Bladewatch.com

The accidental successful cio

Cloud computing is hot – there’s no denying that. However, as with all things in the information technology field, cloud computing isn’t standing still. Even as  you read these words, engineers are hard at work defining and refining just exactly what a cloud computing architecture looks like and how it behaves. Let’s take a peek at what the future holds…

Where The New Ideas Are Being Born

Although cloud computing research is going on in a number of small start-ups as well as at universities world-wide, the work that is being followed the closest is that which is being done by the very large firms. Here’s a quick run down of what they are doing:

• HP /Intel / Yahoo: These three powerhouses have come together to launch the Cloud Research Testbed. The goal is to allow academic researchers to have access to supercomputing resources in order to try out new ideas such as computing chips that have been designed for cloud computing.

• IBM Research: IBM has taken the global approach and launched itsResearch Compute Cloud. This cloud will be used to support business processes

An interesting article talking about cloud, it’s certainly something that’s getting a lot of the press, it’s the new virtualization if you like (so some of my colleagues have said). Regardless as the economics of business and IT change combined with the possibilities of the technology, the way we provision, charge for and support our IT needs to change with it. This will mean some significant changes in the enterprise for some, but interestingly I wonder if the roles (for the short time anyway) might stay the same, we still need the OS guys, the networks and the application teams, but how it’s consumed, how it’s paid for and who you work for, that’s where we might see more change.

Share and Enjoy

Infoworld

BAA is stepping up its IT restructuring programme by announcing plans a £10 million (US$16.5 million) IT consultancy framework.

Under a four year framework, around five suppliers are expected to provide a range of services. These include change and project management; design and transformation; strategy definition and implementation; people development; delivering IT; architecture; and IT in construction.

I wonder if we will see any particular technologies, processes or services as a result of this framework and restructuring programme? I wonder if we’ll see further moves to virtualization, to cloud computing and more efficient processes and information exchange? I’m off to see if I can find out more, very interesting.

Share and Enjoy

E commerce times

In today’s world of hybrid cars and wind farms, consuming energy wisely is on every individual’s agenda. Companies and their IT departments should be no different. Statistics aside, there is no doubt that collectively data centers have a huge “carbon footprint” and are the biggest or one of the biggest consumers of energy in every company. You have already heard about how it is only a matter of time until companies will be forced to make data centers green as a part of their effort to reduce their carbon penalty. If you think this issue is going away, think again.

How you plan your data center strategy in the short and long term will depend on your business and IT needs, but going forward we should be looking at not only the technologies, the air flow, the server, the storage and the network, but the business applications, the mapping of resources and applications to business processes and business needs:

• What is core to our business – what do we absolutely need online now and forever and what can we work without
• What level of availability is needed around our business – does our IT need 24/7 or 9-5?
• How can we be more efficient with the IT with the data center – what can we do to work with what we have, do all the servers need dual power supplies – can we virtualize or decommission – can we consolidate roles and outsource or buy in the services that aren’t core?
• What steps in the build, in the configuration of the servers can be taken to improve their efficiencies and reduce costs, support or in terms of energy efficiency.
• What steps in data center design for example in air flow and the distribution of cooling can be made to improve where we are – is the cooling efficiently distributed, are we cooling too much of one area and missing other parts?
  

Share and Enjoy

SGI

FREMONT, Calif. — August 6, 2009 — SGI® (NASDAQ: SGI) today announced the immediate availability of CloudRack™ X2, a new generation of scalable workgroup clusters for HPC, graphics and Internet applications. The newest addition to the CloudRack product family, CloudRack X2 delivers extreme density and 99 percent power distribution efficiency with Power XE™ in a small, 5-square-foot form factor. This enables a broad range of new deployment scenarios, including to workgroups, laboratories and traditional HPC environments either as a stand-alone unit mounted on casters for in-lab use or installed in industry-standard, 19-inch racks—especially valuable for heterogeneous computing environments.

CloudRack X2 represents the first time that CloudRack trays are available in HPC-dedicated configurations. The new system now supports high-performance processors and high-speed interconnects, including 40 Gb/sec InfiniBand®. As a result, typical HPC installations, including those that run computational structural mechanics, fluid dynamics, electromagnetics, seismic processing, rendering and visualization applications, can immediately capitalize on the many performance, density and thermal advantages of the tray-based CloudRack design.

CloudRack X2 is also ideal for large scale-out deployments into existing rack environments. Customers with traditional rack-mount architectures can leverage the benefits of the CloudRack product line within smaller physical footprints to maximize data center real estate. One-way MicroSliceâ„¢ architecture-based single-socket servers reach extreme price/performance ratios, ideal for Internet applications and workloads.

It’s always great to see further innovations for HPC solutions, in this case from SGI, it will be interesting to see what range of platforms and applications are best suited and if there are any application specific configurations, it does sound very cool and I’m off to check it out and read up the specs.

Share and Enjoy

HP

HP today announced the third in a series of webcasts that provide an in-depth customer view into the inner workings of the technology used in today’s data centers and the business benefits they deliver.

This week’s webcast, titled “Unlocking the Value of Virtualization,” is scheduled for Thursday, Aug. 20.

The HP “CIO Real Talk” webcast series features chief information officers (CIOs) and IT executives from leading companies on a panel moderated by Jonathan Eunice, principal IT advisor at Illuminata. During each audio webcast, IT executives will discuss important industry topics and share real-world experiences as to how they transformed their technology infrastructures to support and drive business objectives. Panelists also will explore the challenges, lessons, tips and techniques related to maximizing technology investments.

I’m looking forward to participating in the webinar, it will be interesting to see what is discussed and what the CIOs and executives are discussing in the virtualization space, do check it out.

Share and Enjoy

SearchServerVirtualization

Is VMware’s new vCenter Chargeback dead on arrival? At its current list price of $750 per managed processor, it may well be. Even though some VMware users understand how Chargeback could help them better utilize resources and reduce sprawl, its price tag will make it a hard sell.

“At $750 per processor, there’s no way in heck I could get management to sign off on what’s basically improved power and cooling,” said Kent Altena, infrastructure architect at Farm Bureau Life Insurance Co. (FBL) in West Des Moines, Iowa.

Altena oversees about 45 VMware ESX hosts running 750 virtual machines, and he has been thinking about ways to push application owners to shut down test and development virtual machines when they’re done with them. FBL currently bills project owners for virtual machines by taking the total cost of a server and calculating the virtual machine’s (VM’s) share of the resources — especially memory. “It works great,” he said, “but it doesn’t give them a model to turn their VMs off. Ever.”

I was reading through a few articles on techtarget and found this one. It’s something that I’ve been discussing more recently, not just in virtualization, but in terms of systems management, actually as a result of a conversation with an IT manager working for a medium business, he’d been complaining that “the price is coming down but everything is extra, even the things that you thought would be included”, and he has a point. As vendors and service providers (not to mention end users), we need to manage expectations, we need to consider not only the marginal cost, but the overall cost, only last week I got a call from a friend working over at a manufacturing business who was asking “did you know that monitoring databases is extra now using…. it was standard in the old version.”.

Going forward it’s an interesting debate, what vendors have to appreciate is that as we deploy more infrastructure, be it physical or virtual, that the marginal capital and support investment doesn’t necessarily increase, that £35 per server for asset management might not be a deal on 10 servers, but as we scale that up it starts representing a significant investment for what could be seen as a hard sell internally – remember that IT is typically seen as a cost. I remember being in a meeting with a business manager who was horrified when he discovered that everything was monitored but not down to the hardware level, “but I thought that had been done he said”, but there’s no budget replied the CIO, you wouldn’t sign it off, “that’s an IT cost was reply”,  as more and more shared infrastructure, shared platforms and virtual environments, how do we manage the customer need for everything to be an IT cost whilst IT manage to deliver at cost? Let us not forget that often chargeback can simply be an average percentage of the IT spend divided by the number of servers, if we took our true costs down to their component parts, how many systems would be ‘off the domain’, not in the backups, running on desktops with a per application configuration and a lack of documentation?

Share and Enjoy

Examiner

Globalization is taking over all the industries and challenges faced by the companies are increasing during this economic downturn. It’s time for top management to evaluate technologies to give their companies a renewed competitive direction. For many businesses in Miami competing in global markets, it is important a strategic revision of their current logistics structure and found new ways to leverage on technological advances to create distinctive advantages against competitors. One of the greatest tools we may find within these new technological resources is the use of virtual teams and virtualization.

Virtualization is a concept that is expanding due the advances on Internet applications. Virtualization allows the company to dramatically improve performance due the creation of a new working environment where the geographic dependency no longer exists, where cultural issues are minimized and the levels of work optimization are increased.

Virtualization can be suitable for small, medium or large companies. With this concept, virtual teams can effectively communicate and interact using a set of web conference tools, run multiple applications and tests, maintain control of the variables and status of projects, orders, shipping, etc.

Check out this article talking about virtualization, it can be a vehicle for reducing costs, for improving delivery, interestingly though (and not being negative), I wonder if we could achieve similar though (granted less exciting and automated/dynamic results) by simply working with what we have to stabilize the core and move on from there. Simple steps like bringing everything up to date, applying best practice configurations and optimizing each server for it’s role, be it a domain controller, a file or application server, even little steps can move your server estate from a high maintenance one, to one that might not be the most up to date, that might be ‘out of support’ but configured and optimized to the best levels that it can be. Let us not under estimate that even if your infrastructure is ‘out of support’, running Windows NT4 on 7 year old servers, that we ‘stabilize the core’, that they have the latest firmware, the latest drivers and security patches, that there’s enough disk space and the page file settings, the disks are defragmented and the settings are just right on a per server/application basis can be the difference between delivery and significant investment.

We need to continue evolving the infrastructure absolutely. But as an IT professional and a business person, we need to achieve the most at the lowest possible cost, spending money is fine providing the returns on investment and opportunity costs are in line with with the capital investment. In essence we need to continue investing in right sized infrastructure for our business, which could be four DL380 G6′s instead of eleven DL580 G1′s, running a mixture of virtual and physical servers. The debate is set to continue and will depend on your business.

Share and Enjoy

MarketWire

PALO ALTO, CA–(Marketwire – July 21, 2009) – VMware, Inc. (NYSE: VMW), the global leader in virtualization solutions from the desktop to the datacenter and into the cloud, today announced VMware global partners, including Dell, Fujitsu, HP, IBM and NEC have announced their support of using VMware Distributed Power Management (DPM) to make their hardware even more power efficient. VMware DPM, part of the VMware vSphereâ„¢ 4 platform, lowers power consumption in the datacenter by aggregating unused capacity and powering off unused servers without disrupting service levels, helping customers slash energy consumption by as much as 20 percent.

“We saw an opportunity to save even more power for our customers by focusing on partially used servers in virtualized environments,” said Dr. Stephen Herrod, chief technology officer and senior vice president of R&D at VMware. “VMware DPM essentially performs server defragmentation. VMware DPM determines the best way to consolidate workloads onto the fewest number of physical servers needed to meet the applications’ performance requirements. VMware DPM then powers off unneeded servers to reduce datacenter energy consumption, powering them back on when the performance needs require more physical resources. This is done automatically, without disruption, while ensuring application SLAs are satisfied. Combined with energy-efficient hardware from our server partners, customers have an opportunity to save costs and make a positive impact in their carbon footprints.”

Anything we can do to further customer opportunities in reducing their costs of running the infrastructure in terms of support, and in energy costs has to be a good thing, the more we can utilize the benefits of the newer more efficient hardware, with rules for powering down idle systems, whilst maintaining the dynamic infrastructure, the more we can continue adding value and managing the customer needs for a cost efficient and dynamic business platform.

I’ll need to read up more about this, I wonder in the enterprise if there might be any concerns about automating elements of the infrastructure with regards to the change and systems management process, an ESX server that’s idle being powered down – is that a change to the infrastructure, and even if it isn’t, how do we update the monitoring, the inventory in line with what’s going on in our live environment?

Share and Enjoy

Network World

When it comes to IT governance, the benefits are obvious: IT is aligned with the business and everyone is happier. But good governance is harder than herding cats. These six steps for success are written by a guest blogger from ISACA, the organization that offers COBIT.  But they’ll work no matter what governance framework you choose.

Microsoft Subnet previously wrote about Microsoft’s efforts to help its customers adopt IT governance best practices, in the form of the Microsoft Operations Framework, newly revised in June. In particular, Microsoft is trying to show how its technologies meld into established governance frameworks like COBIT. To that end, in July, Redmond licensed a free IT governance library from ISACA, a global user association serving 86,000 IT professionals and known for its governance frameworks, COBIT, Val ITand Risk IT. Today, we offer advice on applying the framework, written by guest blogger Brian Barnier, a principal at ValueBridge Advisors and an ISACA volunteer leader who helped write some of its guidelines.

Check out this article talking about IT Governance, more recently we’ve seen quite a trend in improving governance, not just in terms of access/best practice and ownership, but in service delivery, business continuity and accountability. Going forward, how do we manage the need for governance, due diligence and availability in a more complex and layered virtualized environment where my infrastructure is effectively a file, a configuration and a range of equipment? As we move to cloud, who’s governance takes precedence, the client or the supplier?

Share and Enjoy

Philly.com

TOKYO – Hitachi Ltd., Mitsubishi Chemical Corp. and some other organizations plan to jointly develop a next-generation optical disc that can store 25 times more data than a Blu-ray Disc, with the aim of putting the technology into practical use in 2012, industry sources said this week.

The next-generation one terabyte (1,000 gigabyte) disc will be designed to complement cloud computing, which allows for the storage of data and information offline on interconnected databases. It also would increase the efficiency of personal computers.

According to the sources, the next-generation disc will employ hologram technology, in which a laser is used to write 3-D images to record and reproduce data.

Interesting, I wonder if the same technologies could be adopted in long term backups or in virtualization? Could I consolidate my tape backups on to fewer optical discs?

Share and Enjoy