Friday, August 31, 2007
Since it is Blog Day 2007 I thought I would put my two cents in (or 5 links). More than 90% of my blogroll is tech related, so this is some what of a challenge to come up with non-tech blogs. I like the idea though and hope that you find these enjoyable.
1. thesimpledollar.com Financial Talk for the rest of us
2. 24-7 Family History Circle Blog from Ancestry.com
3. Dan Pink Author of A Whole New Mind (favorite book of mine I read recently)
4. Happy Neuron I just like the title :) (good site though)
5. Digitally Imported My favorite online radio station
Monday, August 27, 2007
I immediately went to check blogs from Chuck Hollis and Dave Hitz to see if they had anything to say on the topic. They didn't. Why? Because it is silly. First of all it's a term....nothing more. Journalists and marketing geeks like nothing more than to coin a term to spin the hype machine and see who will pay attention. Storage Service Provider (SSP), SAAS (Storage as a Service), storage cloud, Managed Storage Service....whatever. It is all just 1's and 0's on platters.
Reading the article took me back a bit, as a number of years ago I was a part of a dot-com company that offered online storage. We were like the rest of the dot-coms and offered 1gb, 5gb and 10gb plans, and then had a backup tool that would take your data directly to your online account. We ran the site for many years, and I always remember looking to one company as 'the' big dog of the industry. Storage Networks was the SSP of the time and (I think) had grown quite a large business. They took the enterprise side of the market, while sites like idrive, xdrive, swapdrive and others took the consumer side. Take a look to see what Storage Networks site looked like in 2000.
Today, we have Amazon's S3. I'm still amazed at this offering and almost equally amazed that it hasn't caught on more than it has. I was trying to think if S3 had any real competition and the only thing I could come up with is an equally intriguing offering from Cleversafe. It is fun to see how people are using S3 though. Check out this article about Stardust@home:
The Stardust@home project uses the Amazon Simple Storage Service (Amazon S3) to store and deliver the tens of million of images that represent the data collected from the dust particle aerogel experiment.
Anyway....back to the SSP model. The model is an enterprise one. I really believe that S3 and other online storage services that have survived can only truly cater to home/office users, small companies and some aspects of the hosting world. Enterprise customers want their data close to them and want to know it is managed, protected and monitored at all times. Regulations and data leakage incidents have ensured that enterprise storage is on a short leash at all times.
I do believe that managed storage services (see - even I can't stop from using IT lingo) have a place for some. Inside the data center, or between trusted parties and elaborate SLA and other agreements, it would potentially be beneficial to let someone else manage the scalability, reliability and a few other 'ilities of your storage. If nothing else, perhaps the consumer side of the market will help to push down some of the insane prices that the enterprise players push. Maybe a colo or integrator could offer a pay-for-what-you-use model to the SMB market and see if it flies.
Thursday, August 23, 2007
Wednesday, August 22, 2007
I really wish I could drive down there and snap some shots of construction -- assuming that I could even get anywhere near enough to get a glimpse.
Check out the article here.
I personally like the term grid (for them) better, but I can see a fit with virtualization as well.
"We're actually a lot like VMware in the problem we're solving," said Bernardin. But instead of creating virtual machines, DataSynapse creates "application instances" to maximize application performance, decoupling applications from underlying resources to improve scalability and resilience and set priorities.
"That requires an underlying platform, such as a grid," said Bernardin, who calls grid technology a "precursor" to such advanced functionality.
They expect 50% sales growth this year and a potential IPO next year! I ran across them last year and really like their product offerings. I spoke with a partner rep at the company because I have an idea for a product offering of my own that would incorporate their product. Their GridServer and FabricServer products are extremely cool and worth a look.I would 'almost' venture to say that they are a good target for merger/acquisition. Perhaps the IPO is a backup plan. Adding/acquiring something like 3Tera's Applogic might be a nice complement....?
Check out the article here
Tuesday, August 21, 2007
The first one is well over my head, but extremely cool. It's a year old, but I imagine some of the concepts still apply (or have been revised/improved/tweaked). It's called Hardware Virtualization Rootkits, from Dino A. Dai Zovi and can found here. If you are in to security at all, this guy does some insane research.
The second one I ran across was from a favorite blogger of mine - Christofer Hoff. It's a presentation on Virtualization and Network Security. It's an excellent mix of Virtualization concepts, vulnerabilities and solutions from his former company, Crossbeam. Check out the blog post here and presentation here.
The final is a presentation entitled The Virtualized Rootkit is Dead from Matasano, Symantec and RootLabs. It discusses HVM malware, virtualized malware detection and the Samsara framework. (Dino Dai Zovi is on the Matasono team)
Wednesday, August 15, 2007
- Outside 2.0: As a computer geek, information junkie and perpetual web surfer, this getting outside thing is pretty cool. My family takes an annual trip to visit relatives and a particular town we visit is a wonderful reminder of the real world, wonderful scenery and the perfect place to ditch the phone and laptop (besides, there is no wi-fi hotspot for MILES. On the second day we went to a St. Louis Cardinals game (the 2006 World Champion Cardinals). Although my sons' hero didn't win the game for us like last year, they did win and it was a good (but really really hot) game. The relatives house where we stayed did not have an internet connection, but their neighbors were nice enough to have open wireless access points for me...and a few that had the default router login/password still on it. :)
- I was able to catch up on a number of podcasts that I listen to and try a few new ones out. I was glad to see that Om Malik has a podcast on the red-hot media network Revision3. Since Om made a few (much more qualified) predictions, I thought I would give it a try also. He stated that he thought Ebay will buy Second Life; 'I' Think Google will buy it and integrate a version of a virtual world into Google Earth. Ok, maybe this is just my dream....but it could happen.
- I was also able to read a few books I had waiting in the queue. I listened to an audio version of Al Franken reading his book Rush Limbaugh is a big fat idiot. It was very funny and cool to hear Al reading it. I also read Blink, by Malcom Gladwell. This was a good book as well and now I need to back-track and read the predecessor The Tipping point. I'm thinking of a subscription to Audible as a Chrismas Present this year.
- As an information junkie, I couldn't stay away from blogs for too long. Rich Miller kept me up to speed on the industry and I went back to read several posts from fellow Iowa blogger Trent Hamm. Trent writes a very good (and popular) blog, The Simple Dollar, on 'financial talk for the rest of us'. His story and how he finds the time and ideasvvfor writing his blog was/is inspiring. Check his blog out here.
- For whatever reason I decided to peruse the quarterly 10-Q statements from Terremark, Equinix and Savvis. They were interesting in their lawyer and SOX-filled ramblings, but one thing caught my eye. From my last post, I had done a Google trends search for the term "data center". Interestingly, the second place region that is searching for that term from 2004-2007 is Singapore. In the Equinix 10-Q I read:
"In March 2007, the Company entered into long-term leases for new space in the same building in which the Company’s existing Singapore IBX center is located (the “Singapore IBX Expansion Project”). Minimum payments under these leases, which qualify as operating leases, total 3,674,000 Singapore dollars (approximately $2,394,000 as translated using effective exchange rates at June 30, 2007) in cumulative lease payments with monthly payments commencing in the third quarter of 2007. The Company is building out this new space in multiple phases. As of June 30, 2007, the Company incurred approximately $11,200,000 of capital expenditures to build out the first phase."Singapore seems to be a hot spot, and I have to wonder if we'll see announcements out in the near future about data centers being built there.
- In my Gmail I found a link to a Syska-Hennessy white paper that (due to it being vacation) I had some time to read. It was on outside air economizers and a very interesting read. My new term for the month is enthalpy ( In thermodynamics and molecular chemistry, the enthalpy or heat content (denoted as H or ΔH, or rarely as χ) is a quotient or description of thermodynamic potential of a system, which can be used to calculate the "useful" work obtainable from a closed thermodynamic system under constant pressure. ). As usual, I most likely wouldn't do it justice in trying to explain it, so check it out here.
- Last, I worked some more on a Data Center Site Selection post I have been developing for some time. Through surfing various sites I ran across an article that mentioned the "Pandemic severity index". From a DR and BCP perspective, I thought this was pretty interesting. Check out the Wikipedia article on it here.
I still have a few days left of vacation -- perhaps I should get back to the outdoors and enjoy it while I can. It may take several weeks to get caught up on work, but the time spent with family, 100+ degree temperature and "are we there yet" car rides was definately worth it.
Wednesday, August 08, 2007
Cisco declared Data Center 3.0, Eric Schmidt declared Web 3.0 and I'm going to declare Data Center 3.5.
This Infoworld article made the dare, so I thought I would oblige. :)
Ok...so I suppose I could attempt to put a little bit behind the statement. Let's take a look:
- Data Center 1.0: Everything pre-2002 : from small server rooms all the way up to Exodus and other dot-com boom companies. BTW: When searching for Exodus, I found it funny that this is what I came up with. The Equinix IPO was August 11, 2000
- Data Center 2.0: 2002-2006 : Post dot-com bubble rebuilding. The industry re-groups and numerous other colocation companies come on the scene. Many of the big players start to out-pace the S&P 500.
- Data Center 3.0: 2006-July 2007 : Virtualization technologies; Mega data centers being planned and built by Google, Microsoft, Yahoo and others; Eco-friendly data centers; Blade Servers ; and data center in a box.
- Data Center 3.5: Starting August 8, 2007 : Virtual private data center ; Utility computing, EC2, insane grids and 4+ Billion Megawatt-Hours of electricity per year (just in the U.S.!). I know these are all current technologies, but I think they will really gain momentum from here on out.
Data Center vs. Data Centre
Data Center companies in 2007 (I thought the 365 main spike was kind of funny)
Hopefully Public Relations 4.0 is next.
Monday, August 06, 2007
Sunday, August 05, 2007
Other things I have read here and there, certainly show Google's interest in the global market, underwater fiber to Europe/Asia and peering arrangements.
Check out the Data Center Journal article here.
Google's Fiber (older story) here
Google jobs: Strategic Negotiator-Global Infrastructure
Saturday, August 04, 2007
I had an idea recently and had it 'somewhat' confirmed by my favorite podcast, This Week in Tech. It's potentially far-fetched, but is, for the purpose of this blog, data center related. Before I get to what I am thinking, some background information:
- Google searching for growth in cell phones.
- Google is talking with carriers, pushing the mobile versions of its applications and prototyping it's own phone. If I remember right an iinovate podcast with Eric Schmidt had him mention 'mobile technologies' as the market to watch / grow in the future. Google has relationships with Vodafone, T-Mobile, Sprint and others. Eric Schmidt sits on the board at Apple (i.e. iPhone). Check out these articles for further information:
- Google's $4.6 billion bid for the 700mhz spectrum
The FCC also wants to spur commercial wireless innovation, especially when it comes to creating a "third pipe" for broadband that can provide an alternative to cable/DSL duopoly that prevails in most of the country.
- Google acquires GrandCentral - a follow-me phone service with a ton of functionality.
Ok -- so where I am I going with this? Google obviously has plenty of mega-data centers going up around the world to house the amazing server infrastructure needed to satiate search and applications. But what about customer-access to the net via a 'third pipe' or other
wireless ventures? Well, we all know the story about shipping containers in Google parking garages in Mountain View. What if they had the containers put all over the U.S. and
used them as a small data center to connect communities, serve the immediate area and dominate the third pipe?
Ok, I can't seem to explain my idea the best, but basically -- the have the containers, they'll have the spectrum, customer applications and of course, the cash to lay out the infrastructure. I would think there is a minimal amount of infrastructure (networking and servers) needed to serve a particular market and if they put all of that in a container, dropped it on some of their
dark fiber and then had the mobile devices to connect....well, viola! Now -- let's hope they provide a little better physical security for these containers than the default Sun Black Box.
It's been a long day today (garage sale and tired kids), so I apologize if I am rambling and don't make sense. Thanks again to Leo Laporte of TWiT for confirming and discussing the thought.
Thursday, August 02, 2007
Besides being an excellent paper to read, I wanted to quote what I found to be a funny story inside the paper. It is nice to hear that large corporations can have large blunders. Here is an excerpt from the paper:
"...a company that invested $22M in blade servers without including the site facility representatives in the decision. Just to install the blades, a site infrastructure upgrade investment (site CapEx) of an additional $54M was required For three years of operation, another $21M was required for site OpEx."
Check out the white paper here