.
“You can find ghost servers in a lot of enterprises,” says John Phelps, an analyst at Gartner. “And the larger and more diverse the company, the harder it can be to have a single group or technology platform that provides control over all corporate assets.”
Sun Microsystems, through case studies of two large corporate datacenter operations and anecdotal analysis of efforts with many customers, believes that 8 percent to 10 percent of all servers in large corporations have no identifiable function. In the two datacenter studies, 150 ghost servers were found in an installation of 1,800 servers, and 354 ghost servers were found in an installation of 3,500 servers.
As the server infrastructure grows (virtual or physical) managing the asset as well as everything around it gets more complicated. Take any large data centers with blades and rack servers, and I’ll bet you’ll find at least one server which is either not in use, or not actively doing anything. Taking your asset list, your server inventory, working out what’s doing which role and where it is can be a very effective way of freeing up data center space - that Compaq Proliant 1850 with 256MB ram running NT4 - you sure it’s being used? But keep in mind that sometimes those tier one servers can be the ones that no one has heard off - 68945 particularly if it’s never been rebooted or had an issue.
Next entries »
