POSTER: ConcurORAM: High-Throughput Parallel Multi-Client ORAM

Oblivious RAM (ORAM) mechanisms have improved rapidly in recent years as increasing amounts of data are outsourced. Although several tree-based ORAMs such as PathORAM [8] and RingORAM [6] have achieved near-optimal bandwidth for single client scenarios, their low overall throughput due to high latency of access -- as clients need to wait for or know about and coordinate with each other, lest privacy is lost -- reduces their applicability for multi-client scenarios. In this paper, we propose ConcurORAM, a multi-client concurrent ORAM that eliminates waiting for concurrent clients and significantly increases overall throughput. ConcurORAM works by securely allowing multiple clients to asynchronously access the data set in between eviction rounds by judiciously storing ORAM position map data in a smaller parallel de-amortized pyramid ORAM [10] of higher complexity. In effect ConcurORAM reaps the benefits of parallelism at a lower O(log(N)) overall complexity by identifying and securely accessing the absolute critical data structures that require parallel access with privacy (position map) and designing everything else using append-only data structures that can be then merged securely in a separate eviction step.