The goal of the LSDS group is to support the design and implementation of tomorrow's large-scale data-intensive systems. We investigate new abstractions and infrastructures for building scalable, robust and secure data-intensive applications.
Crossbow: Scaling Deep Learning with Small Batch Sizes on Multi-GPU Servers
A. Koliousis; P. Watcharapichat; M. Weidlich; L. Mai; P. Costa; and P. Pietzuch
In arXiv:1901.02244
sgx-perf: A Performance Analysis Tool for Intel SGX Enclaves
N. Weichbrodt; P.L. Aublin; and R. Kapitza
In ACM/IFIP Middleware '18
Hammer Slide: Work- and CPU-efficient Streaming Window Aggregation
G. Theodorakis; A. Koliousis; P. Pietzuch; and H. Pirk
In ADMS'18
Chi: A Scalable and Programmable Control Plane for Distributed Stream Processing Systems
L. Mai; Kai Zeng; Rahul Potharaju; Le Xu; Shivaram Venkataraman; P. Costa; Terry Kim; Saravanan Muthukrishnan; Vamsi Kuppa; Sudheer Dhulipalla; and S. Rao
Join the EU-funded SecureCloud project and investigate how new hardware trust mechanisms can be used to improve the security of cloud computing applications.
Two PhD positions are available in the broad area of Computer Systems with focus on designing and building systems that support BigData and Data Science workloads on modern and future hardware platforms.
