You Are Browsing ‘server industry’ Category

Cable Messes: Want Meatballs with that?

Spaghetti Cabling that would make a fully-grown sys admin cry.
or… Cable management is for wimps.

Spaghetti Cable Mess

Most network and system administrators like a nice tidy cabled cabinet of servers and network equipment. Sometimes it just doesn’t work out that way. The following are some of my favorite images of nasty cable messes.

===================
A Cable Mess of Beauty

photo by: Cormac Phelan
===================

Spaghetti Cable Mess

========================
Racks and Racks of Spaghetti
photo by: Andrew McKaskill
========================

================
Cable-Bound Racks
photo by: dM.nyc
================
Cable Mess
=========================================
Good Old-Fashioned Telephone Wire Mess in Bangkok
photo by: Nurmi
=========================================
Telephone Wire Mess
====================
Cable Mess (un)Managed
photo by: gruzuk
====================
Cable Mess Un-Managed
================================
The Spaghetti Man on Sixth Avenue, NYC
photo by: Ken
================================
Spaghetti mess Man
====================================
Spaghetti Cables on the CMS Tracker at CERN
photo by: Richard Brauer
====================================
CERN cable mess
==================
Colorful Cable Salad
photo by: Zenji
==================
Cable Mess
==================
A Fine Mess
photo by: BRphoto
==================

====================
Cable Mess Frustration
photo by: Jonathan Arehart
====================
Cable Mess Frustration
==================
Rat’s Nest Cable Mess
photo by: n8foo
==================
Rats nest Cable Mess
==============================
Sound Stage Cables get messed up too
photo by: Laura A Watt
==============================
Sound Cables mess
==============
Cable Chaos
photo by: VespaGT
==============
Spaghetti Mess
==============================================

 

LOPSA’s Sys Admin Horrors

LOPSA (League of Professional System Administrators) is a terrific organization that Vibrant wholeheartedly supports. Afterall, they host National SysAdmin Appreciation Day and also came up with a terrific photo set of Sys Admin horrors including the cable messes below:

=============================
DO NOT TOUCH ANY OF THESE WIRES!

=============================

Do not Touch wires

=============
Burnt Spaghetti

=============

Burned Spaghetti Cables

=================
Spaghetti Cable Hurl
=================

Spaghetti Cables spew

========================
Network Cables or Silly String?
========================

Green Spaghetti Cabling

================
Patriotic Cable Mess
================

Patriotic Spaghetti Cabling

=======================
Network Floss and Modems
(rumored to have run the dialup access at an old ISP)
=======================
Network Floss

Any good ugly cable messes I might have missed?

PS3: Your mother(board) was a blade server!

The Playstation Blog has drawn the curtain strings a salivating tad to show off their very impressive server cluster of rackmounted PS3s that run it’s multi-player game Warhawk.

The Data Center Knowledge blog keanly observes:

The powerful processor is no stranger to the data center, as it is also being used in IBM System X blade servers and System Z mainframes.

Players of Warhawk on Playstation 3s take advantage of a server method they call IGS or “Integrated Game Server”, which is essentially distributed computing. This allows players to take advantage of PS3s nearby to reduce latency, but Sony also wanted to have dedicated racks of PS3s available to gamers as well.

It is this type of innovation potential which made us choose the PS3 as one of the ten servers that changed the world last year and it’s beginning to fulfill the prophecy, at least in the gaming world.

I would love to get my hands on one of these rackmount kits and rig a PS3 into one of our server racks for our next in-house video game happy hour.

Google runs 100 Billion Servers

Google's Sergey Brin and Larry Page as Dr. Evil and Mini-MeOK, maybe not quite yet… but Pandia Search Engine News is reporting that Google runs more than 1 million servers.

Pandia cites a Gartner analyst from the fair elven lands of Norway (the article is in Norwegian, so we’ll have to take their word for it).

Google reports that it spends some 200 to 250 million US dollars a year on IT equipment. We know that Google make use of a large number of cheap off the shelf servers using open source (and free) LINUX.

If we say that Google spends 900 USD on each machine, and the same sum on storage and peripherals, it is a fair guess that Google uses some 1 million servers in its data centers.

———————————————————————————-

Pandia goes on to speculate that Google likely installs 100,000 new servers per quarter in datacenters around the world. That would place Google as the 4th largest server producer in the world after HP, Dell and IBM.

But does Google really have 1 million servers? I’m not so sure. The logic above is fuzzy at best and others have estimated they only have 450,000 systems. My guess is reality falls somewhere in between these impressive numbers.

I wonder how many servers are purely dedicated to facilitating searches for LOLcats?

EATIN UR DATAZ

We know that Google primarily uses white box linux-based server models and blade servers, but should they ever need memory or CPU upgrades, Vibrant sells a wide range of Used Servers. Our sales reps hear “1 million servers” and can’t help but drool.

The Top Enterprise IT Bloggers

There are endless numbers of tech bloggers online, but finding those that cover enterprise IT is a bit more challenging. For the past several months, I’ve been compiling a list of the best that I could find.

Quick observations: Sun is the runaway blogging leader. HP has a ways to go, as does Cisco, but they’re finally getting there. Storage Execs are your buddies in the blogosphere, as they go by their first names (i.e. Chuck’s Blog, Dave’s Blog, Hu’s Blog, etc…)

General IT Bloggers
- Some great writers and thinkers covering Enterprise IT of all flavors:

  • Managing L’Unix – Paul Murphy (ZDNet)
    - If a high-level view on enterprise IT is your bag, then this blog’s for you. Paul has been an IT consultant, programmer and is highly experienced in systems management and operation.
  • Life After Coffee – Jon Emmons
    - Server Admin and Author of Easy Linux Commands, who writes incredibly practical posts on AIX, Solaris, UNIX and Linux.
  • ComputerWorld Hardware Blog – Various
    - Aggregated list of ComputerWorld’s numerous blog posts tagged “hardware.”
  • RootPrompt – Various
    - Another aggregated portal, this time consisting of Unix bloggers.
  • Makezine Blog (DIY Tech)
    - Although they MAKE all kinds of techie goodness, it’s posts like “How to make a terabyte file server for $500″ and “Router design project” that qualify them for this list.

Sun Bloggers
- Blogs from Sun employees and users:

  • Jonathan Schwartz’s Weblog – Jonathan Schwartz
    - The General of the Sun Army, Jonathan Schwartz is the CEO of Sun Microsystems and the envy of every figurehead that wishes they had his ability to write killer blog posts.
  • blogs.sun.com – Various
    - Sun compiles and updates their top Sun.com blog posts here. Sun has a massive army of bloggers.
  • SunHelp.org – Bill Bradford
    - Mr. Bill has been providing resources to the Sun community for over a decade. A great place to stay up to date on developments in the Sun world.
  • The Blog of Ben Rockwood (at Cuddletech)
    - A Sun evangelist and Sun community champion, Ben is the CEO and founder of Cuddletech as well as the Senior Unix admin for Homestead.com.
  • John J. McLaughlin (at Blogspot)
    - John earns this spot mostly on the merits of his terrific weekly System News email newsletter. John has been a Unix user, developer, admin and evangelist for 26 years.

IBM Enterprise Blogs
- Bloggers covering IBM, AIX, OS400, etc…

  • IBMer’s Blogs (various)
    - Summary of IBM bloggers, including an editor’s choice section showing the best recent posts.
  • AIXpert Blog (various)
    - An IBM highlight. A group blog by Ron Barker, Shawn Bodily, Bruce Spencer and John Tesch.
  • I on I5 (Lisa Jobson)
    - Lisa is an i5 specialist and currently the Canadian regional dedicated specialist for iSeries integration with xSeries.

HP Enterprise Blogs
- Covering HP-UX, HP Integrity and Enterprise HP:

  • The BladeWonk (Gary Thome)
    - Great name and this HP blog is off to a good start. No posts since February, but hopefully we’ll hear more on server blades soon.
  • The Storage Technology Watch (Ash Ashutosh)
    - HP’s VP and CTO has begun blogging on enterprise storage technologies.
  • Encompass Blogs (various)
    - Entries from those blogging at Encompass, the HP Enterprise Technology user group.

Cisco and Network Hardware Blogs
- Blogs covering the networking hardware world and Cisco:

  • Network World’s Cisco Subnet (various)
    - NWW’s excellent writers and Cisco user blogs. The top portal for independent voices of Cisco customers, evangelists and detractors.
  • Cisco’s DataCenter Networks Blog (various)
    - Cisco’s best corporate blog (from their blogs.cisco.com portal).
  • TechIQ (VAR Guy)
    - The mysterious and illusive VAR Guy covers many IT channel topics, but his Cisco coverage is surely top-notch.
  • Cisco Blog (Jeremy Cioara)
    - “The world of all things Cisco” is their tagline. Jeremy is a Cisco instructor.

Enterprise Storage Blogs
- Blogs covering mass storage, SAN, NAS, Backup and Recovery, etc…

  • Drunken Data (John Toigo)
    - Great name and a great blog about storage & data management.
  • Storing up Trouble (Chris Mellor)
    - General Storage posts from the UK at TechWorld.com.
  • Stor Wars (Tony Asaro)
    - Excellent material from Tony, senior analyst at the Enterprise Strategy Group.
  • Steve’s IT Rants (Steve Duplessie)
    - Strong storage opinions (and great reads) from the founder of ESG.
  • StorageMojo (Robin Harris)
    - Robin is a senior analyst at the Data Mobility Group and always interesting on storage topics at the Storage Mojo blog.
  • ZeroWait’s Blog (Mike)
    - Independent NetApp reseller on various storage topics.
  • Dave’s Blog (Dave Hitz)
    - NetApp Blog from co-founder and Executive VP, Dave Hitz. Dave does a great job writing interesting posts the likes of this recent entry: “Does Helping Customers Use Less Disk hurt NetApp’s Business?
  • Mark’s Blog (Mark Lewis)
    - EMC corporate blog from Executive VP and Chief Development Officer Mark Lewis.
  • Chuck’s Blog (Chuck Hollis)
    - EMC Insider and VP of Technology Alliances, Chuck Hollis’s Blog.
  • Hu’s Blog (Hu Yoshida)
    - Hitachi VP and CTO, Hu Yoshida provides storage technology insights.

Used Enterprise IT Hardware Blogs
-
Blogs covering the secondary enterprise IT industry:

  • The Remarketer
    - Teacher says you have to love yourself before you can love another. ;) Beyond that, we’re one of only a handful of authentic voices in this space.
  • Brad Reese’s Blog
    - The “People magazine” of the networking world, Brad frequently discusses the used networking marketplace.
  • Corey Donovan’s Blog
    - My blog at Network World’s Cisco Subnet on used and refurbished Cisco topics.
  • Refurb Radio
    - An industry insider podcast on the refurb IT market, with the occasional blog post, too. I host the podcast with Vibrant‘s own Marty Brown.

More?
- I don’t intend this as a static list, so please suggest any other blogs you would like to see here.

Sun’s Project Blackbox stops in Minneapolis

Last week, Sun Microsystem’s Project Blackbox (aka The DatacenterMobile) finally made a stop in Minneapolis, but alas, daycare couldn’t be lined up and I didn’t make it over to the St. Thomas campus to get a tour. I don’t feel like I completely missed out though, because TechnologyEvangelist.com offered a great video of Sun’s Twin Cities visit.

sharedVideo(715960650)

Find out where Project Blackbox is now.

Last-Minute Daylight Savings Time Change Fixes – Enterprise Guide

At lunch yesterday, I was amazed to find out that the two co-workers joining me hadn’t yet realized that daylight savings time goes into effect this coming weekend. Of course, they aren’t responsible for any of our company’s or clients’ IT hardware or computer systems. Otherwise they most certainly would have been aware that Daylight Savings Time is starting a month earlier this year due to new US regulations.

Just in case any IT admins out there have gotten (way) behind on their DST07 updates, here is a summary of where you can find help for various manufacturers:

Daylight Savings Time Change (2007) Resources:

IBM Servers and Storage

- A Grid of IBM’s DST updates

Sun Microsystems

- Sun’s BigAdmin guide to DST07

HP Servers

- HP Support Document

Microsoft Servers

- Overview of Updates for Microsoft Windows OS for servers and workstations


Novell SUSE Linux Servers

- Explains how linux handles DST

Red Hat Linux Servers

- Red Hat Knowledgebase Features DST inf

Hitachi
- PDF Alert on DST for Hitachi Products

Cisco
- Summary of the various Cisco platforms effected by DST

Oracle
- Oracle’s DST FAQ
https://metalink.oracle.com/
- Available in Oracle’s Metalink (for authorized users)

Blackberry
- not exactly enterprise, but included because of the sheer numbers of enterprise admins who use them…

Note… Some manufacturers such as Microsoft and Sun Microsystems are only offering patches on the latest OS offerings. Older versions can be updated, but the cost is pretty daunting. Thus, perhaps this event will trigger all those admins who just can’t get enough of Solaris 5, to finally give 10 a shot.

God speed.

Google’s Hard Drive Study on SATA Disks

This month Google released their compiled data on hard drive performance from their datacenters around the world. When you’re Google, this type of study is far beyond the anecdotal reports that can be found on gear forums like “I picked up these piece of $&@# Drives from (xyz low-end drive manufacturer) and one already crapped out!”

Instead, Google basically posted “We’ve got these datacenters all over the world with over 100,000 cheap SATA and PATA drives between them and a bunch crapped out!”, oh yeah and here is some great insight into the circumstances under which this occurred. Over 5 years, they recorded every failure and many variables and then repurposed it into an independent report unprecedented in size.

Here are some of Google’s hard drive findings:

1) Temperature surprisingly didn’t have much influence on failure rates.

Warm or cold, Google didn’t report much statistical variance here.

2) Older Drives fail more often but drive use didn’t factor into it as much.

We would be concerned about this as a used IT reseller, except we don’t see many hard drives lately anyhow due to privacy concerns. When we rebuild systems these days we seem to primarily use new drives from the manufacturer to complete configurations.

3) Some cheap drives are better than other cheap drives

At one point Google mentioned that certain manufacturer’s drives definitely failed more often than others. Unfortunately, they didn’t mention which manufacturer’s hard drives failed though and it is probably not in Google’s best interests to share that information. If that ever leaks out, it could be really damaging to a manufacturer like IBM, Seagate or Maxtor.

4) SMART is pretty smart (most of the time)

S.M.A.R.T. (Self Monitoring, Analysis and Reporting Technology) scans that show a drive problem are very likely to fail within 60 days. Google determined that after receiving a SMART error, drives are 39 times more likely to fail than a drive that isn’t reporting one. Unfortunately, many other drives will also fail that never show a SMART warning (36% of SATA disks that failed were never detected by SMART).

A Slashdot contributor questioned whether or not Google’s study was useful, but I definitely think it is. As one commenter “spineboy” stated “To me it’s useful – if I get a SMART warning, then I’m definitely backing up my drive and will replace it before it croaks.”

Sun’s DataCenterMobile, the Wienermobile and other type hits

When I first heard that Sun’s Datacenter in a Box was slated to go on a tour of the USA, the first thing that came to mind was naturally the Oscar Meyer Wienermobile. Sun shouldn’t take this as a slight though, I think it’s a brilliant move, and I tip my hat to them as a competitor (we sell used Sun servers). Anyhow, in looking into this, I discovered there is a whole world of these product themed cars.

Here is our List of 11 Traveling Product-mobiles, starting with the legend.

1) Oscar Meyer’s Wienermobile
(The Grand-Daddy of product-mobiles)

Created in Chicago in 1936 to promote Meyer’s wieners, the giant hot-dog on wheels still visits the grocery stores, parades and state fairs of the nation. Six vehicles reportedly travel during the summer hosting 12 lucky college kids. Out of over 1000 applicants only these fortunate few get to attend “Hot Dog High” in Madison, WI and embark on a sticky summer promoting processed meat. “No one can really relate to life on the road besides other hotdoggers,” stated Michelle Norton, an alum.

If you can’t catch the wiener-mobile in your town, take a virtual tour at Kraft’s website instead.

2) Sun’s Datacenter in a Box
(aka “Project Blackbox,” aka “Sun’s DataCenterMobile”)

In a brilliant marketing move, Sun Microsystems took the natural next step with their buzz-grabbing product by taking Project Blackbox on the road. The DataCenter in the Box has been criticized as being more of an attention-grabbing vehicle than a viable product, so why not make it a real vehicle and take it on a road show? (Despite the barbs thrown Sun’s way, I admire their bold innovation and believe it helps the entire Sun market.)

The self-contained datacenter is actually only part of a vehicle, being connected to the back of a semi-trailer, but it serves the same purpose. And you know what, I’ll definitely try to get a look at it when it stops in Minneapolis for two days in March.

3) The Peeps-Mobile
(The Peeps Fun Bus)

Peeps are scary enough in your easter basket, (What is their shelf life… 20 years? 50 years? Would they survive nuclear fallout?) but a giant Peeps “fun” bus is absolutely terrifying.

4) The Hershey’s Kiss-Mobile
(Gene Simmons not included)

Created in 1997 to celebrate the 90th birthday of Hershey’s kisses. Hershey’s dubbed it “America’s sweetest ride” which is clever in the kind of way that makes me wince typing it. Their Sweetest Mom Contest is a much more palatable promo. Mom’s deserve plenty of sweetness after all. Tour schedule here (alas, no Minnesota stop).

5) The Pep-O-Mint Lifesavers Truck
(The original fresh-maker)

The earliest product-mobile we could find, having been mounted on a truck in 1918. Sadly, we couldn’t find photo documentation, but picture this giant lifesaver above (on display in Gouverneur, NY) on top of an old-time truck and you’ll get the picture.

6) The Meow Mix Mobile
(Catmobile delivers Meows on Wheels)

The Meow Mix Mobile actually serves a very worthy cause. It stops to visit cat owners around the US who have difficulty traveling to the store to purchase food for their cats on their own (homebound, elderly and disabled cat owners) through it’s Meows on Wheels program. Or is this really a secret way to discover cat hoarders?

7) The Heinz Pickle-Mobile
(not actual picture)

We couldn’t find a photo of this one either, but Johnny McGuire’s delivery-wagon is a very worthy substitute.

8) Mr. Peanut’s Hot Rod
(Plantar’s Nascar Peanutmobile)

Planters is “The Official Snack of Nascar,” but we can’t quite see the monacle catching on with the racing crowd.

9) Goldfish-Mobile
(“Miles of Smiles” from Pepperidge Farms)

[youtube]tBJDgiwpGos[/youtube]

Take a goldfish, add giant sunglasses and drive it around the country. Brilliant!

10) Kellogg’s TonyMobile
(the cerealmobile)

A traveling cerealmobile featuring Kellogg’s all-stars. This one feals more like a parade float, but who doesn’t love cereal?

11) The SpamMobile
(The “Spambassador” from Austin has the floor)

Last but not least, straight outta Austin, MN, is the appropriately boxy Spam-mobile. It is forged from a trolley car shell and piloted by elite “Spambassadors.” The Spam bus is on the road now so check the schedule to connect with Spam in the place that you are.

Interested in a product-mobile of your very own?
(Who builds something like that?)

Prototype Source does… Not only building prototypes for Volkswagen and GM, this cool Santa Barbara company has found a great niche designing and building product-mobiles for a wide variety of clients.

One parting note… Al Unser Jr. once drove the Wienermobile around the Indianapolis Speedway at over 100 miles per hour. Mr. Schwartz, you’ve officially been challenged.

———————————————————————————————-

Servers in the Movies – Our Top Ten

A few weeks ago, Corey wrote about the top “Ten Servers that Changed the World.” In reaction, I decided to make my own list… The Ten Servers that Changed the Fictional World.

There are two guidelines for this list. One, they must exist only in the world of movies or TV. Second, they need to fit the following definition: A server is a computer system that provides services or data to other computing systems – called clients – over a network or other communication device.

With that said, here they are. In no particular order…

————————————————————————————————————-

1. Teletraan 1 – Transformers (1983 – 2007)
You might not know much about the Transformers yet, but by next summer, you will. If a server is a computer system that maintains and controls other computer systems, then Teletraan is definitely a server.

Teletraan is the computer that assisted and recreated the Autobots and Decepticons in the Transformers TV show. It was the computer that picked out who would be a semi-truck and who would be a cassette recorder. It routinely provided the lower computer systems (i.e. Transformers) with new data via the spy satellite or extensive database. It was hardware built for imaging.

Real-world comparison: When you think of a computer that excels in graphic design, you think of Apple. So, in theory, we could compare Teletraan with Apple’s Xserve. Two dual-core 3GHz Xeon processors with a Xserve RAID for 7TB of Autobot history. On ‘roids.

————————————————————————————————————-

2. MCP – Tron (1982)
The Master Control Program was a massive server running multiple programs at once. The ENCOM corporation had many divisions, each one utilizing this mainframe server for storage and operation. Each program could wander around the mainframe and interact with other programs, ultimately destroying the MCP or operating system and creating big problems. (Not to mention the memory leaks in the RAM were out of control!) Clearly the hardware was in need of a revision. ECC memory and Memory Protection were desperately needed. In the end, the heroes celebrated their triumph over the MCP inside their virtual world. Little did they know, come Monday, the IT admin would probably reload the “damaged” MCP software from a backup-tape – and they would all be deleted shortly after.

Real-world comparison: The MCP could be compared to any modern day operating system. The idea that the MCP would absorb other programs and take over their functions is not unheard of – for example, Apple launched the Dashboard after Konfabulator was created. (And Microsoft has a few anti-trust suits under its belt to further illustrate this point.)

————————————————————————————————————-

3. UNIX environment – Jurassic Park (1993)
The UNIX environment here is a classic geek joke. Everything we saw was real – created by Silicon Graphics and called IRIX. InGen was the corporation funding the island, and from an IT perspective they let the worst possible thing happen: they allowed one programmer to design the infrastructure with no supervision. What’s worse, they obviously required no documentation of what was done. The result was a kid had to hack in and gain ROOT privileges. The likelihood of a young kid knowing a way to get ROOT (and not a more experienced programmer) is pretty hard to swallow. The hardware for this server was probably minimal, running door locks and starting Quicktime movies. “We spared no expense!” You would think that with the millions of dollars they spent on the park, they could have hired a couple newbie programmers and added a server on the backend.

Real-world comparison: Since the actual screen shots used in the movie were based on real software, there isn’t much to compare. This could have been any run of mill UNIX server with various dummy terminals.

————————————————————————————————————-

4. WOPR (War Operations Plan Response) – WarGames (1983)
The good people working at NORAD decided to give total control of the nuclear weapons to a computer, specifically the War Operations Plan Response server. It was a computer they bought, and obviously didn’t research very well. My biggest criticism of the WOPR is that the NORAD folks didn’t look at what they had – it had games on it. Everyone knows, if you want a computer system to be optimized for its sheer power, you delete the games. There’s a reason that Windows Server 2003 doesn’t come with Solitaire.

Real-world comparison: The WOPR hardware itself was probably pretty simple, the radar information from around the country was fed in via fiber optic cables and interpreted as a whole. Then the commands to the missiles could be phoned out and delivered via the 1200bps modem (remember this was 1983). The data analysis wouldn’t be much different from a weather station’s software, with multiple servers being clustered together, commonly called a Beowulf Cluster. (Also used for the Sony Playstation 2.)

————————————————————————————————————-

5. SKYNET – The Terminator(1984)
The SKYNET computer was designed for missile defense, and I’m guesssing nobody who designed this computer system saw “WarGames” (see above). The hardware for this server would have been some sort of distributed system, because if it was at one location, turning it off would have been simple. The software running on SKYNET must have been very buggy at first. (Imagine how incorrectly a child would learn everything if he was allowed to teach himself instead of being taught.) Maybe that’s why SKYNET began killing everyone once it got the chance.

Real-world comparison: Although we can’t see it, most of the internet we use comes from an Akamai server. It’s a basic web/file server but it synchronizes constantly with other Akamai servers. When you download a file, it comes from one of maybe hundreds of servers. That’s why the server doesn’t go down when millions of people hit it at once. All that needs to happen is for Akamai to gain control of our nuclear weapons.

————————————————————————————————————-

6. The Gibson – Hackers (1995)
Hacking the Gibson has become a euphemism for showing off your skills. The actual nature of this machine is much less showy. The name comes from an homage to a sci-fi writer, William Gibson, and was used to calculate oil drilling locations and data. The movie depicts this as a single computer, which doesn’t exist in reality. They do exist, however, in clusters. The server from the movie (aside from the gratuitous GUI) is just a basic server running UNIX. At the end of the movie, the IT team tried to fend off the attacks of hackers around the world. If they only would’ve had a stateful firewall in front of that server, they could have saved a lot of time. The firewall would’ve blocked all non-solicited traffic to the inside network, leaving only the telnet connections in, which could be turned off in a state of emergency.

Real-world comparison: GRID computing is similar to a Beowulf Cluster but differs in the open standard and distances. The cluster needs the computers to be directly connected and running similar software. The GRID concept allows computers all over the world, running different software, to contribute to the overall computing power. An example of this would SETI@Home or Folding@Home. The dream that a single server could analyze as much data as 10,000 desktop computers is just that – a dream.

————————————————————————————————————-

7. The Source – The Matrix (1999 – 2003)
The Source was a server that supplies data and services to multiple subsystems and computer networks, but the actual hardware is hard to imagine. The movie explains the actual world with a unique, yet effective plot mechanism. They don’t know their own history and origins of the machine world. Convenient.

Real-world comparison: A logical real-world equivalent to the Matrix is Second Life. According to Linden Labs, they use servers that have two dual core processors which they call Class 4. They call the network of servers a “grid,” but only because of the layout, not because of distributed computing power. Each server runs a specific area of the world. There are disputes, but one report says there are 2,579 servers. However, the World of Warcraft users will argue that they have the bigger, better world. Second Life doesn’t have as many users as World of Warcraft, but WoW has individual servers (or realms) and a relatively static world in comparison. If you’ve played Second Life, then you’ll understand the comparison. The world is so slow to render and low in quality and resolution that it becomes almost unbearable. It would take some serious horsepower for a system to render a world so realistic, a human would believe in it.

————————————————————————————————————-

8. HAL9000 - 2001: A Space Odyssey (1968)
Hal9000 The HAL9000 or SAL9000 (the earthbound female counterpart) was responsible for making decisions and carrying out orders aboard a spaceship when humans were not willing or capable of doing so. The break-through on this server was the AI, supposedly the most sophisticated artificial intelligence man could produce. When the movie was released, it was believed that this level of intelligence could be created by the year 2001 – but we now understand that it’s much further away than anticipated.

Real-world comparison: This computer has been compared to the Blue Gene. The Blue Gene/L currently holds the top spot on the fastest super computer list with 131,072 processors. It would be a top contender to play HAL9000 on the next space mission. My only suggestion would be to have a kill switch incase HAL decides the humans are conflicting with orders and needs to be “deleted.”

————————————————————————————————————-

9. VIKI – I, Robot (2004)V.I.K.I from iRobot
V.I.K.I. is the governing intelligence behind all the robots in the film, “I, Robot.” In the movie, V.I.K.I. is a positronic brain – a CPU that can create new pathways for each new task it learns. While nobody has perfected the technology, they have patented it. Eventually, the final product would be a computer that could be taught – a technology that seems inevitable based on CPU trends. If you want to know more about positronic brains, watch some Star Trek, The Next Generation.

Real-world comparison: This is a bit of a stretch, so bare with me. V.I.K.I. was the server responsible for sending out updates and keeps tabs on all the robots in the world. When she decided humans were a threat, she sent out a final update that allowed the robots to begin killing humans.We currently have a server/client scenario that is marketed to sit in every familys living room and receive commands from the “mothership.” We call it Xbox Live. So next time you are changing the CD in there, watch your fingers! Microsoft could decide to send a lethal update to our Xboxes!

————————————————————————————————————-

10. Deep Thought - Hitchhikers Guide to the Galaxy (2005)
The Deep Thought computer was created to answer one question. What is the ultimate answer to Life, the Universe, and Everything? As you might imagine, the question is a bit vague. So the computer begins to calculate the answer as it has been commanded. After 7.5 million years, the answer is 42. The amount of computer cycles to compute all possibilities is quite large. Calculating all the possibilities from a question is currently a busy project.

Real-world comparison: There was a paper written by Philippe Oechslin, describing the benefit of Time vs Memory in computational speed. It explained how you could calculate out all the possibilities in an algorithm and save the results, and then the software would only need to access the file instead of force the CPU to do the math. Of course, the most applicable use of this has been for password cracking. Rainbow tables are the result of this study. The basic set of characters doesnt take too long on top of the line hardware. However, every good system administrator knows that enforcing complex passwords in a Windows domain is standard. That means eight characters with a least one capital letter. If you use that character set, you are looking at a bit longer…. Over 32 years on a single CPU computer, still much better than 7.5 million years!

Big is the New Small (for IBM Mainframes)

Smaller chips, smaller servers, smaller datacenters, smaller world…

IBM wants you to forget about all that. If you’ve had your ear to IBM’s train tracks, you’ve heard the message too… Big is the New Small.

Mainframes and Beer

So, when the world is all about blade servers and you want to shift a little love to your system z9 and zSeries mainframe servers, what do you do?

  1. Make Big = Easy. Mainframes have always been considered cumbersome, complex and difficult to use. IBM is investing $100 million over five years to make the systems and the z/OS easier to work with. Implementation and administration is expected to become much less complex.
  2. Understand that students shall inherit the Datacenter.Today’s IT students will be tomorrow’s IT decision makers. IBM realizes this and has taken some smart measures in involving them in zSeries mainframe learning.

    Evidence of IBM’s Youth Movement
    - zNextGen Initiative
    - Photos of IBM’s new faces in zSeries technology
    - Master the Mainframe Contest
    - Big Iron Poetry Collection (Mainframe Haikus)
    - Personal Ads in Student Newspapers

    A couple of highlights here are the zNextGen Initiative and the Master the Mainframe contest.

    In their quest to add 20,000 new IBM Mainframe Experts, they’ve launched a program called zNextGen in cooperation with the SHARE user community. Many of the old grey-bearded mainframe wizards are nearing retirement and IBM hopes this user community will pass on some wisdom to the new generation.

    The Master the Mainframe Contest has been a big success for IBM with over 700 participants in their first attempt. The contest is offered to technology students who (preferably) haven’t programmed on a mainframe before. There are nice prizes, including the obligatory iPod and a very sweet t-shirt (with the slogan “Mainframes do IT 24/7″).

    I think that Seth Godin, author of Small is the New Big, would agree that IBM has been acting like a small, innovative business here in order to make big the new small.

  3. PR that Big Iron. IBM’s PR machine is a sight to behold. Do a search for “IBM mainframe” at Google News and you’ll see the topic in numerous publications, always with the story framed on around IBM’s attempts to develop the mainframe and its future admins.

———————————————————————————————-

I thought they were gone
Today, I learned they were not
Mainframes are quite cool
–Matthew Scouten, Marist College


So, will the mainframe survive?

IBM is doing its darndest to capture mindshare and young minds in order to make it so, but time will tell. Small is a pretty cool concept, too.