SANgo: a storage infrastructure simulator with reinforcement learning support

We introduce SANgo (Storage Area Network in the Go language)—a Go-based package for simulating the behavior of modern storage infrastructure. The software is based on the discrete-event modeling paradigm and captures the structure and dynamics of high-level storage system building blocks. The flexible structure of the package allows us to create a model of a real storage system with a configurable number of components. The granularity of the simulated system can be defined depending on the replicated patterns of actual system behavior. Accurate replication enables us to reach the primary goal of our simulator—to explore the stability boundaries of real storage systems. To meet this goal, SANgo offers a variety of interfaces for easy monitoring and tuning of the simulated model. These interfaces allow us to track the number of metrics of such components as storage controllers, network connections, and hard-drives. Other interfaces allow altering the parameter values of the simulated system effectively in real-time, thus providing the possibility for training a realistic digital twin using, for example, the reinforcement learning (RL) approach. One can train an RL model to reduce discrepancies between simulated and real SAN data. The external control algorithm can adjust the simulator parameters to make the difference as small as possible. SANgo supports the standard OpenAI gym interface; thus, the software can serve as a benchmark for comparison of different learning algorithms.

Download Full-text

The Design and Implementation of Intelligent High-Speed Disk Storage Protocols

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.513-517.889 ◽

2014 ◽

Vol 513-517 ◽

pp. 889-892

Author(s):

Jin Xing Shen

Keyword(s):

Data Storage ◽

High Speed ◽

Storage System ◽

Area Network ◽

Time Data ◽

Storage Area Network ◽

Disk Storage ◽

Network Storage System ◽

Memory Buffer ◽

Storage Area

In this paper, through the analysis of the redundant array of independent disks system (RAID), storage area network system (SAN), network storage system (NAS), based on RAID on the system FPGA is used to design a set of intelligent high speed disk storage protocol, through the test and analysis can meet the continuous data acquisition, the real-time data storage needs, in the software through providing high access speed increases memory buffer, large storage capacity and higher data security.

Download Full-text

A study of Storage Area Networks and issues in its management

Journal of University of Shanghai for Science and Technology ◽

10.51201/jusst/21/05257 ◽

2021 ◽

Vol 23 (06) ◽

pp. 238-245

Author(s):

Varsha Kulkarni ◽

◽

Dr. Nagaraj Bhat ◽

Keyword(s):

Data Center ◽

Large Scale ◽

Virtual Machines ◽

Single Point ◽

Storage System ◽

Area Network ◽

Storage Area Network ◽

Storage Devices ◽

Storage Area ◽

Local Storage

A data center has hundreds of servers and storage devices running on virtual machines that can be deployed and migrated over servers as per the requirement. If each server uses local storage, migration of this storage and restoration is mandatory. An attempt to organize and track storage throughout the data center is quite tedious. Using a dedicated storage system like a storage array, it possible to collectively monitor and manage such a network. A storage area network is essentially a network dedicated to storage devices. A storage area network can interconnect devices in all its layers, therefore improving storage availability. Interconnecting all elements in SAN also reduces the chances of a single point of failure. Using the storage devices collectively improves their utilization. SAN offers to manage and maintain all devices in the network. Although SAN is beneficial, it has drawbacks when configuring, monitoring, and managing components in a large-scale network. This paper consolidates the problems associated with SAN and offers possible solutions to overcome them.

Download Full-text

METHODOLOGY TO DETERMINE GUARANTEED TRAFFIC DELIVERY RATE IN CORPORATE WLAN

ВЕСТНИК ВОРОНЕЖСКОГО ГОСУДАРСТВЕННОГО ТЕХНИЧЕСКОГО УНИВЕРСИТЕТА ◽

10.36622/vstu.2020.16.5.006 ◽

2020 ◽

pp. 38-49

Author(s):

Л.И. Абросимов ◽

М.А. Руденкова ◽

Х. Хаю

Keyword(s):

Wireless Local Area Network ◽

Local Area Network ◽

Discrete Event ◽

Local Area ◽

Multimedia Traffic ◽

Area Network ◽

Event Modeling ◽

Channel Access Control ◽

Mathematical Relationships ◽

The Given

Целью работы является повышение качества обслуживания мультимедийного трафика в корпоративных беспроводных локальных вычислительных сетях за счёт средств, обеспечивающих согласование интенсивности мультимедийного трафика и производительности беспроводной локальной вычислительной сети. Для достижения поставленной цели установлены зависимости математического ожидания времени обслуживания пакета с помощью дискретно-событийного моделирования для заданных структур беспроводной локальной вычислительной сети. Разработана аналитическая модель беспроводной локальной вычислительной сети и получены математические соотношения для расчёта гарантированной интенсивности мультимедийного трафика. С помощью дискретно-событийного моделирования и полученных математических соотношений установлены зависимости гарантированной интенсивности мультимедийного трафика для заданных параметров структуры беспроводной локальной вычислительной сети, параметров беспроводного канала связи и канальных протоколов управления доступом The aim of the work is to improve the quality of service for multimedia traffic in corporate wireless local area networks at the expense of means that ensure the coordination of the intensity of multimedia traffic and the performance of the wireless local area network. To achieve this goal, the dependencies of the mathematical expectation of the packet service time are established using discrete-event modeling for the given structures of the wireless local area network. An analytical model of a wireless local area network was developed; and mathematical relationships were obtained for calculating the guaranteed intensity of multimedia traffic. Using discrete-event modeling and the obtained mathematical relationships, the dependences of the guaranteed intensity of multimedia traffic for the given parameters of the structure of the wireless local area network, the parameters of the wireless communication channel and channel access control protocols were established

Download Full-text

The Analysis of Modern Data Storage Systems

NBI Technologies ◽

10.15688/nbit.jvolsu.2019.1.5 ◽

2019 ◽

pp. 25-30

Author(s):

Vadim Shevtsov ◽

Evgeny Abramov

Keyword(s):

Data Storage ◽

Cloud Storage ◽

Storage System ◽

Area Network ◽

Storage Area Network ◽

File Storage ◽

Hardware System ◽

Data Volume ◽

Storage Area ◽

Block Storage

Today, Storage Area Network and Cloud Storage are the common Storage System. Storage Area Network includes NAS, SAN, DAS systems. Cloud Storage includes object storage, file storage, block storage. Storage Area Network is an important technology because it may give a lot of data volume with a high recovery chance and secure access, work and central management with data. Cloud Storage has many advantages: data mobility, teamwork, stability, scalability, quick start. The main threats include destruction, theft, corruption, unauthentication, replacement, blocking. Storage Area Network components (architecture elements, protocols, interfaces, hardware, system software, exploitation) have a lot of vulnerabilities. Cloud Storage may be attacked by software, functional elements, clients, hypervisor, management systems. A lot of companies design storage solutions: DropBox, QNAP, WD, DELL, SEAGATE.

Download Full-text

A Discrete Event Modeling and Simulation of Wave Division Multiplexing Unidirectional Slotted Ring Metropolitan Area Network

Journal of Computer Science ◽

10.3844/jcssp.2009.456.465 ◽

2009 ◽

Vol 5 (6) ◽

pp. 456-465 ◽

Cited By ~ 1

Author(s):

H.A. Fuaâ€™ad ◽

S. Shamala ◽

M. Othman ◽

Z. Zuriati

Keyword(s):

Modeling And Simulation ◽

Metropolitan Area ◽

Discrete Event ◽

Area Network ◽

Metropolitan Area Network ◽

Event Modeling

Download Full-text

Campus Cloud Storage and Preservation

Principles, Methodologies, and Service-Oriented Approaches for Cloud Computing ◽

10.4018/978-1-4666-2854-0.ch012 ◽

2013 ◽

pp. 284-301

Author(s):

Jinlei Jiang ◽

Xiaomeng Huang ◽

Yongwei Wu ◽

Guangwen Yang

Keyword(s):

Data Storage ◽

Cloud Storage ◽

Storage System ◽

Data Availability ◽

Area Network ◽

Storage Area Network ◽

Technology System ◽

System Maintenance ◽

Data Location ◽

Capacity Data

We are now living in the era of big data. The large volume of data raises a lot of issues related to data storage and management, stimulating the emergence of Cloud storage. Unlike traditional storage systems such as SAN (Storage Area Network) and NAS (Network Attached Storage), Cloud storage is delivered over a network and has such features as easy to scale and easy to manage. With Cloud storage shielding complex technical details such as storage capacity, data location, data availability, reliability and security, users can then concentrate on their business rather than IT (Information Technology) system maintenance. However, it is not an easy task to develop a Cloud storage system because multiple factors are involved. In this chapter, the authors show their experience in the design and implementation of a Cloud storage system. They detail its key components, namely the distributed file system Carrier and the data sharing service Corsair. A case study is also given on its application at Tsinghua University.

Download Full-text

A Control Strategy Based on Deep Reinforcement Learning Under the Combined Wind-Solar Storage System

2020 IEEE 3rd Student Conference on Electrical Machines and Systems (SCEMS) ◽

10.1109/scems48876.2020.9352436 ◽

2020 ◽

Author(s):

Shiying Huang ◽

Ming Yang ◽

Changhang Zhang ◽

Jiangyang Yun ◽

Yuan Gao ◽

...

Keyword(s):

Reinforcement Learning ◽

Control Strategy ◽

Storage System

Download Full-text

Integrating Production Planning with Truck-Dispatching Decisions through Reinforcement Learning While Managing Uncertainty

Minerals ◽

10.3390/min11060587 ◽

2021 ◽

Vol 11 (6) ◽

pp. 587

Author(s):

Joao Pedro de Carvalho ◽

Roussos Dimitrakopoulos

Keyword(s):

Reinforcement Learning ◽

Discrete Event ◽

Mining Operations ◽

Fixed Sequence ◽

Q Learning ◽

Reward Function ◽

Copper Gold ◽

Mining Complex ◽

Learning Reinforcement ◽

Operational Plan

This paper presents a new truck dispatching policy approach that is adaptive given different mining complex configurations in order to deliver supply material extracted by the shovels to the processors. The method aims to improve adherence to the operational plan and fleet utilization in a mining complex context. Several sources of operational uncertainty arising from the loading, hauling and dumping activities can influence the dispatching strategy. Given a fixed sequence of extraction of the mining blocks provided by the short-term plan, a discrete event simulator model emulates the interaction arising from these mining operations. The continuous repetition of this simulator and a reward function, associating a score value to each dispatching decision, generate sample experiences to train a deep Q-learning reinforcement learning model. The model learns from past dispatching experience, such that when a new task is required, a well-informed decision can be quickly taken. The approach is tested at a copper–gold mining complex, characterized by uncertainties in equipment performance and geological attributes, and the results show improvements in terms of production targets, metal production, and fleet management.

Download Full-text