SummaryDelivering video content with a high and fairly shared quality of experience is a challenging task in view of the drastic video traffic increase forecasts, as live video traffic will grow 15‐fold by 2022. Currently, content delivery networks provide numerous servers hosting replicas of the video content, and consuming clients are redirected to the closest server. Then, the video content is streamed using adaptive streaming solutions. However, servers and network links often become overloaded during major events, and users may experience a poor or unfairly distributed quality of experience, unless more servers are provisioned. In this paper, we propose Muslin, a streaming solution supporting a high, fairly shared end users' quality of experience for live streaming, while minimizing the required content delivery platform scale. Muslin leverages on MS‐Stream, a content delivery solution, which aggregates video content from multiple servers to offer a high quality of experience for its users. Muslin dynamically provisions servers and replicates content into servers and advertises servers to clients based on real‐time delivery conditions. We have used Muslin to replay a 1‐day video‐games event, with hundreds of clients and several test beds. Our results show that our approach outperforms traditional content delivery schemes by increasing the fairness and quality of experience at the user side with a smaller infrastructure scale.

We present PProx, a system preventing recommendation-as-a-service (RaaS) providers from accessing sensitive data about the users of applications leveraging their services. PProx does not impact recommendations accuracy, is compatible with arbitrary recommendation algorithms, and has minimal deployment requirements. Its design combines two proxying layers directly running inside SGX enclaves at the RaaS provider side. These layers transparently pseudonymize users and items and hide links between the two, and PProx privacy guarantees are robust even to the corruption of one of these enclaves. We integrated PProx with the Harness recommendation engine and evaluated it on a 27-node cluster. Our results indicate its ability to withstand a high number of requests with low end-to-end latency, horizontally scaling up to match increasing workloads of recommendations.CCS Concepts • Security and privacy;