Akamai Reports Improved Performance with BBR Algorithm
As viewers consume more and more video each day, network congestion has become an increasing challenge for CDNs. Akamai announced today that it has responded to that challenge by completing rollout of the Bottleneck Bandwidth and Roundtrip Time (BBR) TCP congestion algorithm across its Edge Platform. BBR is just one of the algorithms Akamai uses, depending on a variety of factors, and is now the algorithm of choice more than 80% of the time, according to a company spokesperson.
Rather than wait until customers complained of congestion and excessive buffering, Akamai took a proactive approach in rolling out BBR, says Alex Balford, senior product marketing manager at Akamai. "Rolling out BBR is more about Akamai’s ongoing work to evaluate, test, and deploy capabilities that can help improve delivery performance for our customers rather than the result of complaints or specific requests," Balford says. "Particularly given the growth of and competition within the online media industry, delivery performance is a paramount concern for our customers. We're thrilled by the improvements that we’ve observed since the rollout of BBR across Akamai’s edge platform."
In a blog post released today, Balford writes that one customer who uses multiple CDN services saw Akamai move into the top spot for U.S. CDN performance after a 5Mbps increase in average throughput after BBR was applied, and another saw improvements of anywhere from 5% to 18% globally. Overall, for adaptive media delivery, Akamai reports that goodput improved by nearly 19% for files larger than 1 MB.
Google first introduced a version of the BBR algorithm in 2016 and started implementing it a year later, and Amazon CloudFront began using it earlier this year. Akamai's version of BBR is unique, however. "The global rollout took place in November; however, Akamai began experimenting with BBR well over a year ago, tested it for some time, and deployed it opportunistically for certain customers," says Balford. "Using the insight gained from those early experiences with BBR, Akamai took the Google version and optimized it specifically for the Akamai platform and deployed it globally so that all customers using our media products can benefit from the boost in performance."
Balford says that BBR isn't the only new approach to reducing congestion that Akamai is now using. "It's worth noting that around the same time as the BBR rollout, Akamai deployed another platform feature that uses machine learning to make intelligent, real-time decisions regarding which congestion control algorithm to use per end user connection," he says. "Akamai is harnessing the vast amount of data gathered around how we operate in certain geographies, ISPs, how client devices operate, and more. We’re continuing to fine tune this over time to achieve even higher performance levels."
Related Articles
There are four key delivery methods for live content, and understanding the math behind them can help decision makers determine which is the best for their applications.
01 Nov 2020
New developer APIs give providers more control over live and on-demand streams, and the company has improved its Cloud Wrapper and security offerings as well.
14 Oct 2020
Multiple algorithms for encoding, transfer, and playback intersect in the end user's player. So how can you make the numbers work in your favor?
27 Aug 2020
Providing flawless live streaming quality has never been more complex—or essential. However, relying on performance monitoring, real-time communications, and AI can ensure satisfied end-users.
01 Oct 2019
Data breaches aren't just a concern for financial platforms. Hackers also target OTT platforms. Here are three best practices for keeping subscriber data safe.
31 Jul 2019