Optimization Tackles the Capacity Crunch

Mon, 09/20/2010 - 8:18am
Maisie Ramsay

YouTube ate up one-tenth of the world’s bandwidth during the second half of last year, and for good reason: It’s not set up for wireless.

Videos posted to YouTube are meant to be viewed on laptops and PCs, not the miniaturized screens of smartphones. YouTube videos are full of extraneous bytes that can’t be used by cell phones, like stereo sound instead of mono and unnecessarily high resolution.

The extra data contained in the videos moves from YouTube to the wireless network, clogging up already congested data pipes, only to be discarded by devices not set up to handle stereo sound and high-resolution video.

Cisco estimates that mobile video will comprise 66 percent of wireless data traffic by 2014, and carriers aren’t sitting idly by and waiting for their networks to get stopped up. They’re turning to device optimization technologies that make sure data is sent to devices in as efficient a manner as possible.

“Today, some form of optimization is in place with more than half the operators around the world, and those who have not deployed it yet are in the process of getting the technology,” says Randy Cavaiani, a former Novarra executive who now  leads sales and marketing for Nokia’s operator web services group after Nokia acquired the optimization company last spring.

“Even operators who have not previously seen issues are seeing their pipes get clogged pretty quickly because of the mobile Internet,” Cavaiani says. “The problem is raising its head around the globe, and operators are looking at what they can implement quickly.”

Nokia’s solution to this issue is a distributed browser that consists of a server in the cloud and an optional microbrowser on the end-user device. The distributed browser reduces the total payload sent over a wireless network by up to 95 percent by stripping out data that can’t be used by phones and re-rendering websites for better viewing on a handset’s small screen.

Nokia’s technology compresses images, video and other data, reprocesses CSS and Java for use by handsets and reformats Web pages so that a user only has to scroll vertically. This all happens in real time, and Cavaiani says the optimization is invisible to end users, who will only notice an increase in speed.

Depending on the complexity of the website, feature phone users may see the time it takes to render a Web page drop from 30-60 seconds to just eight to 10 seconds, while smartphone users could see rendering time drop to about 5 seconds for some pages.

OpenWave has seized the opportunity in the mobile video space with optimization technology that keeps the user experience in mind by conforming to VQM (video quality management) standards, which compare the quality of original content to the quality of optimized content.

“Many vendors have a one-dimension view of optimization, which is reducing the number of bits going over the network,” says Ram Rajagopalan, senior product marketing manager at OpenWave. “Our solution focuses on the user experience.”

Rajagopalan says OpenWave has fine-tuned its algorithms to allow the company to push the limits on how much data can be compressed before a video starts to degrade in quality.

OpenWave and companies like Bytemobile also use technology that matches the data pipe available to the end user with the bit rate of the content and can adjust to changing network conditions so that a video being watched during off-peak hours will not receive the same amount of compression as a video being watched during the middle of the day.

“The majority of data consumption comes not from Facebook, search or Web browsing but from people watching video on laptops and smartphones,” Rajagopalan says of the traffic on wireless networks. “The real value [in optimization] comes when you can reduce the number of bytes transferred over the network.”

The 4G networks being deployed by the likes of Clearwire, AT&T and Verizon Wireless have been touted as the solution to the capacity crunch, reducing the importance of optimization, but Bytemobile isn’t worried about losing customers to 4G networks any time soon.

Joel Brand, vice president of product management at Bytemobile, believes demand will continue to outpace capacity even with the deployment of next-generation networks.

“It’s endless, the amount of content that we will consume,” Brand says. “As a result, I don’t think any 4G, 5G or 6G technology is going to solve the problem.”

Brand believes that optimization of Web content will continue to be important no matter the network technology. “Carriers are competing with each other on quality of experience," he says. “By allowing them to make better use of their pipe, we’re making the experience better to the end user.”

Part of Brand’s bullishness stems from the fact that device optimization is not just about cutting out unnecessary data; it’s about improving the performance of handsets and providing operators new ways to monetize their networks.

Optimization technology can be leveraged to improve things like battery life and processing speed, as well as layering value-added services on top of Web browsers. Already, the guts behind optimization technology is being used by T-Mobile USA to throttle users’ network speeds once they exceed 5GB.

Compressing video and tweaking Web pages for better viewing on handsets isn’t the end-all-be-all solution to the capacity crunch, but it can go a long way toward addressing the problem.

The optimization technology used in Research In Motion’s BlackBerry devices has resulted in the handsets consuming an average of just 54 MB per month, according to Consumer Reports, compared to 273 MB for iPhone and 150 MB for other brands of smartphones.

Demand for wireless data is increasing exponentially, and device optimization – whether in the form of data compression or widgets built into Web browsers – has emerged as an easily accessible solution to network congestion. The technology is here to stay, and if data demands are any indication, its use is only going to grow.


Share this Story

You may login with either your assigned username or your e-mail address.
The password field is case sensitive.