| |
Current Topic: Technology |
|
InfoQ: Scalability Best Practices: Lessons from eBay |
|
|
Topic: Technology |
4:43 pm EDT, May 29, 2008 |
At eBay, one of the primary architectural forces we contend with every day is scalability. It colors and drives every architectural and design decision we make. With hundreds of millions of users worldwide, over two billion page views a day, and petabytes of data in our systems, this is not a choice - it is a necessity. RelatedVendorContent A Technical Introduction to Terracotta Hibernate without Database Bottlenecks Scale Your Application without Punishing Your Database Why Should I Care About Terracotta? Terracotta 2.6 - Download now for scalability without tradeoffs Related Sponsor Terracotta is Scalability and Availability for Java Applications. It clusters the JVM itself, which dramatically simplifies development and reduces database dependency. In a scalable architecture, resource usage should increase linearly (or better) with load, where load may be measured in user traffic, data volume, etc. Where performance is about the resource usage associated with a single unit of work, scalability is about how resource usage changes as units of work grow in number or size. Said another way, scalability is the shape of the price-performance curve, as opposed to its value at one point in that curve. There are many facets to scalability - transactional, operational, development effort. In this article, I will outline several of the key best practices we have learned over time to scale the transactional throughput of a web-based system. Most of these best practices will be familiar to you. Some may not. All come from the collective experience of the people who develop and operate the eBay site.
InfoQ: Scalability Best Practices: Lessons from eBay |
|
Topic: Technology |
1:25 am EDT, May 29, 2008 |
The Internet Layer If transmission has seen dramatic changes in the past decade then what has happened at the IP layer over the same period? The glib answer is "absolutely nothing"! But that answer would be gliding over a large amount of activity in this area. We've tried to change many parts of IP in the past decade, but, interestingly, none of the proposed changes have managed to gain any significant traction out there in the network, and IP today is largely no different from IP of a decade ago. Mobility [1], Multicast [2] and IP Security (IPSec) [3] remain poised in the wings, still awaiting adoption by the mainstream of the Internet.
ISP Column - June 2008 |
|
Topic: Technology |
12:29 am EDT, May 29, 2008 |
Not all forms of internet access were based on dial-up in 1998. ISDN was on use in some places, but it was never cheap enough as a retail service to take over as the ubiquitous access method for the Internet. There were also access services based on Frame Relay, X.25 and various forms of digital data services. At the high end of the speed spectrum there were T-1 access circuits with 1.5Mbps clocking, and T-3 circuits clocked at 45Mbps. If you were an ISP you leased circuits from a telco. In 1998 the ISP industry was undergoing a general transition of their trunk IP infrastructure from T-1 circuits to T-3 circuits. While it was not going to stop here, squeezing even more capacity from the network was now proving to be a challenge. 622Mbps IP circuits were being deployed, although many of these were constructed using 155Mbps ATM circuits using router load balancing to share the IP load over four of these circuits in parallel. Gigabit circuits were just around the corner, and the initial exercises of running IP over 2.5Gbps SDH circuits were being undertaken in 1998.
ISP Column - June 2008 |
|
Software Integrity » Blog Archive » Glitch Watch - Bingo cashier jailed for exploiting glitch |
|
|
Topic: Technology |
12:07 am EDT, May 29, 2008 |
Glitch Watch - Bingo cashier jailed for exploiting glitch May 16th, 2008 by Nigel Cheshire. Posted in Glitch Watch BanburyThe quiet market town of Banbury, Oxfordshire in England is not a place you would expect to see in the news very often. But, according to the Banbury Guardian, a bingo cashier in the town’s Gala bingo hall was jailed this week for stealing approximately $32,000 from her employer. Apparently, the bingo hall uses a swipe card system, called Buzzcard, that allows patrons to load money onto the card and use it to store winnings. When Emma Meechan, the cashier in question, discovered a misfeature in the system that caused about $335,000 to be mistakenly loaded onto some of the cards, she took advantage of the glitch and pocketed some of the money. Meechan was sentenced to serve 8 months behind bars.
Software Integrity » Blog Archive » Glitch Watch - Bingo cashier jailed for exploiting glitch |
|
Market Maker - Energy Speculators Draw the Heat - NYTimes.com |
|
|
Topic: Technology |
10:57 pm EDT, May 25, 2008 |
Next month, Representative John B. Larson, a Connecticut Democrat, plans to go even further, proposing legislation that would essentially ban over-the-counter trading of energy futures by traders who don’t plan to take physical delivery of the commodity. While Nymex trading would be largely unaffected, billions of other trades could potentially be brought to a halt. The idea for the bill, Mr. Larson says, came from local suppliers of heating oil, gasoline and diesel in Connecticut, who say the price spike can’t be explained by simple supply and demand. While advocates defend the futures market as a way of hedging against higher prices, Mr. Larson doesn’t buy it. “We see this as nothing short of greed on the part of speculators,” he says.
Perhaps the middle ground is to only let companies that have a reason to hedge to trade futures? Market Maker - Energy Speculators Draw the Heat - NYTimes.com |
|
Enlightened Perl Organisation |
|
|
Topic: Technology |
9:54 pm EDT, May 25, 2008 |
Enlightened Perl is an organization which is built around a movement within the Perl community. Its goals are complementary to the Perl Foundation. Specifically, we support certain Perl development efforts that ensure perl's future as an enterprise-grade development platform. Its aims are: * To encourage the usage of the Perl programming language as a modern, high-level development platform. * To promote, sponsor and enable the development of Perl modules, add-ons and restorative code that further such use of Perl. * To collect and distribute appropriate funding and support that encourages this development. * To provide general encouragement to the wider Perl community. To provide resources for the community in the manner of, but not restricted to or defined by, internet based sites, articles and code. * To emphasize the concept of a Perl Enlightenment in the continuing development of the Perl community.
Used poorly, Perl is the worst language in widespread usage. Enlightened Perl Organisation |
|
Navicat - the World's Best PostgreSQL Front End for Windows & Mac OS X |
|
|
Topic: Technology |
9:51 pm EDT, May 25, 2008 |
Download for 30-day trial Please click the button below to start downloading the evaluation version of Navicat for PostgreSQL.Navicat for PostgreSQL would support PostgreSQL database server version 8.0 or above.
30 day trial of a very good tool for Postgres. Very helpful for grok'ing large schemas. Navicat - the World's Best PostgreSQL Front End for Windows & Mac OS X |
|
A quick-and dirty MySQL backend for SVN |
|
|
Topic: Technology |
7:06 pm EDT, May 25, 2008 |
The aim is to alter libsvn_fs (and much smaller parts of svnadmin, and libsvn_repos) to support storing filesystem data in a MySQL database. Configuration data will be stored in a repository directory, along with the existing non-BDB repository stuff. The practical reasons for such a scheme may possibly include increased efficiency, integration with existing SQL databases, another interface to the data (using simple SQL statements), and buzzword points. Since, AFAIK, this hasn't been done before, I'm hoping that the one practical outcome of this experiment will be to provide exciting new data on the above.
A quick-and dirty MySQL backend for SVN |
|
Load balancing howto: LVS ldirectord heartbeat 2 | Novell User Communities |
|
|
Topic: Technology |
5:15 pm EDT, May 25, 2008 |
Load balancing howto: LVS ldirectord heartbeat 2 Submitted By jslezacek on Fri. 05.23.2008 Contents: * Environment * Problem * Solution * Goals * Configuration: linux-director (load balancer) o Ldirectord o Heartbeat 2 * Configuration: real servers * Testing the setup * Caveats * Alternative solutions * Conclusion * External links
Neat howto Load balancing howto: LVS ldirectord heartbeat 2 | Novell User Communities |
|
Why Startups Fail - Tech, Startups, Capital, Ideas |
|
|
Topic: Technology |
9:32 pm EDT, May 23, 2008 |
They spend too much on sales and marketing before they’re ready. Many venture companies move to a high burn rate too quickly and it’s hard to go back. Sometimes even a frugal entrepreneur winds up spending too much either because he doesn’t manage the money or is tempted by having money in the bank. This often happens when a startup raises too much money too early. Other times, this occurs with entrepreneurs who are accustomed to having lots of resources. They ramp up sales before the product is ready. Of course, there’s a lot of work required to get sales early on. But a product with a truly great value proposition that delivers in a measurable way will practically sell itself. Companies that ramp sales and marketing too soon waste a lot of money. Sometimes even when the product is great, the sales process itself isn’t understood to a point where it can be scaled: who are you selling to, how much will they really spend, and what profile of sales person does the company need to hire who will succeed at selling that particular product. All of this has to be understood before sales can efficiently scale. Spending on the sales and marketing operations means there is no return if customers don’t bite. When you spend money on the product that work can be leveraged in future versions. (In fact, the key to effective product delivery is to try a lot of things and see what sticks.) For every venture dollar invested, I estimate that more than two-thirds go into sales costs and only a third into product development. Once you up the burn rate, there’s no easy way back.
Why Startups Fail - Tech, Startups, Capital, Ideas |
|