The University of Texas at Arlington computer science team is using a $600,000 grant from the National Science Foundation to develop algorithms for microservices-based data center services that enable faster use, more efficient data center computing resources.
In addition, the grant will advance the curriculum while supporting a number of doctoral students working on the project and providing research experiences for undergraduate and graduate students.
Hong Jiang, chair, and Wendell H. Nedderman, Distinguished Professor in the UTA Department of Engineering and Computer Science, are leading a project titled “SHF: Small: A Resource Management Framework that Ensures SLO Can distributed extension for microservices”. Microservices are a type of computer software building that Jiang compares to building something with Legos.
Each individual microservice can be viewed as a Lego part and one can enable a specific computing service by simply putting together a set of Lego parts or microservices, instead of having to build a service. service from scratch,” said Jiang. “When you scale a computing service built with microservices, you can just move to the small congested services and scale them without scaling other parts of the service. However, since different computing services can share microservices, how to ensure performance for individual computing services becomes an important challenge.”
Another challenge Jiang is trying to solve is how to make efficient use of microservice resources.
“The ability to scale services at the microservice level gives you the best performance and response times, but you often don’t use the full power of the system,” says Jiang. “It’s like having a car with 600 hp but only using 300 hp from the engine. It’s expensive and unnecessary.”
Zhijun Wang, a computer science research associate, and Hao Che, a computer science professor, were part of the research team. Wang says the solution the team has proposed should address those challenges.
“Microservices are already there,” says Wang. “What we’re trying to do is ensure performance for individual computing services while maximizing resource usage.”
Such a maximization is a huge economic benefit for businesses, says Che.
“For every additional 100 milliseconds a customer is waiting on your website, the chance that a customer might leave increases by a few percent,” says Che. “Questions from customers browsing your website are imperative to being answered quickly and accurately. It means real money loss if they travel somewhere else. “