File Assignment Policy in Network Storage System

Network storage increase capacity and scalability of storage system, data availability and enables the sharing of data among clients. When the developing network technology reduce performance gap between disk and network, however, mismatched policies and access pattern can significantly reduce network storage performance. So the strategy of data placement in system is an important factor that impacts the performance of overall system. In this paper, the two algorithms of file assignment are presented. One is Greed partition that aims at the load balance across all NADs (Network Attached Disk). The other is Sort partition that tries to minimize variance of service time in each NAD. Moreover, we also compare the performance of our two algorithms in practical environment. Our experimental results show that when the size distribution (load characters) of all assigning files is closer and larger, Sort partition provides consistently better response times than Greedy algorithm. However, when the range of all assigning files is wider, there are more small files and access rate is higher, the Greedy algorithm has superior performance in compared with the Sort partition in off-line.

HUSTserver:Implementation for Reliable and High-Performance Network Attached Storage System

Network attached storage (NAS) with the properties of improved scalability, simplified management, low cost and balanced price performance, is desirable for high performance storage systems applied to extensive areas. Unfortunately, it also has some disadvantages such as increased network workload, and inconvenience in disaster recovery. To overcome these disadvantages, we propose a channel bonding technique and provide hot backup functions in the designed NAS system, named HUSTserver. Channel bonding means merging multiple Ethernet channels into integrated one, and that the data packets can be transferred through any available network channels in a parallel mode. The hot backup function provides automatic data mirroring among servers. In this paper, we first describe the whole system prototype from a software and hardware architecture view. Then, multiple Ethernet and hot backup technologies that distinguish HUSTserver from others are discussed in detail. The findings presented demonstrate that network bandwidth can be scaled by the use of multiple commodity networks. Dual parallel channels of commodity 100 Mbps Ethernet are both necessary and sufficient to support the data rates of multiple concurrent file transfers. And the hot backup function introduced in our system provides high data accessibility.

Security System in United Storage Network and Its Implementation

With development of networked storage and its applications, united storage network (USN) combined with network attached storage (NAS) and storage area network (SAN) has emerged. It has such advantages as high performance, low cost, good connectivity, etc. However the security issue has been complicated because USN responds to block I/O and file I/O requests simultaneously. In this paper, a security system module is developed to prevent many types of attacks against USN based on NAS head. The module not only uses effective authentication to prevent unauthorized access to the system data, but also checks the data integrity. Experimental results show that the security module can not only resist remote attacks and attacks from those who has physical access to the USN, but can also be seamlessly integrated into underlying file systems, with little influence on their performance.

Storage Assignment Optimization in a Multi-tier Shuttle Warehousing System

The current mathematical models for the storage assignment problem are generally established based on the traveling salesman problem（TSP）,which has been widely applied in the conventional automated storage and retrieval system（AS/RS）.However,the previous mathematical models in conventional AS/RS do not match multi-tier shuttle warehousing systems（MSWS） because the characteristics of parallel retrieval in multiple tiers and progressive vertical movement destroy the foundation of TSP.In this study,a two-stage open queuing network model in which shuttles and a lift are regarded as servers at different stages is proposed to analyze system performance in the terms of shuttle waiting period（SWP） and lift idle period（LIP） during transaction cycle time.A mean arrival time difference matrix for pairwise stock keeping units（SKUs） is presented to determine the mean waiting time and queue length to optimize the storage assignment problem on the basis of SKU correlation.The decomposition method is applied to analyze the interactions among outbound task time,SWP,and LIP.The ant colony clustering algorithm is designed to determine storage partitions using clustering items.In addition,goods are assigned for storage according to the rearranging permutation and the combination of storage partitions in a 2D plane.This combination is derived based on the analysis results of the queuing network model and on three basic principles.The storage assignment method and its entire optimization algorithm method as applied in a MSWS are verified through a practical engineering project conducted in the tobacco industry.The applying results show that the total SWP and LIP can be reduced effectively to improve the utilization rates of all devices and to increase the throughput of the distribution center.

Design of Enterprise Storage Architecture for Optimal Business Continuity

This paper presents a solution for optimal business continuity, with storage architecture for enterprise applications, which will ensure zero data loss and quick recovery. The solution makes use of Internet protocol storage area network （IPSAN）, which is used for data management without burdening the application server, as well as mix of synchronous and semi-synchronous replication techniques to replicate data to remote disaster recovery site. We have presented the detailed design of both synchronous and semi-synchronous with case study of using open source database postgres to prove our point for optimal business continuity. The theoretical presentation is also given for the same.

Attribute-Based Re-Encryption Scheme in the Standard Model

In this paper, we propose a new attribute-based proxy re-encryption scheme, where a semi-trusted proxy, with some additional information, can transform a ciphertext under a set of attributes into a new ciphertext under another set of attributes on the same message, but not vice versa, furthermore, its security was proved in the standard model based on decisional bilinear Diffie-Hellman assumption. This scheme can be used to realize fine-grained selectively sharing of encrypted data, but the general proxy rencryption scheme severely can not do it, so the proposed schemecan be thought as an improvement of general traditional proxy re-encryption scheme.

Distributed Storage Cluster Design for Remote Mirroring Based on Storage Area Network

With the explosion of information nowadays, applying data storage safety requirements has become a new challenge, especially in high data available cluster environments. With the emergence of Storage Area Networks （SANs）, storage can be network-based and consolidated, and mass data movements via Fiber Channels （FCs） can be of very high speed. Based on these features, this paper introduces a dual-node storage cluster designed for remote mirroring as a concurrent data replication method to protect data during system failures. This design takes full advantage of a SAN system＇s benefits, and it adopts a synchronous protocol to guarantee a fully up-to-date data copy on the remote site. By developing a Linux kernel module to control the I/O flow and by using the technologies of software Logic Unit Number （LUN） masking, background online resynchronization and a self-management daemon, we have achieved a reliable mirroring system with the characteristics of server-free data replication, fault tolerance, online disaster recovery and high performance. In this study, we implemented the design in a remote mirror subsystem built on a software Fiber Channel Storage Area Network （FC-SAN） system.

Towards energy-efficient storage placement in large scale sensor networks

Data storage has become an important issue for energy efficient data management in sensor networks. In this paper, we investigate the optimized storage placement problem in large scale sensor networks, aiming to achieve minimized energy cost. In order to efficiently deal with large scale deployment areas with irregular shape, we propose to utilize the hop as the computation unit instead of the node, such that computation complexity can be greatly reduced. We propose methodologies to solve the optimization problem both in situations for limited and unlimited numbers of storage units. The ultimate goal of this paper is to give fundamental guidance for optimized storage placement in large scale sensor networks. Simulation results show that our methodologies can greatly reduce the overall energy consumption compared to other strategies.

Study on I/O response time bounds of networked storage systems

In order to predict and improve the performance of networked storage systems,this paper explored the rela-tionship between the system I/O response time and its per-formance factors by quantitative analytical method.Through analyzing data flow in networked RAID storage system,we established its analytical model utilizing closed queueing networks and studied the performance bounds of the system I/O response time.Experimental results show that the theo-retical bounds are found to be in agreement with the actual performance bounds of the networked RAID storage system and reflect the dynamic trend of its actual performance.Furthermore,it concludes that the CPU processing power and cache hit rate of the central storage server are the key factors affecting the I/O response time as the concurrent jobs are lower,while the network bandwidth and cache hit rate of the central storage server become the key factors as the concurrent jobs go higher.

Design and Implementation of a Storage Virtualization System Based on SCSI Target Simulator in SAN

The ideal storage virtualization system is compatible with all operating systems in storage area networks (SANs). However, current storage systems on clustered hosts and multiple operating systems are not practical. This paper presents a storage virtualization system based on a SCSI target simulator in a SAN to solve these problems. This storage virtualization system runs in the target hosts of the SAN, dynamically stores the physical information, and uses the mapping table method to modify the SCSI command addresses. The system uses the bitmap technique to manage the free space. The storage virtualization system provides various functions, such as logical volume resizing, data mirroring, and snapshots, and is compatible with clustered hosts and multiple operating systems, such as Windows NT and RedHat.

FlexPolh adaptive event polling for network-intensive applications

In today＇s data centers supporting Internet-scale computing and input/output （I/0） services, increasingly more network-intensive applications are deployed on the network as a service. To this end, it is critical for the applications to quickly retrieve requests from the network and send their responses to the network. To facilitate this network func- tion, operating system usually provides an event notification mechanism so that the applications （or the library） know if the network is ready to supply data for them to read or to receive data for them to write. As a widely used and repre- sentative notification mechanism, epoll in Linux provides a scalable and high-performance implementation by allowing applications to specifically indicate which connections and what events on them need to be watched. As epoll has been used in some major systems, includ- ing key-value （KV） systems, such as Redis and Memcached, and web server systems such as NGINX, we have identified a substantial performance issue in its use. For the sake of efficiency, applications usually use epoll＇s system calls to inform the kernel exactly of what events they are interested in and always keep the information up-to-date. However, in a system with demanding network traffic, such a rigid main- tenance of the information is not necessary and the excess number of system calls for this purpose can substantially de- grade the system＇s performance. In this paper, we use Redis as an example to explore the issue. We propose a strategy of informing the kernel of the interest events in a manner adaptive to the current network load, so that the epoll system calls can be reduced and the events can be efficiently deliv- ered. We have implemented an event-polling library, named as FlexPoll, purely in user-level without modifying any ker- nel code. Our evaluation on Redis shows that the query through- put can be improved by up to 46.9% on micro-benchmarks, and even up to 67.8% on workloads emulating real-world ac- cess patterns. FlexPoll is a generic mechanism thus it can be adopted by other applications in a straightforward manner, such as NG1NX and Memcached.

Design and Implementation of SCSI Target Emulator

A SCSI target emulator is used in a storage area network （SAN） environment to simulate the behavior of a SCSI target for processing and responding to I/O requests issued by initiators. The SCSI target emulator works with general storage devices with multiple transport protocols. The target emulator utilizes a protocol conversion module that translates the SCSI protocols to a variety of storage devices and implements the multi-RAID-level configuration and storage visualization functions. Moreover, the target emulator implements RAM caching, multi-queuing, and request merging to effectively improve the I/O response speed of the general storage devices. The throughput and average response times of the target emulator for block sizes of 4 KB to 128 KB are 150% faster for reads and 67% faster for writes than the existing emulator. With a block size of 16 KB, the I/O latency of the target emulator is only about 20% that of the existing emulator.

SCSI Target Simulator Based on FC and IP Protocols in TH-MSNS

Storage area networks （SAN） generally use fiber channel （FC） disks, but they are expensive, hard to scale, and incompatible with previous storage arrays. This paper describes an SCSI interface disk array, which is controlled by a simulated target controller, as the target device of an SAN system. A detailed target- mode SCSI middle level design and implementation is described with interfaces to efficiently supervise the entire target simulator. The target simulator has control flexibility, wide applicability, reduced SAN system cost, and compatibility with current SAN systems. The present mass data storage system uses a 2-Gb/s fiber network. Performance and compatibility tests demonstrate the efficiency of the target simulator design.

Design and Implementation of a Fibre Channel Target Driver Supporting SCSI

Fibre channel storage area networks (FC-SAN) are effective solutions to address storage man- agement problems caused by very large volumes of data. But the expense of fibre channel devices limits FC-SAN applications. The use of IP networks instead of fibre channel networks will reduce SAN cost, but will also reduce the performance. Therefore, small computer system interface (SCSI) devices were consid- ered to replace FC disks to reduce the SAN cost. A driver for the FC network adapter and the FC target, de- signed and implemented to support this structure, obeys the SCSI protocol and works in target mode with 200 MB/s bandwidth. The FC target architecture and implementation were compared with the FC initiator. The SCSI command transfer process in the FC layer was described. The performance test results show that the maximum I/O throughput reachs 167 MB/s for read requests and 196 MB/s for write requests (FC band- width is 200 MB/s), verifying that the FC target is very efficient. The modularization, efficiency, and low cost of the FC target will enable SAN and fibre channel to be more widely used in applications.