SAN Storage is a critical component of many enterprise infrastructures, and the need to support steadily growing amounts of digital content, provide 24/7 access, and comply with regulatory requirements has only increased the importance of controlling costs and maximizing the efficiency of storage resources. 

The most common solution to these issues is the deployment of the Fibre Channel based storage area network 9SAN), which is typically easier to manage and back up, provides higher utilization and offer lower total cost of ownership than internal storage direct attach storage (DAS) in large environments.

However, Fibre Channel SANs also tend to have a higher cost of entry that other types of storage, these costs combined with a steep management learning curve can place medium-size enterprises.

Internet SCSI (iSCSI) technology is changing the face of networked storage, providing a cost-effective and easy to manage alternative to Fibre Channel Networks. iSCSI provide the advantages in scalability, availability, and manageability of Fibre Channel but at lower cost, helping eliminate this barrier to entry for organizations that want to deploy a SAN without investing in a Fibre Channel infrastructure.

iSCSI storage can be implemented in several ways, including as an alternative to Fibre Channel for enterprise that have not yet implemented a SAN, as a complement to an existing Fibre Channel SAN, and as a complement to network attached storage (NAS), for enterprises seeking integrated file and block level storage access.

Comparing traditional storage approaches

To understand the advantages offered by iSCSI technology, enterprises should understand the strengths and weaknesses of traditional approaches: DAS, Internal Storage, Networked storage such as NAS and Fibre Channel SANs.

Direct Attach Storage – DAS is a logical extension of internal storage, and typically consists of a rack of external hard drives utilized by a single server to expand capacity although innovations.

Other improvements like multiple drive option: Serial ATA (SATA) and Serial Attached SCSI (SAS), enhanced scalability and data availability and the introduction of applications with built-in replication capabilities like Microsoft Exchange Server 2007 have further increased the usefulness of DAS but the number of attached hosts is limited, DAS can still create the same sort of management difficulties as internal storage in large environments.

Internal StorageInternal storage is typically simple and inexpensive but does not scale well. It has the storage capacity limitation or maximum processing power is reached, administrators must add server to meet the new requirements, potentially under utilizing either their storage or processing resources.

Networked storage – NAS and SANs enable many hosts to share storage resources, helping enhance storage utilization, simplify management, and reduce total cost of ownership compared with internal storage and DAS. NAS is primarily used for application data. Given that the initial entry cost for SAN is typically much higher than internal storage or DAS, deployment has primarily been restricted to large enterprises with the resource to invest in this type of infrastructure.


Understanding iSCSI

This is where iSCSI fills a longtime gap in enterprise storage – iSCSI enables small and medium-size enterprises to deploy SAN technology that may have been previously out of reach, while enabling large enterprises to create a complementary storage tier for secondary servers or expand their existing networked storage to include departments and work group, and consolidate storage for increased capacity utilization.

The iSCSI protocol allows SCSI commands to be sent over an Ethernet network, enabling enterprises to build SANs with standard equipment such as Gigabit Ethernet switches and Network Interface Card (NICs). An iSCSI SAN access and transfer data in blocks, with each connected server seeing the remote storage array as a local hard drive. (This is different implementation from NAS, which abstracts and organizes block data so that connected systems see the storage as files.)


Clarifying common misperceptions

Several common misperceptions have persisted regarding how iSCSI compares to traditional Fibre Channel deployments. SCSI becoming an increasingly mainstream technology and large vendor like Dell, EMC, IBM, HP offering iSCSI products, these notions are no longer necessary true.

Performance – The first misperception is that iSCSI cannot provide he performance necessary for enterprise applications. Many enterprises compare Fibre Channel 4 Gbps bandwidth to Ethernet 1Gbps bandwidth and assume that Fibre Channel is four times faster than iSCSI. However, many key applications, the “size of the pipe” is not the bottleneck for performance. Many key applications have random I/O data patterns and the performance bottleneck ends up being the time it takes to write and read data from hard disk drives, not the network bandwidth.

Network Security – A very common misperception is that iSCSI SANs are not as secure as Fibre Channel SANs. In fact, when logically or physically separated, iSCSI networks are just as secure as Fibre Channel. An easy way to secure iSCSI traffic is to implement a dedicated iSCSI network using an industry standard Gigabit Ethernet switch. If iSCSI is used over a shared data network, implementing virtual LANs (VLANs) typically should be sufficient to secure iSCSI data from unwanted access.

Manageability – Another misperception is that iSCSI is more difficult to manage than Fibre Channel. Storage device management – for example, creating a RAID groups and virtual disks – is typically network independent, and thus iSCSI and Fibre Channel exhibit equivalent manageability in this respect. Many IT staffs are more familiar with Ethernet than Fibre Channel, iSCSI networks may be easier to manage.