01-09-2016, 12:17 PM
1452092975-iscrc.pdf (Size: 388.41 KB / Downloads: 6)
ISCSI is proposed as a possible solution to building future storage systems. However, using iSCSI raises numerous questions
about its implications on system performance. This lack of understanding of system I/O behavior in modern and future systems
inhibits providing solutions at the architectural and system levels. Our main goals in this work are to understand the behavior
of the application server (iSCSI initiator), to evaluate the overhead introduced by iSCSI compared to systems with directlyattached
storage, and to provide insight about how future storage systems may be improved. We examine these questions in the
context of commodity iSCSI systems that can benefit most from using iSCSI. We use commodity PCs with several disks as
storage nodes and a Gigabit Ethernet network as the storage network. On the application server side we use a broad range of
benchmarks and applications to evaluate the impact of iSCSI on application and server performance. Our analysis reveals how
iSCSI affects application performance and shows that building next generation, network-based I/O architectures, requires
optimizing I/O latency, reducing network and buffer cache related processing in the host CPU, and increasing the sheer
network bandwidth to account for consolidation of different types of traffic. ISCSI is the block-level storage protocol that lets
users create a separate storage network using Ethernet. This protocol uses Ethernet as a transport for data from servers to
storage area Networks. Now a day, fiber channels are becoming dominant in storage area networks. With fast developing
network technologies iSCSI made all companies to seriously think about the implementation of future storage network. ISCSI
storage network based upon Cisco’s catalyst 6500 gigabit Ethernet switch and SN5420 storage router using fiber channel for
storage access.
Keywords: ISCSI, Internet Protocol (SAN), Fiber channels
1. INTRODUCTION
In a new era of computing, data storage has changed its role from secondary with respect to CPU and RAM to
primary importance in today’s information world. Online data storage doubles every 9 months [7] due to an evergrowing
[8] demand for networked information services. In general, networked storage architectures have evolved from
network-attached storage (NAS) [7], storage area network (SAN) [4] to most recent storage over IP (IP SAN). NAS
architecture allows a storage system/device to be directly As Connected to a standard network, typically via Ethernet.
SAN technology, on the other hand, provides a simple block level interface for manipulating nonvolatile magnetic
media. Typically, a SAN consists of networked storage devices interconnected through a dedicated fiber channel (FC-4
protocol) network. The basic premise of a SAN is to replace the “point-to-point” infrastructure of server-to-storage
communications with one that allows “any-to-any” communications. A SAN provides high connectivity, scalability,
and availability using a specialized network protocol: FC-4 protocol. Deploying such a specialized network usually
introduces additional cost for implementation, maintenance, and management. Internet SCSI (iSCSI) is the most
recently emerging technology with the goal of implementing the IP SAN. Compared to FC-4, implementing SAN over
IP (IP SAN) has several advantages [4]: IP SAN can run over standard off-the-shelf network components, such as
switched Ethernet, which reduces the cost. One can extend and expand the switched network easily and quickly while
riding the cost/performance improvement trends of Ethernet. IP SAN can exploit existing IP-based protocols, and IP
SANs using iSCSI can be manage during existing and familiar IP-based tools such as SNMP, while Fiber Channel
SANs require specialize management infrastructure. A network that incorporates IP SANs need use only a single kind
of network infrastructure (Ethernet) for both data and storage traffic, whereas use of fiber channel protocol (FCP)
requires a separate kind of infrastructure (fiber channel) for storage.
Internet Protocol (IP) dominates local and wide area networks, and data storage requirements grow unabated, it
seems inevitable that these two forces converge. The Internet Small Computer Systems Interface (iSCSI) protocol unites
storage and IP networking. iSCSI enables the transport of block-level storage traffic over IP networks. It builds on two
widely used technologies — SCSI commands for storage and IP protocols for networking. iSCSI is an end-to-end
protocol for transporting storage I/O block data over an IP network. The protocol is used on servers (initiators), storage
devices (targets), and protocol transfer gateway devices. ISCSI uses standard Ethernet switches and routers to move the
ISCSI-The Future of the Storage Network
Ms. Sulbha V. Hemke1
, Dr. A. D. Gawande2
and Prof. L. K. Gautum3
1
Student, M.E. Computer Engineering, Sipna COET- Amravati
2Head of Computer Science and Engg.
3Department of Computer Science and Engg.
Web Site: www.ijaiem.org Email: editor[at]ijaiem.org, editorijaiem[at]gmail.com
Volume 2, Issue 4, April 2013 ISSN 2319 - 4847
Volume 2, Issue 4, April 2013 Page 236
data from server to storage. It also enables IP and Ethernet infrastructure to be used for expanding access to SAN
storage and extending SAN connectivity across any distance iSCSI is an end-to-end protocol for transporting storage
I/O block data over an I/P network.
2. SCSI CONCEPT:
The Small Computer Systems Interface (SCSI) is a popular family of protocols for communicating with I/O
devices, especially storage devices. SCSI is an industry standard system interface, which allows multiple peripheral
devices to be installed on the system. The number of devices that can be installed depends on the type of SCSI Host
adapter that you have installed in your system. iSCSI is based on the Small Computer Systems Interface (SCSI), which
enables host computer systems to perform block data input/output (I/O) operations with a variety of peripheral devices.
Target devices may include disk and tape devices, optical storage devices, as well as printers and Scanners. The
traditional SCSI connection between a host system and peripheral devices is based on parallel cabling, which has
inherent distance and device support limitations. For storage applications, these limitations have fostered the
development of new technologies based on networking architectures such as Fiber Channel and Gigabit Ethernet. IP
Storage networks based on serial gigabit transport layers overcome the distance, performance, scalability, and
availability restrictions of parallel SCSI implementations.
By leveraging SCSI protocols over networked infrastructures, storage networking enables flexible high-speed
block data transfers for a variety of applications, including tape backup, server clustering, storage consolidation, and
disaster recovery. The Internet SCSI (iSCSI) protocol defines a means to enable block storage applications over TCP/IP
networks. The SCSI architecture is based on a client/server model, and iSCSI takes this into account to deliver storage
functionality over TCP/IP networks. The client is typically a host system such as file server that issues requests to read
or write data. The server is a resource such as a disk array that responds to client requests. In storage parlance, the
client is an initiator and plays the active role in issuing commands. The server is a target and has a passive role in
fulfilling client requests, having one or more logical units that process initiator commands. Because of the proliferation
of the public Internet, corporations and other organizations currently have in place infrastructure and resources
dedicated to managing their TCP/IP traffic, over LANs and the Internet itself. They have deployed firewalls, routers
and other infrastructure solely to take advantage of the Internet's benefits, and protect them from malicious outside
intervention. iSCSI builds upon this built-in networking infrastructure and expertise, and provides many benefits to
users of storage -- every computer user -- at relatively low technology acquisition cost. The iSCSI protocol has two
halves to it - the initiator resides on a server or desktop computer, and sends commands to the iSCSI target, which
resides on the storage appliance. The target performs the work requested by the initiator, and sends a reply back. All
communications take place via TCP/IP.
ARCHITECTURE:
Essentially, STICS is a cache that bridges the protocol and speed disparities between SCSI and IP. Fig shows a
typical SAN implementation over IP using STICS. Any number of storage devices or server computers can be
connected to the standard Internet through STICS to form a SAN. Instead of using a specialized network or specialized
switch, STICS connects a regular host server or a storage device to the standard IP network using off-the-shelf
components. Consider STICS 1 in the diagram. It is directly connected to the SCSI HBA of Host 1 as a local storage
device. It also acts as a cache and bridge to allow Host 1 to access, at the block level, any storage device connected to
the SAN such as NAS, STICS 2, STICS 3, etc. In order to allow a smooth data access from Host 1 to the SAN, STICS
1 A STICS connects to the host via SCSI interface And connects to other STICS’ or NAS via Internet. Provides SCSI
service, caching service, naming service, and IP protocol service.
The basic structure of STICS, which consists of five main components: