Bringing the data centre SLA into the 21st century
Thu 13 Oct 2016
A researcher from the University of Texas has won a grant worth nearly $800,000 from the National Science Foundation to conduct research into improving SLA workflows for data centres.
It’s an ironic problem to be tackling this late into the cloud evolution. While the advent of on-demand services has made micro-granular provisioning a fact of life – perhaps most famously via Amazon Web Services (AWS) – the Service Level Agreement has been stuck in a state of massive over-provisioning by default for the last twenty years, due to the need to comprehensively anticipate customer needs. This leads to resource wastage, with incumbent costs passed on to the client.
Now Hao Che, an associate professor at the Computer Science and Engineering Department at the University of Texas, in Arlington, will address the issue over three years of a $799,950 grant (which appears to be nudging a threshold of its own) to research potential models that could produce workable and unassailable SLAs without such expensive legal buffer zones.
“Today’s cloud service providers,” said Che “mainly provide resource-centric services to their customers, offering hardware and software resources only, leaving the hard question to the customer themselves to deal with. For example, how many resources are needed to meet the service-level objectives. With our model, the customers will be able to know exactly what and how much resources need to be purchased to meet their needs.”
The research project also has the ambit of improving clarity of choice in the consumer sector.
SLAs relating to data centres have to cover a number of contingencies which tend to zoom up and down the DC stack in terms of provisioning, including penalties, system availability, customer support, security guarantees, response times, application maintenance and development, and penalties for early termination and other revisions in the contracted arrangements.
Che was formerly a systems architect in the private sector and an adjunct professor at Pennsylvania State University, with an emphasis on network resource management and network architecture.