iSCSI learns new tricks

Many people still believe iSCSI can’t cut it in the enterprise, which is better served by Fibre Channel. iSCSI vendors are doing their best to prove that assumption wrong.

Ever since 10Gbps Ethernet arrived, iSCSI has competed in the numbers game — especially when stacked against 8Gbps Fibre Channel. But raw throughput isn’t everything. To win the debate, Fibre Channel partisans simply look down their noses and observe that no Ethernet-based storage protocol, iSCSI included, can compete with FC’s inherent performance advantages.

That conclusion still holds true to a certain extent, but the performance gap separating FC and iSCSI keeps narrowing. Eventually, other factors such as ease-of-use and flexibility may be more important. To decide which tool is right for the job, you need a solid feel for what differentiates FC from iSCSI; simply repeating that FC is “faster,” while true, glosses over vital distinctions.

Fibre Channel on top

No amount of fancy new technology will change the fundamental difference between the iSCSI and Fibre Channel protocols: Fibre Channel, regardless of the physical media that it runs on, operates as a layer-two (data link) protocol, while iSCSI operates at the layer-seven (application) level.

Though this distinction has a number of ramifications, the most critical one is that Fibre Channel implements its own flow control and error correction mechanisms, while iSCSI depends on equivalent mechanisms built into TCP/IP.

Processing Fibre Channel frames takes significantly less time because TCP and IP headers do not need to be generated and decoded for each frame sent. Since Fibre Channel is a purpose-specific protocol, all frame management is performed in dedicated hardware and thus can operate with far less latency. Meanwhile, since iSCSI depends upon TCP/IP to work, encoding and decoding these headers is a necessary step. Increasingly, it’s possible to offload much of this work (such as generating error-correction checksums) into hardware that decreases system CPU load, but some latency is still added.

Second, because of comprehensive built-in flow control mechanisms, Fibre Channel is able to shuffle data across the network without any frame loss due to congestion. This makes Fibre Channel an assured delivery or lossless protocol; unless there’s some form of misconfiguration or hardware failure, any frame that goes in one end will come out the other without being lost along the way. If there’s too much data to send across the pipe or one side of the conversation can’t keep up, flow control mechanisms restrict the amount of data being sent and prevents loss-resulting congestion from occurring.

The same cannot be said for Ethernet, which was originally developed as a best-effort packet delivery protocol, implying that packets will be lost when congestion occurs. Because of that, higher-level protocols (such as TCP) must be able to detect that the loss has taken place and deal with retransmitting the lost data segments. In general, this works well in terms of ensuring the packets are eventually delivered — there’s never an issue of data corruption that can result from packet loss. However, in networks where very large amounts of data are being moved and congestion is frequent, performance can suffer heavily.

iSCSI’s advantages

Though iSCSI may suffer from some performance disadvantages, it makes up for them with simplicity and flexibility. Unlike traditional Fibre Channel, iSCSI doesn’t require its own dedicated network infrastructure; instead, it can share the same Ethernet network that the rest of the data center runs on. And since it runs as TCP/IP, iSCSI traffic can be routed outside of a single broadcast domain — say, across a WAN to facilitate replication. This can’t be done with Fibre Channel without expensive Fibre Channel-to-IP (FCIP) gateways or, occasionally, proprietary vendor-specific solutions that use IP networks.

iSCSI is also very simple to use. While zoning new storage in a Fibre Channel network typically requires a significant amount of knowledge and effort, the same task can be accomplished using iSCSI in far less time and with much of the same skill set that most network administrators already have. To be sure, tuning iSCSI to reach its full performance potential (including such tasks as configuring MPIO) requires extra know-how, but it ends up being easier for most network admins to learn.

Enter Fibre Channel over Ethernet

The advent of Fibre Channel over Ethernet (FCoE) has significantly changed the landscape. FCoE is made possible by new extensions to traditional Ethernet collectively termed “data center bridging” (DCB) or sometimes “converged enhanced Ethernet” (CEE). DCB essentially extends the Ethernet protocol to support the same types of lossless throughput provided by Fibre Channel by implementing similar flow control and prioritization mechanisms.

While implementing DCB almost always requires new hardware built to the task, it finally allows the consolidation of traditional FC networks into Ethernet networks. Though there is an initial cost to implement DCB-capable switches, this places FCoE much closer to iSCSI in terms of the overall amount of flexibility it offers. However, like traditional Fibre Channel, FCoE is still a layer-two protocol — that is, it can’t be routed outside of a broadcast domain and it retains some of the complexity that made Fibre Channel more difficult to manage.

iSCSI tags along for the ride

With DCB-capable switches becoming more common in the data center, some traditionally iSCSI-only storage vendors have jumped at the opportunity to take advantage. Specifically, Dell EqualLogic is in the process of fielding an early production release of its 5.1.x firmware for the PeerStorage line of iSCSI storage. Among other new features and bug fixes, this new release will bring DCB compatibility to the entire line of storage, effectively allowing iSCSI to benefit from the same lossless, prioritized functionality that made FCoE a reality. To my knowledge, this is the first time that a major iSCSI vendor has offered support for DCB, but you can bet it probably won’t be the last.

Putting it all together

While iSCSI will always suffer from the fact that it is a higher-level and, thus, higher-latency protocol, it will also continue to benefit from being generally easier to deploy and understand. Many vendors have already taken the path of allowing customers to choose which kinds of storage interconnects they want for a given implementation, potentially allowing the same storage platform to support traditional FC, FCoE, and DCB-enhanced iSCSI all at the same time. As a result, which transport to use in a given environment need not be an all-or-nothing choice.

This article, “iSCSI learns new tricks,” originally appeared at InfoWorld.com. Read more of Matt Prigge’s Information Overload blog and follow the latest developments in storage at InfoWorld.com. For the latest business technology news, follow InfoWorld.com on Twitter.