Hot answers tagged

28

Given that the Cat2960-S is a desktop/access switch, with very, very small buffers, you would likely experience a lot of output drops. A datacenter switch, such as a 4948E, would be a better choice for an iSCSI application. To understand the reasoning behind this, you have to remember that an ethernet switch is either transmitting, or not transmitting on ...


14

http://packetlife.net/blog/2012/jan/16/catalyst-2960s-iscsi-optimization/ I followed Jeremy Stretch's post here a while back, moving our iSCSI SAN off a production switch onto a dedicated 3750G (stack of 2). While I can't speak specifically to the 2960S, the tweaks offered in that post seem to be doing well on the 3750G's. I've got a handful of ...


11

In a word: deep buffers, preferably per-port buffers iSCSI does not tolerate frame drops. Delays are sort of ok, but a switch dropping a frame will create all manner of issues. (Finding buffer size -- and type -- in manufacturer specs can be a challenge. They aren't always provided, and what is provided isn't always true.)


8

Make sure the switch is non-blocking. That is, the switch backplane can support all ports at line rate with no oversubscription. For a 24-port switch, the minimum backplane bandwidth is 24 x 10G, or 240Gb.


7

I've had experience where customers have chosen to use the lower end switches (3750, 2960) for iSCSI and they are pretty terrible. The issue gets even worse when you have your storage connected at a higher speed (like 10gbit for the SAN and 1gbit for the end servers). I would look towards more of the DC class switches, if you want to stick with Catalyst ...


6

"Switched fabric" (SW) is mainly used for Fibre Channel networks as its contrasts the other FC topologies point-to-point (P2P) and arbitrated loop (AL). The point in SW being that there are multiple, active paths between two nodes. Classic Ethernet requires a tree topology - however, the concept of a fabric isn't uncommon with Ethernet, especially when used ...


5

The Best Practice Guide for HP Lefthand iSCSI Storage stated, "For optimal switch performance, its recommended that the switch have at least 512 KB of buffer cache per port. Consult your switch manufacturer specifications for the total buffer cache. For example, if the switch has 48x 1 Gb ports, the recommendation is to have at least 24 MB of buffer cache ...


4

Be sure to understand what your iSCSI implementation on the end systems has to say about LACP (Port Aggregation) and/or if it mandates path isolation, to have two completely independent network paths between any initiator and any target. If either is true, then MLAG (Multichassis Link Aggregation) capabilities probably won't be required.


3

Setting the switch to use MTU of 9k globally will not be an issue for host device interfaces that are programmed to use 1500 as a maximum. The host will compile frames that have an MTU size of 1500 maximum and while the switch accepts a maximum of 9000, there will be no fragmentation and it will accept the frames without discarding them. An issue would only ...


3

To understand the terms you have to understand some history. A "host bus adaptor" was originally something used to connect a SCSI storage bus to a host computer. You could in theory have multiple computers on the same SCSI bus but limits on total devices made it impractical to have more than a couple. The networking world used the term "Network Interface ...


3

In complement to Ron's answer, you need to look for Jumbo Frame. That is not mandatory, but it will significantly improve the bandwidth. A good point too is the latency. It should ideally be below the µs. You may also go for a full SFP+ switch, because depending on the equipment, you will have a mix of interfaces.


3

IOPs depend mainly on throughput capabilities and latencies of your storage hardware and (to a much lesser extent) latencies in your network. In other words, you can't "translate" a certain infrastructure to IOPs (you can make an educated guess though). However, you need to design your SAN to manage the bandwidth you require. Once you exceed the bandwidth ...


2

N-port is a node port, actually it's HBA on a server or a port on a storage array. It is used to connect a node to a Fibre Channel switch. On Brocade switch you would usually have F-port, which is "Fabric port" used to connect with node port. In rare situations you can get L- or FL- port on the switch when you have to deal with mostly deprecated FC-Loop ...


2

A SAN appears as a raw block device and can do the things that a block device can but a filesystem can't. Two very common cases include data storage where an application is using the block device directly (as some databases can), or to serve as a boot volume for a diskless server (some servers and blade systems can speak iSCSI from firmware well enough to ...


2

Source/Destination addresses here are 24 bit FC Port addresses which consist of 3 bytes: Domain ID, Area ID and Node ID. Details you can find, for example, here: https://www.storagefreak.net/2014/10/fibre-channel-addressing


1

2950's are too slow, and their buffers way too small for iSCSI. 2960's will work in a pinch, but their buffers are really small as well. (I've run iSCSI SANs on 2960S's. For low traffic levels it will work; at higher levels frame drops will become a nightmare.)


1

Point-2-point mode is the opposite of arbitrated-loop mode (AL). You usually do not want AL if there's a switch. Set all ports to P2P at all times unless you really do need AL mode (when you chain multiple devices on a single port). In AL mode, ports pass on traffic to other ports behind them in the loop. At any time, there can only two ports active in a ...


1

I don't believe this is possible, although Juniper's documentation is very vague. It is my understanding that in order to do FCoE, the EX4550 must be a stand-alone device, not part of a Virtual-Chassis. There is vague reference in the EX4600 doco that says: Note: Standalone switches support FCoE. Virtual Chassis (VC) and mixed- mode Virtual Chassis ...


1

A NAS is a (possibly) simple storage device, featuring various protocols for storage access, often including iSCSI. There is an extremely wide range of different devices, features and performance grades. Most NASes provide disk redundancy, some power and network, few controller redundancy. A SAN is a network between storage devices and hosts, providing and ...


Only top voted, non community-wiki answers of a minimum length are eligible