Peng Xu

Building LSM-Tree-based key-value stores (LSM stores) on Cloud Block Storage (CBS) is becoming increasingly popular due to its convenience compared to local block storage devices. However, the constrained bandwidth of CBS and the competition between bandwidth and paid IOPS results in performance degradation and high tail latency in LSM stores, which cannot provide stable service-level agreements (SLAs). We propose RaKV, an efficient LSM store with stable SLAs for CBS, to improve write performance and reduce latency. This store successfully tackles challenges by utilizing instance memory and managing data flow between components. First, RaKV designs a dynamically partitioned L ₀ placed in memory and a parallel L ₀ - L ₁ compaction mechanism to minimize write amplification and write stalls, mitigating the performance impact of the CBS’s constrained bandwidth. Second, RaKV proposes a bandwidth-aware admission control mechanism that dynamically adjusts the number of incoming user requests based on the capacity of the L ₀ layer and the available cloud storage bandwidth, thus preventing write stalls and ensuring SLA consistency. Finally, RaKV implements an efficient read/write I/O isolation and scheduling mechanism to coordinate bandwidth and IOPS demands, thus stabilizing read SLAs and reducing read latency. Our evaluation results show that RaKV improves random write performance by 25.8% compared to the state-of-the-art key-value store ADOC, with more consistent throughput and lower read latency than CruiseDB.

Sparse general matrix-matrix multiplication is widely used in data mining applications. Its irregular memory access patterns limit the performance of general-purpose processors, thus motivating many FPGA-based hardware innovations in recent years. Nevertheless, existing accelerators fail to efficiently support heterogeneous input matrix sparsity, which is universal in various real-world applications. With in-depth experimental analysis, we observe that their performance is bottlenecked by their fixed tiling mechanisms, which only alleviate the irregularity of one input matrix. Based on the observation, we propose SpaHet, a software/hardware co-design to accelerate heterogeneous-sparsity based sparse matrix multiplication. SpaHet adopts a dual-adaptive sliding window mechanism to cover the reuse characteristics of both input matrices simultaneously. With a specialized exploration algorithm, the window-based mechanism can automatically find the optimal tiling strategy instead of applying a fixed one based on empirical experience. A sparsity-aware merge tree is also proposed to maximize the output matrix reuse via accumulating intermediate results thoroughly. Our results on a Xilinx Alveo U280 accelerator card show that SpaHet outperforms state-of-the-art CPU-, GPU- and FPGA-based solutions by 7.71×, 1.1×, and 2.74× in performance, respectively.

Cloud storage is gaining popularity because features such as pay-as-you-go significantly reduce storage costs. However, the community has not sufficiently explored its contract model and latency characteristics. As LSM-Tree-based key-value stores (LSM stores) become the building block for numerous cloud applications, how cloud storage would impact the performance of key-value accesses is vital. This study reveals the significant latency variances of Amazon Elastic Block Store (EBS) under various I/O pressures, which challenges LSM store read performance on cloud storage. To reduce the corresponding tail latency, we propose Calcspar, a contract-aware LSM store for cloud storage, which efficiently addresses the challenges by regulating the rate of I/O requests to cloud storage and absorbing surplus I/O requests with the data cache. We specifically developed a fluctuation-aware cache to lower the high latency brought on by workload fluctuations. Additionally, we build a congestion-aware IOPS allocator to reduce the impact of LSM store internal operations on read latency. We evaluated Calcspar on EBS with different real-world workloads and compared it to the cutting-edge LSM stores. The results show that Calcspar can significantly reduce tail latency while maintaining regular read and write performance, keeping the 99 ^th percentile latency under 550μs and reducing average latency by 66%. In addition, Calcspar has lower write prices and average latency compared to Cloud NoSQL services offered by cloud vendors.

With the current large scale deployment of machine learning technologies, such as those on cloud servers and edge and IoT hardwares, machine learning systems have been widely prevalence. Practical requirement has driven their performance increase in both academia and industry. However, the application requirement varies greatly across different applications, and directly using off-the-shelf systems might not be sufficient in many cases. In this work, we first propose to implement a series of techniques to optimize performance of convolution operation, one of the most important operations, in constructing deep learning networks. Besides, we also propose to apply the automated empirical optimisation of software approach to improve the performance of operators in machine learning system, most notably across various hardware platforms. Evaluation compared to existing libraries on different hardware devices has proved the efficiency of our proposed method.

The explosive growth of modern web-scale applications has made cost-effectiveness a primary design goal for their underlying databases. As a backbone of modern databases, LSM-tree based key-value stores (LSM store) face limited storage options. They are either designed for local storage that is relatively small, expensive, and fast or for cloud storage that offers larger capacities at reduced costs but slower. Designing an LSM store by integrating local storage with cloud storage services is a promising way to balance the cost and performance. However, such design faces challenges such as data reorganization and metadata overhead issues. In this paper, we propose ROCKSMASH, a fast and efficient LSM store that uses local storage to store frequently accessed data and metadata while using cloud to hold the rest of the data to achieve cost-effectiveness. To improve metadata space-efficiency and read performance, ROCKSMASH uses an LSM-aware persistent cache that stores metadata in a space-efficient way and stores popular data blocks by using compaction-aware layouts. We implemented ROCKSMASH by embedding these designs into RocksDB. The evaluation results show that ROCKSMASH improves the performance by up to 1.7 × compared to the state-of-the-art schemes and delivers higher reliability and cost-effectiveness.