Log-Structured Merge-Tree (LSM)-based key-value stores (LSM-KV stores) form the backbone of modern NoSQL systems due to their superior write performance. However, they are vulnerable to severe performance degradation including write stalls that can reduce throughput to nearly zero. Prior studies primarily investigate these issues under write-intensive workloads, attributing performance degradation to resource saturation or imbalances between foreground and background operations. Yet, their conclusions remain inconsistent, largely due to variations in hardware and system configurations. Moreover, real-world applications typically exhibit diverse read-write access patterns and workload intensity which are rarely examined in existing research. In this work, we conduct a comprehensive investigation of performance degradation across a broad spectrum of read-write mixed workloads, from write-dominant to read-intensive scenarios. We analyze how workload intensity, hardware resources, and key LSM-KV configuration parameters—such as background threads, block cache size, sub-compaction, and component-constraint mechanisms—jointly shape performance behaviors. Our study uncovers the root causes of degradation in both write-heavy and read-heavy regimes and highlights several underexamined factors whose interactions substantially influence system efficiency. These findings extend prior observations by revealing previously overlooked performance interactions and provide actionable guidance for configuring LSM-KV stores to mitigate degradation across diverse workloads.

As the volume of retained data continues to increase, it is important to design primary storage systems that efficiently respond to access requests while also providing strong sustainability by reducing carbon emissions. Just over two decades ago, the Massive Arrays of Idle Disks (MAID) architecture was introduced as an energy-efficient alternative to traditional HDD-based always-on storage, employing aggressive spin-down strategies to reduce power consumption. However, high access latencies and hardware limitations led to its decline. In this work, we propose a tiered SSD+MAID storage model that combines the low-latency advantages of SSDs with the energy and carbon efficiency of a MAID system, thus offering a modern alternative to MAID while achieving lower carbon emissions than all-SSD storage. To assess the sustainability impact of such a tiered storage system, we develop a comprehensive carbon emission model that incorporates access patterns, update behaviors, and HDD spin-up dynamics. This model captures both operational and embodied carbon costs, enabling evaluations of primary storage with sustainability in mind. Through real-world workloads, we evaluate the proposed SSD+MAID system and show that it can provide a good trade-off between performance, price, and sustainability.

Modern cloud-scale cold storage data centers have begun to support right-provisioning of a rack’s resources (power, cooling, etc.), which allows only a small fraction of all hard disks to be active (spinning) concurrently at any given time to reduce the cost of ownership. Data deduplication is a traditional approach to split files into chunks and eliminate duplicate chunks, which can also cut costs for cold storage systems. However, when combined with right-provisioning, classical deduplication may make a file deduplicated and stored across the disks some of which are not active currently, thus leading to unacceptable access performance caused by spinning up and down of the disks.

In-memory key-value stores are often used to speed up Big Data workloads on modern HPC clusters. To maintain their high availability, erasure coding has been recently adopted as a low-cost redundancy scheme instead of replication. Existing erasure-coded update schemes, however, have either low performance or high memory overhead. In this paper, we propose a novel parity logging-based architecture, HybridPL, which creates a hybrid of in-place update (for data and XOR parity chunks) and log-based update (for the remaining parity chunks), so as to balance the update performance and memory cost, while maintaining efficient single-failure repairs. We realize HybridPL as an in-memory key-value store called LogECMem, and further design efficient repair schemes for multiple failures. We prototype LogECMem and conduct experiments on different workloads. We show that LogECMem achieves better update performance over existing erasure-coded update schemes with low memory overhead, while maintaining high basic I/O and repair performance.

People's dependence on search engines allows various computer viruses to spread faster and stronger. Most scholars have neglected the influence of search engines on virus propagation and immunity. It is impossible to immunize all users at the same time with a huge system like social networks. So the main problem is how to pick a fixed-scale node cluster as the source of immunity in the network, which can make other individuals immune and continue to spread (called immune seeds). The immune seeds are scattered on some web pages of search engines to reduce the network virus infection rate. We establish two models, one is the model of computer virus early propagation based on the search engine, and the other is the model of the virus propagation and immunization model. Then we propose an improved immunization strategy: Community Partition immunization strategy based on the target immunization strategy. And we use four real datasets and two simulated datasets to do the simulation experiments, which shows that search engine can promote the propagation of the virus and the immune seeds, and the efficiency of the Community Partition immunization strategy is slightly higher than the target immunization strategy based on degree under the same conditions.