By decoupling compute and memory resources into independent pools that are provisioned and managed separately, disaggregated memory (DM) is promising to break the scaling constraints for memory systems and improve resource utilization. However, it also comes with a new challenge to design a high-performance hashing index to manage the vast memory pool with weak computing power. In this paper, we reconsider this problem and find that existing hashing indexes for DM still experience two fundamental yet unresolved limitations: (i) amplifying traffic under high insertion concurrency, and (ii) introducing significantly high insertion latency, stemming mainly from the directory synchronization and item relocation in resizing process. We resolve the above limitations by designing Pulse, a finegrained hierarchical hashing index for DM. Pulse comprises the following three design primitives. It proposes a multi-level index structure, which breaks the conventional flat directory into multiple sub-directories that are organized hierarchically, achieving fine-grained directory synchronization. Pulse also maintains a small portion of hashed keys in the memory pool, which aids in item relocation during resizing, thereby reducing resizing traffic and promising system stability. Pulse finally exploits operation parallelism by tailoring the doorbell batching mechanism with selective signaling. We conduct extensive experiments using a variety of benchmarks, showing that Pulse can improve$3.46 \times$of the throughput and reduce 76.3% of the tail latency compared to state-of-the-art hashing indexes.

Persistent memory (PMem) brings new design considerations in realizing high-performance and scalable hashing indexes. We uncover that existing hashing indexes for PMem still suffer from traffic amplification and memory inefficiency. We present MetoHash, a memory-efficient and traffic-optimized hashing index on hybrid PMem-DRAM memories. MetoHash proposes a three-layer index structure spanning across CPU caches, DRAM, and PMem for data management. It aggregates the incoming key-value items in CPU caches for fast inserts, which are then arranged in DRAM and flushed to PMem, to eliminate traffic amplification. MetoHash also uses fingerprinting to reduce unnecessary probings over PMem and removes duplicate items during bucket relocations. We implement MetoHash on PMem with persistent and volatile CPU caches, and show that compared to state-of-the-art hashing indexes for PMem, MetoHash improves the throughput by 86.1–257.6% under various workloads.

Unsupervised time-series anomaly detection (TSAD) aims to identify anomalies in industrial sensing signals to ensure production safety. As Industry 4.0 emerges, TSAD deployment must migrate from resource-rich cloud to resource-limited edges for real-time and fine-grained control. In this case, it raises new targets with high accuracy, high timeliness, and low consumption for TSAD. However, existing models focus solely on achieving high accuracy by building neural networks with deep structures and large parameters. Consequently, these models demand prohibitive training durations and computational overhead, which makes them unsuitable for edge deployment. To solve this issue, an unsupervised lightweight and fast TSAD model, namely, LFTSAD, is proposed via point-level and sequence-level reconstruction discrepancy in this article. First, to achieve high timeliness and low consumption, LFTSAD uses two two-layer multilayer perceptron networks (MLPs) to construct a lightweight contrastive architecture with few parameters. Second, leveraging the lightweight architecture, a dual-branch reconstruction network is designed to generate corresponding reconstruction discrepancies from point-level and sequence-level perspectives. Finally, a novel anomaly scoring scheme is designed to combine point-level and sequence-level reconstruction discrepancies for more accurate anomaly detection. To the best of our knowledge, this is the first work to develop a lightweight All-MLP-based TSAD model for resource-limited edge devices. Extensive experiments demonstrate that LFTSAD is 3-10 times faster in timeliness, consumes only 1/2 of the resources, and achieves accuracy that is either comparable to or superior to several deep SOTA models. The source code of LFTSAD is here https://github.com/infogroup502/LFTSAD.