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 Harness's Universal Recommender 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.

Stream processing frameworks allow processing massive amounts of data shortly after it is produced, and enable a fast reaction to events in scenarios such as data center monitoring, smart transportation, or telecommunication networks. Many scenarios depend on the fast and reliable processing of incoming data, requiring low end-to-end latencies from the ingest of a new event to the corresponding output. The occurrence of faults jeopardizes these guarantees: Currently-leading high-availability solutions for stream processing such as Spark Streaming or Apache Flink's implement passive replication through snapshotting, requiring a stop-the-world operation to recover from a failure. Active replication, while incurring higher deployment costs, can overcome these limitations and allow to mask the impact of faults and match stringent end-to-end latency requirements. We present the design, implementation, and evaluation of active replication in the popular Apache Flink platform. Our study explores two alternative designs, a leader-based approach leveraging external services (Kafka and ZooKeeper) and a leaderless implementation leveraging a novel deterministic merging algorithm. Our evaluation using a series of microbenchmarks and a SaaS cloud monitoring scenario on a 37-server cluster show that the actively-replicated Flink can fully mask the impact of faults on end-to-end latency.

Even though the cloud computing paradigm has proven benefits, it faces a serious problem that can compromise its commercial success. It concerns the lack of efficient approach for using optimally the available resources. For this, several approaches have been proposed. However, they suffer from several shortcomings. Often only one objective is taken into account, expressing all operations in terms of cost. Furthermore, business processes should be insured with elasticity and multi-tenancy mechanism while adjusting the available resources to the dynamic load distribution. The proposed approach aims to optimize two conflicting objectives, namely the number of migrations of tenants and the cost incurred using a set of resources. It allows to take into account the multi-tenancy property and the Cloud computing elasticity, and is efficient as shown by an extensive experimentation based on real data from Bonita BPM customers.

In this paper, we describe a framework that allows to automate and repeat business process execution on different cloud configurations. We present how and why the different components of the experimentation pipeline-like Ansible, Docker and Jenkins have been set up, and the kind of results we obtained on a large set of configurations from the AWS public cloud. It allows us to calculate actual prices regarding the cost of process execution, in order to compare not only pure performance but also the economic dimension of process execution.