Remzi H. Arpaci-Dusseau

This paper is included in the Proceedings of the 15th USENIX Conference on File and Storage Technologies (FAST ’17).

This paper is included in the Proceedings of the 21st USENIX Conference on File and Storage Technologies.

Towards Efficient, Portable Application-Level Consistency

Warming up Storage-Level Caches with Bonfire

Bonfire is a mechanism for accelerating cache warmup based on detailed analysis of block-level data-center traces that provides insights into heuristics for warmup as well as the potential for efficient mechanisms.

NyxCache: Flexible and Efficient Multi-tenant Persistent Memory Caching

This paper is included in the Proceedings of the 24th USENIX Conference on File and Storage Technologies.

HARE introduces a holistic allocation model that captures the demand correlation between cache size and other resources, and uses a novel two-phase harvest/redistribute method to optimize resource allocation across tenants, maximizing the throughput of each while maintaining fairness.

InfraBench: Evaluating Infrastructure Agents Across Layers, Lifecycle, and Risk

InfraBench is presented, a benchmark suite for evaluating AI agents on realistic infrastructure tasks across the full system stack and full operational lifecycle with fine-grained risk assessment and shows that even the strongest agent cannot secure a full score across all tasks.