Keynote: Short-lived Clouds

Serverless computing and its user base have skyrocketed in the past years due to the attractive promise of elasticity, low cost, and fast deployment without the need of provisioning a fixed virtual machine (VM) infrastructure, and consequently eliding all of its associated costs to operate, run, and maintain. While the specs and interfaces of the commercial offerings differ somewhat, they all provide not only finer granularity billing than regular VMs, but also unprecedented platform elasticity by supporting the launch of a myriad of functions concurrently while exhibiting much faster start-up times than that of any VM-based offering. In practice, however, serverless computing has been relegated to a niche of simple and sporadic workloads and remains largely impractical to a wide variety of applications which require more complex workflows of functions and communication patterns.In this talk, we make the case that serverless computing is the back bone of future cloud computing platforms, and, in order to circumvent and alleviate its current limitations, applications should be re-designed around it, i.e., become serverless native, along with providing a few opportunities and challenges for realizing the future ideal short-lived or ephemeral cloud that operates efficiently and on-demand without any need of provisioning, allowing for an ideal match of resources of supply and demand.
Dr. Javier Picorel
Engineering Manager at Huawei Cloud R&D in Huawei's Munich Research Center.
His group's mission at Huawei is to build the next-generation TCO-efficient cloud infrastructure. His research interests encompass the broad area of computer systems and computer architecture, with an emphasis on serverless computing, data and AI systems, and hardware-software co-design. He received a PhD in computer science from EPFL in 2017, and is also the recipient of several awards during his tenure at Huawei.