Speaker: Daniela Dorneanu

334630_368231316529350_800943754_o

I am a passionate Computer Engineer and in the last 2 years my professional path included internships at CERN (Technical Student) and IBM Zurich (Great Minds Program) and a Master in Computer Engineering at Politecnico di Milano.

I consider that the great development of Internet allows us infinite ways to ease our life, but we don’t yet have designed the proper systems to fully take advantage of this. Virtualization was one of the main technologies which allowed the fast growth of Cloud computing. The entrance cost for the Cloud has lowered a lot and many companies use this today, to outsource their hardware maintenance. However this addresses confidentiality, integrity and availability and scalability from a new perspective as they become part of the offer. How does software need to keep up with the new security, reliability and privacy challenges? This direction is
very fascinating for me and and I would really like to develop new systems which would for example allow a automatic usage of the cloud resources or provide secure deletion of the resources we want to remove from a system.

I am very passionate about Linux systems and I am very enthusiastic of the project I was developing at IBM with the Linux Kernel. My work was highly appreciated and taken into account in the two papers accepted at:
- NDSS 2013, San Diego, CA Attack Surface Metrics and Automated Compile Time OS Kernel Tailoring
- HotDep 2012,Hollywood, CA Automatic OS Kernel TCB Reduction by Leveraging Compile-Time Configurability

Currently I am working as a technical student at CERN were I have the opportunity to explore working with an open-source cloud infrastructure based on Openstack. My goal is to update the license monitoring system and build a tool to report the current usage of the
commercial licenses at CERN. This tool aggregates data from approx 500 to 1000 working environments and provides licenses to approximate 4000 users. This is very exciting and as the Openstack cloud is still in early adoption here at CERN we aim to provide through this integration a generic process that other services will follow.