The SDTimes has an article up about a new operating system Microsoft is working on called “Midori”. It is based on their “Singularity” OS, with everything being written in managed code then natively compiled.  Rumor has it that this is the follow-on to the Windows platform… we’ll see if it ever materializes commercially. SDTimes bases the article on some internal documents they got access to, which may be why we haven’t seen this level of detail before (see the entry in Wikipedia). From the article:

According to the documentation, Midori will be built with an asynchronous-only architecture that is built for task concurrency and parallel use of local and distributed resources, with a distributed component-based and data-driven application model, and dynamic management of power and other resources.

The Midori documents foresee applications running across a multitude of topologies, ranging from client-server and multi-tier deployments to peer-to-peer at the edge, and in the cloud data center. Those topologies form a heterogeneous mesh where capabilities can exist at separate places.

In order to efficiently distribute applications across nodes, Midori will introduce a higher-level application model that abstracts the details of physical machines and processors. The model will be consistent for both the distributed and local concurrency layers, and it is internally known as Asynchronous Promise Architecture.

…operating system services, such as storage, would either be provided to the applications by the OS or be discovered across a trusted distributed environment.

Read the rest…

Share/Save

High Scalability has a great link to a video TechTalk with Cuong Do, YouTube’s engineering manager. He talks about the challenges YouTube faces (past and present) to meet it’s skyrocketing user demand, as well as the infrastructure that allows them to scale. I enjoyed the anecdotes: especially the frantic email sent at 2am alerting the dev team that they only had 3 days of storage left… I always thought Google/YouTube would be immune to emergencies like that… ignorance on my part

(requires Adobe Flash plugin… click HERE to watch it on YouTube)

I found this information interesting:

• The application code is written mostly in Python (the web app is not the bottleneck… the database RPC is)
• They use Apache for page content and lighttpd for serving video
• Thumbnails are now served by Google’s BigTable
• They’re running SuSE Linux with MySQL
• HW RAID-10 across multiple disks was too slow. HW RAID-1 with SW RAID-0 was faster because the Linux I/O scheduler could see the multiple volumes and would therefore schedule more I/O

You can read a good summary of the talk HERE from the High Scalability website.

TechCruch has a good article of the YouTube API.

Share/Save

CIO has a good interview with Rob Reeg, president of Mastercard’s Global Technology and Operations. He discusses their infrastructure and processing architecture. Definitely worth looking at if you’re interested in how credit card transactions are processed.

Interviewer: How big of an infrastructure do you have to support and maintain? It must be huge.

Reeg: Actually from a pure server footprint standpoint… we probably have fewer actual footprint servers because of techniques like virtualization that help us leverage one box to do multiple things.

Where it gets interesting is philosophically: We try to put [transaction] processing as close to our customers, the banks, as possible. When we talk about the global network, we have small servers that sit with the bank customers that connect to our network. What it does is it gives us intelligence there at the end of the network. So as a transaction comes through, we can take a look at that transaction and decide how do we best process that transaction for the benefit of all those four parties in the model.

As to processing, the majority of transactions we’re looking at relate to how do we process them as fast as possible in the most accurate way. The way to do that is by peer to peer: If you’re using your card in Europe, in London, say, and you swipe your card as you check out of hotel, we can route that transaction to the hotel’s acquiring bank in London directly to your issuing bank and get that message back for approval without ever going through St. Louis or some big data center in the middle of all that.

You can read the full article HERE.

Share/Save

Update 7/22/2008: The issue may be more complex than it first looks (of course, the media always over-simplifies things). Click HERE to read an insider’s account of the situation.

—

Okay, THIS is funny because of the glaring security mistakes made by San Francisco’s Department of Technology (or Department of Ignorance, after this one). From the New York Times:

A disgruntled city computer engineer has virtually commandeered San Francisco’s new multimillion-dollar computer network, altering it to deny access to top administrators even as he sits in jail…

Prosecutors say Childs, who works in the Department of Technology… tampered with the city’s new FiberWAN (Wide Area Network), where records such as officials’ e-mails, city payroll files, confidential law enforcement documents and jail inmates’ bookings are stored.

Officials also said they feared that although Childs is in jail, he may have enabled a third party to access the system by telephone or other electronic device and order the destruction of hundreds of thousands of sensitive documents.

This is like security 101… you never give this much power to any single person. On critical systems like this, you always have check-and-balances, outside security code reviews, and strict audits. The S.F. DoT was basically driving around without insurance and got in an accident… I don’t feel sorry for them. It’s really sad how ignorant the world is about security (sigh).

Share/Save

TechUser has a great article about possible reasons Microsoft would be interested in purchasing Yahoo’s paid-search business (click HERE for the back story). For those of you new to the subject, paid-search is where advertisers bid against each other to get better/more frequent placement of their ads next to search results (think Google AdWords). This is big business. Believe it or not, it’s what drives Google… search, Gmail, Google Docs, etc all revolve around their ad business.

The afore mentioned article puts forth a compelling argument that Microsoft is only interested in the infamous ‘361 patent held by Yahoo. This is the patent on the whole idea of paid-search. Apparently, Yahoo has been dening Microsoft a good licensing deal on the patent, so Microsoft is retaliating. From the article:

Microsoft is still chafing under Yahoo’s influence and is desperate for unfettered access to the ‘361 patent. It is quite possible that the size of the royalties Microsoft is paying to Yahoo are forcing Microsoft to neglect its paid search operations in order to minimize payments to Yahoo, and to minimize the size of an eventual settlement with Yahoo.

Microsoft is completely aware of the ludicrousness of its attempts to buy Yahoo’s paid-search assets and Microsoft’s earlier acquisition bid seems to have been an attempt to soften up Yahoo’s opposition to a paid-search asset acquisition.

The entire theory is interesting. Click HERE to read the full article.

Share/Save

Systems Management News has up an interview with Ron Gorodetzky, enterprise architect for Digg. It’s an interesting look at the challenges Digg faced scaling to meet it success (over 26 million unique visitors a month). They’re using a LAMP stack (Linux, Apache, MySQL, PHP), with MogileFS as their backend distributed file system. To help manage their infrastructure, Digg uses Puppet.

Ron highlights a commom problem all architects face when they try to scale their software: the database.

“The first pain point we hit was just database stuff. The first thing you’ll notice is when you start to grow these queries, the database can’t commit as much time to committing a certain query as it used to,” said Gorodetzky. “You’ll find the normal things that work, suddenly don’t. You’ll find that, one day, you’ll see a spike in your graphs telling you that something’s going slower. Once you do that, you get to the point where the database part is as fast as it can be, you cache things.

You can read the full article HERE.

Share/Save

A group of researchers from Indiana University, Harvard Medical School, et. al have completed the first map of the outer layer of the brain. While this is cool research, I think people tend to extrapolate this stuff out too far… more on that later. Here’s a quote I found interesting:

“This is one of the first steps necessary for building large-scale computational models of the human brain to help us understand processes that are difficult to observe, such as disease states and recovery processes to injuries…” [emphasis mine]

As an engineer, I’m more interested in modeling the brain for artifical intelligence reasons (call me narrow and selfish if you want). To be clear, I don’t believe that simply cloning the brain will create a living conciousness… I’m more interested in augmenting our existing reasoning power (controversial, I know).

I tend to question the comments made by one of the researchers (maybe I’m reading too much into it):

“We can measure a significant correlation between brain anatomy and brain dynamics. This means that if we know how the brain is connected we can predict what the brain will do.”

Maybe on some generic scale, but I consider us more than just a biological computer. I’ll lay out why I believe this using inductive logic. First, I believe people are fully responsible for their own actions (using the Bible as my base). To be responsible, a person has to be able to make a choice about what actions they will take (free-will vs determinism). Now, assume we are just a biological computer, with predictable, deterministic actions. If this is true, then our behavior is dictated by our biology and we only appear to have free-will. Without free-will, we are not responsible for our actions. Thus, this can’t be true.

Okay, now some would argue that we are a biological computer but with some non-deterministic properties. If this is the case, we still have no responsibility for our actions because they are simply slaves to these random fluctuations. Thus, I hold firm to my belief that we have a soul that transends the physical.

You can read the full results of the study HERE. Or as a PDF.

Share/Save

Okay, I love stories like this: someone’s found a loophole in the AT&T/iPhone marketing schemes. Apparently, the cheapest way to get an unlocked iPhone is to sign up for a 2 year contract with AT&T, get a subsidized iPhone for $199, and then cancel the contract and pay the early termination fee. Again, this only applies if you want an UNLOCKED iPhone… if you’re okay with AT&T, just keep the subsidized plan. You can read the full article at Gizmodo HERE.

Here’s the math from the Gizmodo article:

$299 - 16GB iPhone w/contract
$36 - Activation
$175 - Early Termination Fee
$70 - One month of service

Total Price Using The Shady Method: $580

So…

8GB iPhone Canceled: $480
16GB iPhone Canceled: $580
8GB iPhone Unlocked from AT&T: $599
16GB iPhone Unlocked from AT&T: $699

Are we missing something?

Share/Save

Okay, so I’m on this research kick to investigate the infrastructure solutions of some of largest distributed systems in the world. Given that most of those are internet companies… well, that should explain the trend in my posts (more to come).

Baseline has a good article (albeit a little dated… 2006) describing Google’s server infrastructure and some of the history behind it. Google really pioneered the use of COTS (Commodity Off-The-Shelf) hardware to reduce cost; an idea that works great in some situations. They’re up to around 450,000 servers worldwide now… quite impressive. Click HERE to read the full article.

Here’s an except I found interesting regarding their infrastructure history:

Google and its information-technology infrastructure had humble beginnings… when the server infrastructure consisted of a jumble of PCs scavenged from around campus.

“But this is the start of the story,” he adds, part of an approach that says “don’t necessarily do it the way everyone else did. Just find some way of doing it cheap and effectively—so we can learn.”

By 1999, the Google.com search engine was running in professionally managed Internet data centers run by companies like Exodus. But the equipment Google was parking there was, if anything, more unconventional, based on hand-built racks with corkboard trays…

His team assembled racks of bare motherboards, mounted four to a shelf on corkboard, with cheap no-name hard drives purchased at Fry’s Electronics. These were packed close together (like “blade servers before there were blade servers,” Merrill says). The individual servers on these racks exchanged and replicated data over a snarl of Ethernet cables plugged into Hewlett-Packard network switches.

… corkboard, huh… not a good idea:

Later Google data centers tidied up the cabling, and corkboard (which turned out to pose a fire hazard) vanished from the server racks.

Read the rest…

Share/Save

Okay, I’m always interested to know how things work… especially when there is a correllation to what I’m working on (large distributed systems). Data Center Knowledge has a summary of Wikipedia’s back-end infrastructure and links to more in-depth presentations. Take the plunge if you’re nerdy enough…

Click HERE to read the article.

Click HERE (PDF) for the in-depth details.

Share/Save