MCAS(Memory Centric Active Storage)是一种用于高性能持久数据存储的持久内存层。它从一开始就设计为通过内存持久性和复制提供具有低延迟保证和数据持久性的键值功能。为了减少数据移动并进一步提高性能,我们提供了对用户定义的“下推”操作(称为活动数据对象)的支持,这些操作可以在与一个或多个键相关联的值内存上直接安全地执行。ADO机制允许复杂的基于指针的动态数据结构(例如,树)被存储在持久存储器中并对其进行操作。为此,我们研究了MCAS-ADO在处理企业存储系统元数据以实现连续数据保护(CDP)方面的实际用例。这需要不断更新复杂的元数据,必须保持一致和持久。在本文中,我们i.)提出了MCAS-ADO系统的体系结构;展示CDP用例是如何实现的,最后iii.)在此用例的上下文中对系统性能进行评估。
英文摘要:
MCAS (Memory Centric Active Storage) is a persistent memory tier for high-performance durable data storage. It is designed from the ground-up to provide a key-value capability with low-latency guarantees and data durability through memory persistence and replication. To reduce data movement and make further gains in performance, we provide support for user-defined "push-down" operations (known as Active Data Objects) that can execute directly and safely on the value-memory associated with one or more keys. The ADO mechanism allows complex pointer-based dynamic data structures (e.g., trees) to be stored and operated on in persistent memory. To this end, we examine a real-world use case for MCAS-ADO in the handling of enterprise storage system metadata for Continuous Data Protection (CDP). This requires continuously updating complex metadata that must be kept consistent and durable. In this paper, we i.) present the MCAS-ADO system architecture, ii.) show how the CDP use case is implemented, and finally iii.) give an evaluation of system performance in the context of this use case.
CIDER: Boosting Memory-Disaggregated Key-Value Stores with Pessimistic Synchronization
College of Computer Science and Artificial Intelligence, Fudan University(复旦大学计算机科学与人工智能学院);National Key Laboratory of Parallel and Distributed Computing(并行与分布式计算国家重点实验室);Duke Kunshan University(杜克昆山大学)
Lessons learned from the early performance evaluation of Intel Optane DC Persistent Memory in DBMS
非易失性存储器(NVM)是一种新兴技术,其具有大容量存储设备的持久性特性(例如,HDD和SSD),同时提供传统DRAM存储器的低访问延迟和字节可寻址性。这种独特的功能组合在构建数据库管理系统(DBMS)时开辟了几个新的设计考虑因素,例如取代DRAM(作为主工作空间内存)或块设备(作为持久性存储),或者同时补充两种DBMS组件(例如访问方法,存储引擎,缓冲区管理,日志记录/恢复等)。然而,与NVM交互需要改变应用软件以最佳地使用设备(例如,小高速缓存行的mmap和clflush,而不是大页面缓冲器的write和fsync)。在对NVM的DBMS引入(可能是主要的)代码更改之前,开发人员需要清楚地了解NVM在各种条件下的性能,以帮助做出更好的设计选择。在本文中,我们提供了广泛的性能评估进行了最近发布的NVM设备,英特尔Optane DC持久性存储器(PMem),在不同的配置与几个微基准测试工具。此外,我们评估OLTP和OLAP数据库工作负载(即,TPC-C和TPC-H)与Microsoft SQL Server 2019配合使用时,使用NVM设备作为内存缓冲池或持久性存储。从经验教训中,我们分享了一些建议,为未来的DBMS设计与PMem,例如,简单的硬件或软件的变化是不够的,最好的使用PMem在DBMS中。