By Brian Boyko

Although the stock market seems to have started recovering, trading in the New York Stock Exchange took a heavy tumble yesterday.

There are two main theories about the market plunge. The first was a $100B drop in the value of the Shanghai stock market.

The other is computer related. According to the Seattle Times, Dow Jones & Co.'s computers weren't up to the task of handling the huge volumes of trades at the NYSE - a spike of 4.5 billion trades, or twice the usual average. Switching over to a backup computer caused a massive "catch-up" in the Dow Jones industrial average - which recorded an instant plunge of 200 points, which couldn't have had a positive effect on investors, who reported that computer systems were slow to respond.

(Continued...)

Continue reading "Performance Problems Lead to Drop in the Dow" »

Business Intelligence Lowdown: How to be a Terrible Network Admin - 25 fail safe tips.

I guess "practices to avoid in network administration" didn't have quite the ring to it. Most of these are common sense tips; but because there's so many of them, even the best of us could use a reminder - there's probably something you'll read about and realize that you missed something.

Anyway, we really like #3:

3. Say no to knowing your network: Monitoring your servers and your network interface is unnecessary, simply because you have no need to know how much traffic is traveling about your network. Regular observation of your network provides you with knowledge on normal, trouble-free usage, which in turn helps you identify problem spots and potential causes when you're called to troubleshoot and plan for future growth. This is not necessary for a terrible network administrator.

ComputerWorld: How IT Makes Johnny More Productive.

This story by Kathleen Melymuka talks about a study from Sinan Aral and Erik Brynjolfsson about exactly how IT helps worker productivity by studying individual executive recruiters and their interactions with technology in the workplace.

If you look across e-mail and social networks, database and phone, the surprise was that overall, IT use is not associated with an increase in speed. In fact, it's associated with slower speed. But we found that heavier IT users are much heavier multitaskers, so over time, they're completing more projects and bringing in more money for the firm.

Anyone who tells you more technology is associated with faster speed has not looked at the story of the Franklin stove.

Throughout much of American history, technological improvements have been marketed as labor saving devices. But cooking times increased dramatically when the "labor saving" stove was invented - because people were no longer content with the simple meals that can be cooked over open pit-fires, and asked for more time-intensive meals that were now possible.

Similarly, despite many advances in technology, it takes a person roughly the same amount of time to clean a house as it did to clean a house in the 19th century - it's just that our standards for cleanliness have become much higher.

So this seems entirely plausible to me - that we get more done, but we don't necessarily get things done faster.

By Tim Smith, John Mao, and Ben Erwin. Compiled by Brian Boyko.

Dale Imler wrote in:

I would like to see some information on how to use the SuperAgent and Reporter Analyzer for executive reporting and capacity planning.

The big consideration with capacity planning is that it shouldn't be done in absence of performance considerations, and you can't consider performance with insufficient data. Mostly, SuperAgent helps to positively identify those cases where the solution to intermittent performance is increased capacity, and help eliminate false positives where the solution might be something completely different.

Additionally, SuperAgent can measure the degree in performance gain you can expect following an infrastructure upgrade. If you have WAN circuits of various types, SuperAgent shows application performance over each type, and may reveal patterns associated with that type.

If you're using some sort of WAN optimization, you should keep performance in mind when making utilization based capacity planning, because some types of WAN optimization are designed to raise utilization rates in order to extract better performance. So, if WAN optimization is present and performance is excellent, high utilization needs to be interpreted differently than if WAN optimization wasn't there. Again, this means measuring end-to-end performance with a product like SuperAgent.

(Continued...)

Continue reading "In response to Dale Imler's question..." »

What would you like to see us cover in the next few weeks?

DailyWTF: Bunker Buster

Talk about misplaced priorities. This story from the DailyWTF (ah, DailyWTF - what would I do without you?) starts with a story about poor network performance, but ends with the world's most secure insecure server.

The President smirked. As they waited for the elevator, he told Jack about the facility. It was a decommissioned military bunker with a 40,000-square foot, two-story underground facility. They had air scrubbers, generators, food, water -- enough supplies to last completely cut off from the world for at least two months. They even had decontamination showers and holding cells. Yes, holding cells.

I know network security is a big deal, but I've never known an IT department to need a holding cell. (Decontamination showers, on the other hand…)

Mike-O-Matic: The Three Kinds of IT Shops: Which is Yours?

Mr. O'Matic breaks down the different types of IT departments into three groups - places that think of everything in terms of databases, places that look at everything in terms of application frameworks, and places that put an emphasis on design and the front end.

If you are a developer working in a culture that is different from how you approach technology, you are likely to butt heads a lot with your coworkers. It's sort of a square-peg-in-round-hole situation. Once I was working at a place that was squarely in the SQL Squad camp. Our inventory management software was very feature-rich, but was overkill for many of our customers. As the person who handled all of our customer support, I was acutely aware of this pain. "What the hell is a Master Item?" they would ask. "Why can't I just add a jar of pickles?"

Unfortunately, you need to have at least 13 hearts before you pick up the Master Item. However, you can get the pickles from the old man under the tree for 70 rupees.

IBM: Windows to Linux Roadmap

The article isn't surprising - the source is. This is a roadmap for developers migrating from Windows to Linux from IBM's e-business architect, Chris Walden.

This roadmap is designed to help you take the experience and knowledge that you already have in computing and redirect it to working in Linux. It's not the only reference you'll ever need, but it will help you get past some of your first obstacles and adjust to a new and, I think, exciting approach to computing. As you follow this roadmap, you'll discover many new resources to help you learn, troubleshoot, and manage Linux.

by Dr. Cathy Fulton

WAN optimization devices are certainly great pieces of technology. They do much to alleviate the problems that companies have with keeping applications which travel over the WAN responsive. But deployment is not without risks and companies need to look at cost vs. performance benefits.

There are several benefits to WAN optimization. One is offloading WAN links – when there is less congestion every application performs better. Another benefit is minimizing the impact of latency on the network. Even if it doesn’t reduce the traffic on the network, TCP and application optimizations can reduce the effective latency experienced by your data.

You might even get a benefit from increasing WAN utilization while still receiving good performance - by fulfilling a latent demand. A good example of this would be a company that receives orders over a network. With WAN optimization, the holy-grail scenario is that people who had been frustrated because of poor performance would now place orders on a more responsive order system. You have to have exactly the right circumstances and environment to hit the points where utilization goes up while end-to-end response time decreases, but it is certainly possible.

But WAN optimization is not a panacea. It works well on a subset of performance issues, but can actually be detrimental for others. In terms of mitigating the effects of congestion, it works best on small to mid timescales; it certainly does not eliminate the need for longer timescale techniques including capacity planning. And it does not work well with every application – some perform worse when “optimized”. If you’re interested in what sort of benefit you’re going to get from your WAN optimization efforts, you need to look at the amount of congestion on your link and the composition of that traffic.

(Continued...)

Continue reading "WAN Optimization Devices: What you need to consider before deployment." »

DailyWTF: Practice Makes Perfect

Rule of thumb - when you're given 6-12 months to do a SQL server upgrade, don't just come in one weekend and upgrade all the servers.

When Kevin returned from vacation on the day following New Year's, he noticed an email sent from the issue tracker with the subject "Test email from SQL2005 Upgrade." After logging in to the issue tracker server, he went directly to the manager to ask why the server was upgraded.

"Oh that," the manager replied, "that was because I needed practice doing an upgrade."

Dmiessler.com: Lsof

This Linux/Unix command gives you the ability to show all your network connections, show only your TCP or UDP network connections, show all the networking related to a given port, a specific host, and much more.

lsof is the Linux/Unix über-tool. I use it most for getting network connection related information from a system, but that's just the beginning for this amazing and little-known application. The tool is aptly called lsof because it "lists open files". And remember, in Unix just about everything (including a network socket) is a file.

I love Unix, but why is it that on one hand, you have file names like "ls" for "list" and "rm" for "remove" and "lsof" for "lists open files" and on the other hand you have names like "gstreamer-libdvd-0.3.0-1.i386.rpm."

CodeIDE: Live Coding Interface

This is just the coolest thing ever if you're a beginning or hobbyist programmer. Type or paste the text of your Basic, Pascal, C++, Perl, Javascript, or HTML code, and test your input and output on the web page itself.

By Joel Trammell

Soon, March Madness will come up and, like every year, there will likely be increased strain on corporate networks as people want to watch the games online. So wondering whether or not your network is ready for March Madness is a topic worthy of discussion.

But being prepared for the spike in network traffic generated by March Madness is only a small part of network preparedness. Unlike many other business assets, networks are very susceptible to external events.

Most corporate assets can be controlled, and the decision makers in the organization know how those assets are going to be used, or not used.

If I have a building with 472 offices, and I hire 473 people and want give to everyone an office, I have to move to a bigger building. But that’s not something that just happens without warning. Two hundred people don’t just walk into the building one day and expect to get a job. That’s something that I can control and prepare for.

The problem with the network, as a corporate asset, is that it’s no longer a closed system, like it was in “the old days.” The network is now affected by external events – be it a major news story, a major virus – it’s usually some unexpected demand that can’t be possibly predicted and can affect the network with little or no warning. March Madness, at least, can be predicted.

(Continued...)

Continue reading "March Madness and other Network Crises: 4 Questions You Should Be Able to Answer" »

By Ben Erwin, Product Manager for NetQoS Performance Center 3.0
& Steve Harriman, Vice President of Marketing at NetQoS:

We've recently held a Webinar to discuss NetQoS Performance Center 3.0. We had a Q&A session there. We couldn't get to all of the questions asked at the Webinar, so we've done our best to answer those questions here in this blog.

We also have a copy of the earlier Q&A posted, as well as a link to the entire Webinar.

(Continued...)

Continue reading "Q&A on NetQoS Performance Center 3.0, Part 2" »

WebTVWire: Texas Judge Denies "Deep Linking" Appeal.

To get to this article:
1) Go to www.webtvwire.com.
2) Browse to February 9th, 2007 in the listings.
3) Click on the story that is entitled "Judge Denies Appeal that Deep Linking Directly to Video and Audio Streams is Not Illegal"

We could just tell you to go to http://www.webtvwire.com/judge-denies-appeal-that-deep-linking-directly-to-video-and-audio-streams-is-not-illegal/ - but apparently that's illegal in Texas.

The judge ruled that "the link Davis provides on his Web site is not a 'fair use' of copyright material" and ordered all links to the audiocasts to be removed. SFX sued supercrosslive.com because they believed that listeners should only access the audiocasts through the website, in order for logos of sponsors to be seen.

"Deep Linking" - or linking to an individual Web site page, as opposed to linking to the index site of the page - is an invention. Before business found its way onto the Internet, the system was designed so that you could link to any page on any Web site that you wanted - the entire point of it was to organize information, filtering out the information you don't need from the information you do need. In this case "sponsor logos" are most certainly information you don't need.

Network World - Layer 8: VoIP services can cause a home security alarm system to malfunction.

While this is mainly for the Vonage customer that also has a home security system that dials-out through the phone line, this is something to keep in mind when moving to a VoIP rollover in the enterprise - if your security system uses the phone line.

* VoIP requires continuous power. If you lose power, you will not be able to use any phone connected to VoIP. More importantly, the security system will not send a signal to your monitoring center.

* Software changes to the alarm system or updates from the service provider could render the system useless.

Now, granted, enterprises typically have backup generators and the like to continue to provide power, but you never when British guys playing German guys with fake American accents (to fool Bruce Willis) intend to rob your company of bearer bonds on Christmas Eve.