TuneUp.ai originated from Project ASCAR and Project CAPES of the Storage Systems Research Center (SSRC) at the University of California, Santa Cruz. We focus on using artificial intelligence and machine learning to improve the tuning process of complex computer systems, starting from what we are good at, the storage systems.
Yan Li is the main figure behind this project. After discovering a machine learning method that could increase the performance of some Lustre workloads by 35% automatically, without any need for human input, he decided to put more effort into commercializing this project. In 2016 and 2017, Yan Li, working with Professor Darrell Long, Ken, and Oceane from SSRC, UCSC, developed a groundbreaking method that used Deep Reinforcement Learning to tune parameters of high performance storage systems. This new system was called CAPES. CAPES was even better than ASCAR and had demonstrated that it could increase the I/O throughput of a Lustre cluster by as much as 45% at saturation point, without supervision.
Now we are thrilled to put these, and many more advanced AI tuning methods that we have been developing, into your hands, to make your system run faster without any manual tuning effort.