Querying data stored in cloud object stores often leads to network bottlenecks, particularly when large datasets need to be transferred over wide area networks (WANs) for processing. Encryption further complicates this challenge by requiring entire encrypted objects to be fetched from the object store before analysis. To address this, we push down filtering and perform secure computing near storage using a Data Processing Unit (DPU) integrated into the cloud server.

Queries of multi-TB Mass Spectrometry (MS) repositories provide deep insights into biological processes and pose challenging data processing problems. The key bottleneck for running these queries is the number of small random reads. Byte-addressable persistent main memory (PMEM) technologies enable real-time MS search systems by delivering low-latency, high-bandwidth storage. This work presents P-Massive, real-time multi-terabyte scale MS search system. P-Massive takes advantage of PMEM and the underlying nature of its data access patterns to maximize performance. We evaluate P-Massive across various storage hierarchies and project forward over the next decade to understand how MS query systems might evolve. Our evaluation shows that P-Massive offers a cost-effective solution that achieves near-DRAM performance. A single query takes 1.7 seconds in P-Massive, 69× faster than state-of-the-art implementation. In an end-to-end, user-facing application, P-Massive delivers a 90% shorter wait time than the latest MS search tool, returning results within seconds rather than minutes.

In this paper, we consider an aerial ad-hoc network construction problem using UAVs in a disaster scenario. We aim to reconnect the communication-wise isolated urban area with the outside communication infrastructure. Our main goal is to perform both network exploration and relay deployment tasks at the same time by taking a progressive optimization toward a self-organizing network construction. We propose a novel UAV exploration-and-deployment algorithm that gradually explores the region of interests and achieves full network coverage in a fast manner. Then, we present an effective network refinement algorithm based on clustering that minimizes the number of UAVs for deployment by finding out essential UAVs, while keeping the similar network coverage performance. Simulation results demonstrate that our proposed scheme significantly reduces the execution time for network exploration and deployment compared to a baseline counterpart. Also, our cluster-based network refinement algorithm provides a very lightweight yet effective solution, well-balancing between UAV resource and computation overhead.

We consider a 3D network construction problem in the post-disaster scenario, where large urban areas are communication-wise isolated from the outside environment due to the severely damaged network infrastructure. Our main goal is to reconnect the isolated regions with the outside environment using unmanned aerial vehicles (UAVs) by building 3D aerial ad-hoc networks. Prior to network construction, we aim to capture the global map information over region of interests (RoI) by exploring all obstacles in the unknown region. We propose an efficient technique for collaborative 3D terrestrial exploration using multiple UAVs based on our distributed path planning algorithm, which finds collision-free exploration paths. Then, we present an optimal full-coverage 3D aerial ad-hoc network construction by deploying the minimum number of UAVs to indispensable spots while obtaining maximum network coverage. Simulation results demonstrate that our proposed exploration scheme outperforms several counterpart algorithms in terms of traversal time and redundant visit rate. Also, our network construction algorithm guarantees almost full coverage toward terrestrial space with only minimal UAV usage.

We consider the problem of path planning using multiple UAVs as message ferries to deliver delay-sensitive information in a catastrophic disaster scenario. Our main goal is to find the optimal paths of UAVs to maximize the number of nodes that can successfully be serviced within each designated packet deadline. At the same time, we want to reduce total travel time for visiting over a virtual grid topology. We propose a distributed path planning algorithm that determines the next visit grid point based on a weighted sum of travel time and delivery deadline. Together with path planning, we incorporate a task division mechanism that collaboratively distributes the unvisited grid points with other UAVs so that the entire travel time can substantially be reduced. Simulation results demonstrate that our distributed path planning algorithm mixed with task division outperforms all baseline counterpart algorithms in terms of on-time service node rate and total travel time.