“Companies don’t have to purchase as much infrastructure to sell services to customers. “
Researchers have developed a new traffic engineering model that uses stock market financial risk theories to optimise data traffic and help reduce infrastructure costs.
Engineers at the Massachusetts Institute of Technology (MIT) have created the TeaVar model which takes into account the failure probabilities of the links that connect all of the world’s datacentres and internet traffic.
Currently, cloud service providers use underground fibre optic cables that are the backbone of connecting datacentres and cities across the world. In order to handle all of this traffic most firms rely on traffic engineering (TE) software that proportions out data bandwidth and network traffic priorities.
One of the biggest challenges for this type of engineering is dealing with the unexpected failure of data links, this can be caused by a host of reasons such as signal quality dropping or a construction worker simply slicing through a line as they work.
To mitigate data loss due to link failure organisations leave links free and unused so they can take up the slack in the event of infrastructure failure.
MIT researcher Jeremy Bogle states that current TE software is not doing enough to optimise traffic. Even though the software does find different link paths, it fails to take into account the reliability of the links. Bogle states that: “They don’t say, ‘This link has a higher probability of being up and running, so that means you should be sending more traffic here.’ Most links in a network are operating at low utilization and aren’t sending as much traffic as they could be sending.”
TeaVar Traffic Engineering Model
The model was developed in collaboration with Microsoft and is essentially a mathematical model that can improve the performance of cloud-computing networks through risk awareness calculations.
The model is based off of stock market calculations which take conditional value into account, for example MIT states that: “A one-day 99 percent conditional value at risk of $50, your expected loss of the worst-case 1 percent scenario on that day is $50. But 99 percent of the time, you’ll do much better.”
In their paper the researchers note “Given the high cost of wide-area backbone networks (WANs), large service providers (e.g., Amazon, Facebook, Google, Microsoft) are investing heavily in optimizing their WAN TE, leveraging Software-Defined Networking (SDN) to globally optimize routing and bandwidth allocation to users.”
MIT researchers conducted experiments, based on real-world data, which showed the TeaVar model can support three times the traffic in comparison to more traditional TE methods. Understanding the risk value of the links helps to establish guaranteed data lines that can be relied on and optimised to a higher degree.
Manya Ghobadi, the TIBCO Career Development Assistant Professor in the MIT Department of Electrical Engineering and Computer Science stated that: “Having greater utilized infrastructure isn’t just good for cloud services — it’s also better for the world.”
“Companies don’t have to purchase as much infrastructure to sell services to customers. Plus, being able to efficiently utilize datacenter resources can save enormous amounts of energy consumption by the cloud infrastructure. So, there are benefits both for the users and the environment at the same time.”
“Mostly, link failures are due to the age of equipment, so the probabilities of failure don’t change much over time. That means our probabilities are more reliable, compared to the stock market.”