Open Source Telecom, Time to Embrace Web Economics
by Guest Editor - Mar. 04, 2008Comments (9)- Related Blog Posts
- Memcached: A Good or Bad Sign for MySQL?
- Open-Xchange Finds Free Webmail Promotion Boosting Its Bottom Line
- Great Insights From PGConf 2008
- MySQL to Offer Proprietary Add-Ons
- Recession: A Boon for Open Source?
Written by Allan Leinwand, a venture partner with Panorama Capital and founder of Vyatta. He was also the CTO of Digital Island.
The use of open-source software and commercial, off-the-shelf (COTS) hardware is making a profound impression on the telecom market -- one that seems destined to follow the path blazed by web economics. To date, the telecom industry has not yet fully embraced open-source and COTS hardware, something that I believe is a fait accompli -- regardless of any course of action taken by traditional vendors.
The overarching vision for open-source telecom is to enable telecom economics that are more in-line with web economics. For web-based businesses, massive use of open-source software and COTS hardware has enabled advertising-supported business models that provide truly disruptive services to the end user: free email, free videos, free storage and processing power, and so on.
From these fundamental businesses, services proliferate and innovation gets turbo-charged. By contrast, today's telecom vendors rely on business models that drive up end-user prices using inflexible, proprietary and expensive components to support lofty product margins. In the world of web economics built on open-source software, this model is destined to break.
Embracing open-source telecom starts with the core assumption that the telecom market is not inherently different than the numerous other markets served by open-source software. While proprietary telecom vendors have spent decades on software and have convinced some that their algorithms are more sophisticated and performance-sensitive that those of other markets, I question that theory. Linux -- an entire operating system that runs a multitude of complex applications and algorithms and supports numerous devices -- is open-source software. MySQL -- a complex database behind countless web transactions -- is open-source software. And these are just two examples in a sea of complex open-source software systems available today.
Are the proprietary vendors foolish enough to think that the market believes the software algorithms in telecom equipment are more complicated than an entire operating system or database system? As evidenced by examples such as Asterisk, Snort, and Quagga, if done correctly, open-source software can be built to handle the complex algorithms for telecom markets.
Another principle driving open-source telecom is the belief that the performance of COTS hardware is sufficient for a large portion of the telecom market. Using COTS hardware and open-source software for telecom does limit the total available market; it is clearly not feasible for the highest end of the market (such as a service provider’s core router network) nor for the lowest end of the market, which requires limited functionality and sees very high volumes (such as the small-office home-office devices sold by vendors such as Netgear and DLink).
However, for everything in between, the combination of open-source software and COTS hardware perfectly hits the mark – using Intel-based processors and associated hardware ecosystem provides more than adequate performance at web economics prices and only promises further performance in the near future. For example, in the open-source networking market in which Vyatta, a company which I helped co-found, operates, there are multiple ways to make telecom economics more like web economics. While proprietary vendors charge $3,500 for one gigabyte of memory in their proprietary systems, the COTS hardware equivalent memory is less than $100.
The exact same ratio can be seen for a gigabit Ethernet port offered by a proprietary vendor vs. the exact same port delivered in a COTS hardware package. The performance of the memory and gigabit Ethernet port in both environments is identical (and are most often using the same underlying silicon chipsets). It is the leverage achieved by using open-source software on top of COTS hardware that is game-changing economics.
Some telecom vendors (3Com, Cisco, Juniper) have begun to mumble the word “open” in their marketing literature. I believe this is a stop-gap approach to freeze the market from racing toward open-source software -- these vendors know full well that truly providing open-source software will cripple their current lofty product prices and margins. Telecom vendors need to embrace the fact that users in a majority of their markets have embraced web economics and those economic realities are here to stay. The use of open-source software and COTS hardware will create a foundation for more services at lower end-user prices -- the same catalyst that sparked Web 2.0. This foundation will, in turn, allow telecom vendors to be competitive, stay relevant and embrace the changes coming their way.
Please tell us what you think. Do you think open source software can have a large, ongoing impact on telecom?
Share Your Comments
