Implications of non-volatile memory as primary storage for database management systems

  title={Implications of non-volatile memory as primary storage for database management systems},
  author={Naveed Ul Mustafa and Adri{\`a} Armejach and {\"O}zcan {\"O}zturk and Adri{\'a}n Cristal and Osman S. Unsal},
  journal={2016 International Conference on Embedded Computer Systems: Architectures, Modeling and Simulation (SAMOS)},
  • N. Mustafa, Adrià Armejach, O. Unsal
  • Published 17 July 2016
  • Computer Science
  • 2016 International Conference on Embedded Computer Systems: Architectures, Modeling and Simulation (SAMOS)
Traditional Database Management System (DBMS) software relies on hard disks for storing relational data. Hard disks are cheap, persistent, and offer huge storage capacities. However, data retrieval latency for hard disks is extremely high. To hide this latency, DRAM is used as an intermediate storage. DRAM is significantly faster than disk, but deployed in smaller capacities due to cost and power constraints, and without the necessary persistency feature that disks have. Non-Volatile Memory… 

Figures and Tables from this paper

Impact of Storage Space Configuration on Transaction Processing Performance for Relational Database in PostgreSQL
What impact on database transaction efficiency has usage of modern hard drive versus solid state drive is shown and how selection of filesystem type for DBMS local storage influences transaction efficiency in supported databases is demonstrated.
NV-eCryptfs: Accelerating Enterprise-Level Cryptographic File System with Non-Volatile Memory
NV-eCryptfs is presented, an asynchronous software stack for eCryptfs, which utilizes NVM as a fast storage tier on top of slower block devices to fully parallelize encryption and data I/O, and design an efficient NVM management scheme to support the fast parallel cryptographic operations.
Towards a Native Architecture of In-NVM DBMS
  • S. Kuznetsov
  • Computer Science
    2019 Actual Problems of Systems and Software Engineering (APSSE)
  • 2019
There is the need for additional research into many aspects of the native architecture of an in-NVM DBMS, the choice of which is influenced only by the goals of simplicity and efficiency.
Reducing processor-memory performance gap and improving network-on-chip throughput
This thesis proposes modifications in storage engine (SE) of a DBMS aiming at fast access to data through bypassing the slow disk interfaces while maintaining all the functionalities of a robust DBMS and proposes a selection scheme that switches routing algorithm of an NoC with changing traffic pattern of an application.
Efficient Machine Learning Methods for Hard Disk Drive Yield Prediction Improvement
This work introduces yield prediction improvement by statistical analysis and machine learning methods including the multiple linear regression (MLR), artificial neural networks (ANN), classification and regression tree (CART), and introduces technique to group quantity of data for yield prediction by considering consistency number, instead of grouping by calendar period as used in traditional method.
Feature Selection to Improve Performance of Yield Prediction in Hard Disk Drive Manufacturing
Yield prediction with MLR shows higher accuracy than yield estimation traditionally performed by human engineers, and the proposed novel learning steps can help HDD process engineers to predict yield with the better performance.


Let's Talk About Storage & Recovery Methods for Non-Volatile Memory Database Systems
This work implemented three engines in a modular DBMS testbed that are based on different storage management architectures, and presents NVM-aware variants of these architectures that leverage the persistence and byte-addressability properties of NVM in their storage and recovery methods.
High performance database logging using storage class memory
The detailed design of an SCM-based approach for DBMSs logging is presented, which achieves high performance by simplified system design and better concurrency support and solutions to tackle several major issues arising during system recovery, including hole detection, partial write detection, and any-point failure recovery are discussed.
IBM solidDB: In-Memory Database Optimized for Extreme Speed and Availability
The structural differences between in-memory and disk-based databases, and how solidDB works to deliver extreme speed are explored.
Storage Management in the NVRAM Era
This paper demonstrates that using NVRAM as a drop-in replacement for disk allows near-instantaneous recovery, but software complexity necessary for disk limits transaction throughput, and considers the possibility of removing software-managed DRAM buffering.
DataBlitz: A High Performance Main-Memory Storage Manager
The Dali system is a main memory storage manager designed to provide the persistence, availability and safety guarantees one typically expects from a diskresident database, while at the same time providing very high performance by virtue of being tuned to support in-memory data.
NVRAM-aware Logging in Transaction Systems
This paper shows that using NVRAM only for the logging subsystem (NV-Logging) provides much higher transactions per dollar than simply replacing all disk storage with NVR AM, and shows that the software overheads associated with centralized log buffers cause performance bottlenecks and limit scaling.
System software for persistent memory
PMFS, a light-weight POSIX file system that exploits PM's byte-addressability to avoid overheads of block-oriented storage and enable direct PM access by applications (with memory-mapped I/O), is implemented.
Reducing energy consumption of queries in memory-resident database systems
This paper exploits the structured organization of memories, which enables a selective mode of operation in which banks are accessed selectively, and proposes hardware techniques that dynamically control the memory by making the system adapt to the access patterns that arise from queries.
A recovery algorithm for a high-performance memory-resident database system
It is argued that existing database recovery schemes fall short of meeting the requirements of such a system, and a new recovery mechanism is presented which is designed to overcome their shortcomings.
Non-Volatile Memory : Emerging Technologies And Their Impacts on Memory Systems
The introduction of these new memory technologies will probably result in more complex memory hierarchies, but is likely to allow the construction of memory chips that are non-volatile, low-energy and have density and latency close to or better than current DRAM chips, improving performance/eiency and allowing memory systems to continue to scale up.