Work-in-Progress: CAMiSE: Content Addressable Memory-integrated Searchable Encryption

Abstract

Searchable symmetric encryption (SSE) aims to support efficient query-execution directly over encrypted databases. Practical implementations of SSE suffer from performance bottlenecks due to randomised memory accesses and computation-intensive cryptographic operations. We propose CAMISE – a fully associative memory-integrated framework for designing SSE systems with fast query processing over large databases. We show a novel usage of custom-designed Content Addressable Memory (CAM) to minimise storage-access latencies during query execution in SSE systems. We prototype a well-known SSE scheme, namely Oblivious Cross Tags (OXT), within this framework. Our implementation achieves 5x-7x speed-up over traditional software-based implementations while scaling smoothly to real-world databases with millions of records.