---

BIZSCI IN BRIEF
Virtualization. You think consolidating the work of twenty servers onto one is impressive? Consider that fact that a modern Dell PC can handle billions of instructions per second, more than any thousand employees in most companies could ever task. If that's the case, then maybe your consolidation goal should be ALL servers onto one, .... or two for redundancy ... or three for added Disaster Recovery. You think 100+ to one server consolidation is not possible? Read on ...

---

We are a large law firm with about two thousand workstations and two hundred production servers in data centers. As a CIO with a Windows shop for most of the last decade, I have leaned to live with the predominant WinTel Cisco IT Infrastructure design. But If you know me, you know I also think that same IT architecture is absolutely nutty. Billion dollar enterprises shouldn't have to eek a computing existence from a slow network of graphical workstations that devote nearly all their tremendous resources to communications overhead, self maintenance, security monitoring, waiting on network, waiting on storage and just about anything but completing the work of the company.

But that's another thread.

Today I want to give an update on an IT topic I am enthusiastic about: Virtualization. Virtualization is amazing because it can help make the nutty, better. More...

Tags: X86 Mainframe, Virtualization

This is a continuation from Part 1

Our goal was to move all production and just under one hundred test and development devices to a single environment, then double the equipment in that environment (for redundancy) in the primary SF area data center and add a another environment to the DC DR site. A year ago I calculated minimum computational requirements for the firm (another project) and plugged in the numbers for the entire company. I calculated that between two and four servers (with an x86 mainframe design) could probably cover the entire firm ... that's a ratio of 100:1 or 150:1 depending on how you calculate it.
We budgeted 2-3x redundancy, i.e. 6 servers.

In Jan our engineers, Jim, Tom, Ryan, Chris, Rolando and the rest of Albert's team began working in earnest on the project. As you can guess, the hardware design required a little more than:

Step 1: Pull Dell out of box ,
Step 2: Install VMware
Step 3: Auto migrate servers.
More...

Tags: X86 Mainframe, Virtualization

This is the second article in the Forward to the Mainframe? series
Part 1: Revisiting the Mainframe Utility Computing Model
Part 3: A Virtualization Machine - the X86 Mainframe

The history of the Luddites is a interesting study, particularly for practitioners in the computer industry. In the early 1800's, Ned Ludd's textile culture defended hand looms over the new large knitting frames. This was despite the fact that the larger looms produced much more cloth with far less skilled labor.

Instead of looms, today's Information Technology workers repeatedly choose servers without the virtualization or processing redundancy needed to keep critical applications running. And they choose these servers over commercially available mainframes and other highly available systems with proven track records.

Let's look at the factors that led and continue to lead an intelligent and informed business computing market to chose lesser technologies.
More...

Tags: X86 Mainframe, Efficiency

In the next few articles, we will take a deep dive into how data centers can be made more reliable and less expensive. Since mainframe design is an important part of this discussion, there will be frequent mention of the mainframe in future articles. In this article, we attempt to grok the term "mainframe".

Let's start of by pointing out that while I will often cite IBM and it's products when describing the historical development of the mainframe, the mainframe is not proprietary to IBM. Quite to the contrary, I believe the mainframe design, inclusive of IBM's innovation, represents the most effective effort to "get technology right" for general business and scientific computing purposes.

As I have stated in a prior article on the mainframe computing model, the transition from mainframe technology to networked PC Server technology, was typically not a quest for better technology. Instead, companies that purchased networked PC Servers were generally seeking a better computing buying model. Companies found that servers enabled them to grow rapidly without pain because of opportunities to add capacity and new features at lower cost. In contrast, those same companies felt increasing apprehension about their mainframe investments, as changing business requirements and increases in business output always led to higher mainframe costs. More...

Tags: X86 Mainframe, Virtualization

This is the third article in the Forward to the Mainframe? series
Part 1: Revisiting the Mainframe Utility Computing Model
Part 2: WinTel, the Luddite Revolution

Forget everything you think you know about mainframe technology. Everything that follows
X86 Mainframe Home Page More...

Tags: X86 Mainframe, Efficiency, Green, Virtualization

Mainframe Utility Model vs. the Server Ecosystem

A mid-size firm might run more than two hundred (200+) production Wintel servers, each with hard drives, memory chips, power supplies and fans expected to fail within four (4) years which equates to about two hundred (200) weeks. If you stagger equipment purchases, then statistically, one of the two hundred servers will either fail or need service each and every week.

A midsize manufacturer or bank might invest in a single IBM zSeries mainframe. Managed properly, the company should experience no more than one outage over the four to seven year life of the system. More...

Tags: X86 Mainframe

This is the first article in the Forward to the Mainframe? series
Part 2: WinTel, the Luddite Revolution
Part 3: A Virtualization Machine - the X86 Mainframe

IBM's Mainframe Business Model

As the original utility computing device, the mainframe business model enables a company to invest in a system that it can grow with.

With a mainframe purchase, you buy a machine that typically can do more than you need. This "utility" computing model enables you to buy only what you need. Thus, instead of paying a price of two million dollars for unlimited use of a mainframe, you might pay one million dollars to obtain the computing power you need today. More...

Tags: X86 Mainframe, Efficiency, Green, Virtualization

I've got this idea about a simple little box that would be sold by a big company. The box would be amazing because it would support most anything a CIO would want to do. More...

Tags: X86 Mainframe, Open Source, Virtualization