No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Design and Implementation of SCSI Target Emulator
Abstract
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.
... also implemented a storage device emulator, but with an analytical model for the simulator. Some storage device emulators do not alter performance to mimic any particular device and essentially map from one interface to another [88,29]; these are unrelated. ...
Performance models for storage devices are an important part of simulations of large-scale computing systems. Storage devices are traditionally modeled using discrete event simulation. However, this is expensive in terms of computation, memory, and configuration. Configuration alone can take months, and the model itself requires intimate knowledge of the internal layout of the device. The difficulty in white-box model creation has led to the current situation, where there are no current, precise models. Automatically learning device behavior is a much more desirable approach, requiring less expert knowledge, fewer assumptions, and less time. Other researchers have created behavioral models of storage device performance, but none have shown low per-request errors. By making use of only a few high-level domain-specific assumptions, such as spatial periodicity and the existence of a logical-to-physical mapping, neural nets can learn to predict access time with low per-request errors. Providing neural nets with specific sinusoidal inputs allows them to generate periodic output, which is necessary for this problem. Weight sharing in the neural net accounts for regularities in the structure of the input, which reduces data requirements and error by reducing the number of free parameters. A trigonometric change of variables in the output of the neural net removes a discontinuity in the objective function, which makes the problem more amenable to neural nets and reduces error. Combining these approaches, we demonstrate that a neural net can predict access times with a mean absolute error of about 0.187 ms over a small portion of a hard disk drive, and a mean absolute error of about 2.120 ms over half of a hard disk drive.
The design of the underlying storage subsystems for multimedia applications faces significant challenges for capability Ahigh
I/O performance and availability requirements. A RAID system is defined as a storage system that distributes data redundantly
across array of disks and can provide high throughput as well as higher availability. In this paper, we present a novel cache
scheme (for short Stripe-cache) for building a multimedia oriented RAID system. This efficient stripe-cache has the following
innovations: (1) Hierarchy architecture with different block structure according to the underlying data layout that exploits
temporal locality and spatial locality (2) Timing-transfer and replacing scheme forwardly move data block in cache for special
efficiency. We built up the multimedia oriented RAID system as a device driver module upon the X86-Linux platform with the
stripe-cache. Evaluation results show that the system can offer much higher I/O performance in handling multimedia applications
than lots of conventional storage systems.
The speed of storing and fetching data on SCSI disks has a great restriction on the efficiency of SAN based on Fiber Channel Network. In this paper, a high-efficient FC-SAN storage method attributed to the statistics conclusions of the file system workload is designed and implemented. By evaluating the load heavy or light, the system selects DISK or RAM devices as the main storage media according to the different I/O operations, i.e., file operations of larger size will be done on disk devices while reading/writing smaller size or file attribution operations will be done in RAMDISK devices. This method reduces the operations that store and fetch data on physical disks, adjusts the speed difference between CPU and DISK, and increases the efficiency of the system as a whole. Experimental testing proves good results and demonstrates that the application of RAM devices for I/O operation could increase the read efficiency of FC-SAN at about 50% to 100% and write performance at about 10% to 40%.
The iSCSI protocol enables accessing SCSI I/O devices over an IP network. TCP is used as a transport for SCSI I/O commands. We describe the design considerations and decisions in defining the iSCSI protocol: why we use TCP how multiple TCP connections can be used to increase performance and reliability, why we require allegiance of a command to a particular TCP connection, the importance of Direct Data Placement, various levels and complexity of error recovery, security and naming issues.
The iSCSI protocol specifies how to access SCSI storage devices over a TCP network. In this article we give a brief introduction to the iSCSI protocol and a brief comparison with alternative technologies. We then discuss the basic features of the iSCSI protocol: sessions, naming, discovery, security, data placement, and recovery.
This paper describes the architecture of a set of kernel components for developing and testing storage area net- work transport protocols under Linux. This software is in- tended for several uses: as a general prototype for network transport protocol development; as a reference implement- ation of the iSCSI protocol currently under development for standardization by IETF; as a basis for conformance testing for iSCSI; and as a testbed for development of in- teroperability test suites for iSCSI.
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.
The initial iSCSI products provide a means to connect FC SAN islands across IP networks. This paper describes the implementation of an IP-SAN where the disk subsystem is a virtual array of individually Ethernet attached IP-addressable disks. By replacing the conventional peripheral bus and loop interconnects with a switched Gigabit Ethernet network, the virtual disk array scales continually and dynamically with the simple addition of Ethernet switches and disks, as well as system-wide disk sharing and inherent high availability. In fully exploiting the IP and Ethernet technologies and user knowledge, this architecture pushes the IP SAN evolution toward a truly scalable, manageable, yet flexible and cost-effective data storage system that will be a seamless part of the networked infrastructure.
Organizations are processing and storing an exploding volume of
information. Because of this, they are looking for new ways to access
their data more quickly and reliably across their networks. This has led
organizations to begin using storage area networks (SANs) based on fibre
channel technology. SANs link storage devices (such as disks, disk
arrays, and tape drives) to create a pool of storage that users can
access directly. In essence, SANs uncouple storage devices from the LAN
and put them on their own network. This approach relieves bandwidth
congestion on the LAN and permits much faster data access than today's
LAN-based storage streams. SANs also permit better storage management
and more fault tolerance. SANs have traditionally used Small Computer
Systems Interface (SCSI) technology, but now fibre channel technology is
allowing SANs to transmit data at higher speeds over greater distances,
which should accelerate the adoption of the storage networks. Fibre
channel is also more scalable than SCSI. Fibre-channel-based SAN
technology has so many advantages that it will become increasingly
popular, particularly as researchers correct its problems and develop
ways for it to transmit data over even longer distances
Internet SCSI), draft-ietf-ips-iscsi-20
- J Satran
- Iscsi
Satran J. iSCSI (Internet SCSI), draft-ietf-ips-iscsi-20.txt
PAN Jiaming (潘家铭) et al：Design and Implementation of SCSI Target Emulator
(January, 2003), http://ietf.org/html.charters/ipscharter.html.
Technical report: Design and implementation of TH-MSNS
- Shu Jiwu
- Yao Jun
Shu Jiwu, Yao Jun. Technical report: Design and
implementation of TH-MSNS. Department of Computer
Science and Technology, Tsinghua University, China,
http://graduate.cs. tsinghua. edu.cn/~yaojun/, 2003. (in
Chinese).
Zheng Weimin. A design and implementation of target simulator module in SAN system
- Li Bigang
- Shu Jiwu
Li Bigang, Shu Jiwu, Zheng Weimin. A design and
implementation of target simulator module in SAN system.
Primary Commands-3(SPC-3)
- Ansi Scsi
ANSI. SCSI Primary Commands-3(SPC-3), Sep. 2002,
http://www.t10.org.
A design and implementation of target simulator module in SAN system
- Li Bigang
- Shu Jiwu
- Zheng Weimin
Li Bigang, Shu Jiwu, Zheng Weimin. A design and
implementation of target simulator module in SAN system.