An official website of the United States government
Recent advancements in deep learning techniques facilitate intelligent-query support in diverse applications, such as content-based image retrieval and audio texturing. Unlike conventional key-based queries, these intelligent queries lack efficient indexing and require complex compute operations for feature matching. To achieve high-performance intelligent querying against massive datasets, modern computing systems employ GPUs in-conjunction with solid-state drives (SSDs) for fast data access and parallel data processing. However, our characterization with various intelligent-query workloads developed with deep neural networks (DNNs), shows that the storage I/O bandwidth is still the major bottleneck that contributes 56%--90% of the query execution time. To this end, we present DeepStore, an in-storage accelerator architecture for intelligent queries. It consists of (1) energy-efficient in-storage accelerators designed specifically for supporting DNN-based intelligent queries, under the resource constraints in modern SSD controllers; (2) a similarity-based in-storage query cache to exploit the temporal locality of user queries for further performance improvement; and (3) a lightweight in-storage runtime system working as the query engine, which provides a simple software abstraction to support different types of intelligent queries. DeepStore exploits SSD parallelisms with design space exploration for achieving the maximal energy efficiency for in-storage accelerators. We validate DeepStore design with an SSD simulator, and evaluate it with a variety of vision, text, and audio based intelligent queries. Compared with the state-of-the-art GPU+SSD approach, DeepStore improves the query performance by up to 17.7×, and energy-efficiency by up to 78.6×.
more »
« less
This content will become publicly available on July 10, 2026
Unlocking the Unusable: A Proactive Caching Framework for Reusing Partial Overlapped Data
Cache systems are widely used to speed up data retrieving. Modern HPC, data analytics, and AI/ML workloads generate vast, multi-dimensional datasets, and those data are accessed via complex queries. However, the probability of requesting the exact same data across different queries is low, leading to limited performance improvement when a traditional key-value cache is applied. In this paper, we present Mosaic-Cache, a proactive and general caching framework that enables applications with efficient partial overlapped data reuse through novel overlap-aware cache interfaces for fast content-level reuse. The core components include a metadata manager leveraging customizable indexing for fast overlap lookups, an adaptive fetch planner for dynamic cache-to-storage decisions, and an async merger to reduce cache fragmentation and redundancy. Evaluations on real-world HPC datasets show that Mosaic-Cache improves overall performance by up to 4.1× over traditional key-value-based cache while adding minimal overhead in worst-case scenarios.
more »
« less
- PAR ID:
- 10612121
- Publisher / Repository:
- ACM
- Date Published:
- Page Range / eLocation ID:
- 129 to 136
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Data movement is a common performance bottleneck, and its chief remedy is caching. Traditional cache management is transparent to the workload: data that should be kept in cache are determined by the recency information only, while the program information, i.e., future data reuses, is not communicated to the cache. This has changed in a new cache design named Lease Cache . The program control is passed to the lease cache by a compiler technique called Compiler Assigned Reference Lease (CARL). This technique collects the reuse interval distribution for each reference and uses it to compute and assign the lease value to each reference. In this article, we prove that CARL is optimal under certain statistical assumptions. Based on this optimality, we prove miss curve convexity, which is useful for optimizing shared cache, and sub-partitioning monotonicity, which simplifies lease compilation. We evaluate the potential using scientific kernels from PolyBench and show that compiler insertions of up to 34 leases in program code achieve similar or better cache utilization (in variable size cache) than the optimal fixed-size caching policy, which has been unattainable with automatic caching but now within the potential of cache programming for all tested programs and most cache sizes.more » « less
-
Typical cybersecurity solutions emphasize on achieving defense functionalities. However, execution efficiency and scalability are equally important, especially for real-world deployment. Straightforward mappings of cybersecurity applications onto HPC platforms may significantly underutilize the HPC devices’ capacities. On the other hand, the sophisticated implementations are quite difficult: they require both in-depth understandings of cybersecurity domain-specific characteristics and HPC architecture and system model. In our work, we investigate three sub-areas in cybersecurity, including mobile software security, network security, and system security. They have the following performance issues, respectively: 1) The flow- and context-sensitive static analysis for the large and complex Android APKs are incredibly time-consuming. Existing CPU-only frameworks/tools have to set a timeout threshold to cease the program analysis to trade the precision for performance. 2) Network intrusion detection systems (NIDS) use automata processing as its searching core and requires line-speed processing. However, achieving high-speed automata processing is exceptionally difficult in both algorithm and implementation aspects. 3) It is unclear how the cache configurations impact time-driven cache side-channel attacks’ performance. This question remains open because it is difficult to conduct comparative measurement to study the impacts. In this dissertation, we demonstrate how application-specific characteristics can be leveraged to optimize implementations on various types of HPC for faster and more scalable cybersecurity executions. For example, we present a new GPU-assisted framework and a collection of optimization strategies for fast Android static data-flow analysis that achieve up to 128X speedups against the plain GPU implementation. For network intrusion detection systems (IDS), we design and implement an algorithm capable of eliminating the state explosion in out-of-order packet situations, which reduces up to 400X of the memory overhead. We also present tools for improving the usability of Micron’s Automata Processor. To study the cache configurations’ impact on time-driven cache side-channel attacks’ performance, we design an approach to conducting comparative measurement. We propose a quantifiable success rate metric to measure the performance of time-driven cache attacks and utilize the GEM5 platform to emulate the configurable cache.more » « less
-
Modern NVMe solid state drives offer significantly higher bandwidth and low latency than prior storage devices. Current key-value stores struggle to fully utilize the bandwidth of such devices. This paper presents SplinterDB, a new key-value store explicitly designed for NVMe solid state drives. SplinterDB is designed around a novel data structure (the STBε-tree), that exposes I/O and CPU concurrency and reduces write amplification without sacrificing query performance. STBε-tree combines ideas from log-structured merge trees and Bε-trees to reduce write amplification and CPU costs of compaction. The SplinterDB memtable and cache are designed to be highly concurrent and to reduce cache misses. We evaluate SplinterDB on a number of micro- and macro-benchmarks, and show that SplinterDB outperforms RocksDB, a state-of-the-art key-value store, by a factor of 6–10x on insertions and 2–2.6x on point queries, while matching RocksDB on small range queries. Furthermore, SplinterDB reduces write amplification by 2x compared to RocksDB.more » « less
-
Modern NVMe solid state drives offer significantly higher bandwidth and low latency than prior storage devices. Current key-value stores struggle to fully utilize the bandwidth of such devices. This paper presents SplinterDB, a new key-value store explicitly designed for NVMe solid state drives. SplinterDB is designed around a novel data structure (the STBε-tree), that exposes I/O and CPU concurrency and reduces write amplification without sacrificing query performance. STBε-tree combines ideas from log-structured merge trees and Bε-trees to reduce write amplification and CPU costs of compaction. The SplinterDB memtable and cache are designed to be highly concurrent and to reduce cache misses. We evaluate SplinterDB on a number of micro- and macro-benchmarks, and show that SplinterDB outperforms RocksDB, a state-of-the-art key-value store, by a factor of 6–10x on insertions and 2–2.6x on point queries, while matching RocksDB on small range queries. Furthermore, SplinterDB reduces write amplification by 2x compared to RocksDB.more » « less
