By Barb Goldworm
published: Wednesday, January 30 2008
The start of a new year always causes me to reflect back over past years and/or decades and consider what has and hasn’t changed during those years. This can be fun to do at parties – “remember when gas was 25 cents a gallon.” “ Remember life before email, text messaging, cell phones and (gasp) when the only phone in the house was a wall phone with a really long cord?” Being in the computer industry (for just about forever), the party conversation invariably turns to reflections over how computers have (and haven’t ) changed. So to kick off the new year, here’s a look back over thirty years of virtualization and what’s new and what’s not so new, and what we can learn from the past.
Virtualization then…
In 1978, life in a corporate data center meant big iron, and I mean big, dominated by IBM. Mainframes were the size of my entire office, and then some. Most computing ran in batch mode with interactive computing done from terminals connecting to a time sharing application running on the mainframe. The operating system options were DOS/VS, MVS or the new kid on the block – VM. What was new about these OS environments at the time was …. Virtualization. In the case of DOS/VS and MVS, it was virtual storage. Prior to VS, there was only physical (real) memory, and software had to be written to run within whatever real (physical) memory was on the computer. VM, specifically VM/370 at that time, was a relatively new tool in the datacenter, and was the favorite new hot technology of all the really cool geeks (who were called MIS systems programmers back then, before it became IT). VM ran on the bare-metal, loading the VM/370 Control Program (CP). (VM was originally called CP67 from 1967). Guest OS options were various flavors of DOS, VS1 and MVS.
VM was a great way for developers to test out different versions of their operating systems, all on the same physical machine. And it allowed a phased migration from one version to a new version, leaving some applications running on the old version, while moving apps over to the new version. As IBM worked hard to get customers to move to their newest generation of mainframes, VM became the preferred migration tool, running it on the new box, moving everything from the old box over intact under the old OS, then slowly converting to the newer OS, (after months or years) of testing. New mainframes could be ordered with an easy VM installation package (called an IPO – Installation Productivity Option), optionally with a guest operating system preconfigured under it running DOS/VSE. MVS guests had to be installed separately.
To make VM run faster, you ordered a special feature on the mainframe called VM Assist – a hardware based, virtualization assist. Devices within a VM could be defined as virtual devices, where the VM Control Program (CP) owned it and handled all the I/O, or they could be dedicated, bypassing the virtualization layer, and allowing the guest OS to do the I/O directly, improving performance, among other reasons.
VM was a developer’s dream environment, allowing complete control over his virtual machine, isolated from everything else running. A culture quickly developed of VM gurus who then became known as VM bigots, because once they came over, they never wanted to go back. (OK, I confess, I was an IBM VM specialist at the time, and I was one of those VM bigots). Lots of cool VM tools sprung up, including management tools, and the VM subculture was in full swing.
Virtualization now…
So here we are in 2008, and virtualization is the hot “new” technology, again. It supports guest VMs running distinct and often different Operating Systems isolated from each other, giving the owner complete control over the VM. It offers a great way to convert to the latest hardware and software, insulating the OS and applications from any hardware specific issues. Developers love it, and once people get comfortable (e.g. in VMware) they never want to back. Processors have virtualization hardware assist capability (Intel-VT or AMD-V, and now the next generation of assists addressing paging and I/O improvements in a virtualized environment) which improve the performance significantly. Cool tools are springing up from lots of vendors (affectionately called the virtualization ecosystem), and management is becoming the hot spot going forward.
The terminals of old have been replaced by PCs, which today have the CPU power, memory and disk capacity that mainframes used to have back then. But wait, as virtualization gains traction, the idea of running the PC desktop capability in a VM is becoming the hot new thing, and IT is looking at replacing PCs with…thin or zero clients (aka terminals that have now grown up to include graphics capabilities).
So what can we learn??
They say those who forget the past are doomed to repeat it. So what can we learn from those good old days to make virtual life smoother this time around?
1. Let the hardware do its job –Take advantage of all the hardware assists possible, with as much memory as practical. Once you’ve succeeded at the server level, consider whether/where extending virtualization to the desktop makes sense, and whether it ‘s the right time for a move to thin clients talking to VM desktops.
2. Optimize the software – Spend the time to optimize both the virtualization layer and the guest OS VMs, for your environment. Monitor, measure, tune. Repeat.
3. Train the people – Virtualization offers tremendous flexibility, which also introduces complexity and levels of indirection. Understanding not only the basics, but the subtleties can make or break a successful implementation.
4. Take advantage of the management tools – These tools are what will allow the flexibility to work for you rather than against you.
So after 30 years, the whole world of corporate computing has changed, and yet, not. And I guess I’m still a VM bigot.
|
Barb Goldworm is president and chief analyst of Focus Consulting, a market research, analyst and consulting firm focused on systems, software and storage. Barb has spent 30 years in various senior management, marketing, sales, engineering, technical and industry analyst roles with IBM, StorageTek, Novell, Enterprise Management Associates and several successful startup ventures. Barb began working with virtualization at IBM in the late 1970s (as a VM technical specialist), and later was VP of Marketing for the company that developed the first PC software distribution product in the mid-1980s (acquired by Novell in 1992). In 2006, she was commissioned by Wiley Publishing to write a book on blades - she agreed but changed the focus of the book to include a major emphasis on virtualization - the book was released in 2007, entitled "Blade Servers and Virtualization: Transforming Enterprise Computing While Cutting Costs", available on Amazon.com, Borders and Barnes and Noble.
Barb is a frequent keynote speaker, chair, and presenter at industry events. In 2007, she chaired the Server Blade Summit on Blades and Virtualization, and has been the keynote speaker for numerous Virtualization Summits this year. She also created and chaired the Interop Network Storage Track and has been one of the top 3 ranked analyst/ knowledge expert speakers at SNW.
Barb has published extensively since the 1990s, and has been a regular contributor to Network World, Computerworld, Computerworld Storage Networking World Online, Tech Target, and now Virtual Strategy Magazine. She has authored numerous research studies, landscape reports, and business and technical white papers on systems, software, storage, storage networking and enterprise management. Barb can be reached at
This e-mail address is being protected from spam bots, you need JavaScript enabled to view it
|
