Xsigo Server Fabric from Jon Toor on Vimeo.

Unified computing solutions are now available from multiple vendors. Who will prevail? Download this thoroughly researched 23-page report from InformationWeek Analytics that exhaustively compares solutions from HP, IBM, Cisco, and Dell/Xsigo.

It’s an interesting read.

InformationWeek offers it only on their pay-site ($99 to join), but we’re making this report titled, “Unified Computing: Who Will Win the Stack Wars,” available to you for a limited time free of charge.

Register for the report here.

The central question is this: “The battle lines are drawn, but do the benefits of a prebuilt system outweigh the drawbacks of subscribing to a single vendor’s technology vision and product roadmap?”

The report looks at this question — and the question of why unified computing at all — in detail.

Here are a few highlights:

1) “The next big thing in unified computing is the Virtualization of Everything: data and storage networks, block and file storage, even network appliances.”

Needless to say, we would agree completely. In deployment after deployment, we’ve seen that an end-to-end approach — virtualizing the servers and the infrastructure — makes the server virtualization project significantly more effective.

2) “The ultimate goal for the next-generation data center is a cloud-like environment where each application can be dynamically provisioned, configured, orchestrated and decommissioned using set resource profiles.”

Once you have achieved step 1 (the “omni-virtualized data center”) you’re significantly closer to a cloud-like environment. With that end-to-end approach I can:

• Dynamically connect any network or storage to any server
• Link resources without regard for physical limitations such as cards cables and switch ports
• Deliver whatever number of connections and bandwidth each server needs

Those attributes put me a closer to that fully flexible cloud of interchangeable assets.

3) “Bundled unified computing stacks occupy that uncomfortable position common to most all-in-one offerings: When striving to be all things to all people, you end up dissatisfying almost everyone.”

You can look at this point from two angles. One is that any bundled solution is likely to quickly diverge from its prix fixe origins, so some level of integration becomes a fact of life.

The second is the open solutions such as Dell/Xsigo’s have an immediate appeal. They’re designed with simplicity and interoperability in mind, with use of widely-tested, well-known components. Instead of trying to be all-in-one, they strive for easy integration.

Other interesting discussions in the article raise questions such as:

• Of the technologies in the stacks, which are IT managers most likely to deploy in the next 12 to 24 months?
• Why the fixation on blades?
• For storage in these stacks,  what are the pros/cons of FC and iSCSI?

Register for the report here.

Enjoy!

The dark side of technology is obsolescence. Technology evolves constantly, leaving a swath of behind-the-times gear in its wake. This problem becomes especially vexing as we transition to private cloud architectures where we’d like all resources, both new and old, to be interchangeable. If I can’t connect my new network to my older servers, those servers are no longer interchangeable.

In a recent Virtualization Review column, Elias Khnaser likens virtual I/O to an insurance policy. Elias states, “Virtual I/O allows you buy a sort of insurance policy against any type of future fabric that you might introduce later in the network. Xsigo has an appliance that you install between the servers and all the different types of fabrics available. Xsigo will accept all these fabrics as inputs and will then output them in a standard format, which means if a new technology is developed tomorrow, you can simply virtualize that behind the Xsigo appliance and maintain the same cabling and the same adapter in your servers.”

Why This Matters

A cloud infrastructure by definition relies on flexible connectivity. The whole point is to make server resources interchangeable so you have the flexibility to move workloads around, scale your environment in response to changing demands, and boost utilization by minimizing wasted server time.

But this relies on the ability for any server to connect to any network or storage when needed. (Conversely, it also requires that connectivity can be removed from a server when it would introduce compliance or security concerns.)

The obvious problem is this: what happens when new networks or storage are added? Are they connected to all of my servers or only to some of them? There are lots of reasons why servers might get excluded. They might lack the number of ports needed or types of ports. Or it might be difficult to justify the expense of connecting them. Or there could simply be schedule pressures.

Whatever the reason, when servers get left out, they are no longer fully interchangeable resources. You’ve just destroyed your “cloud.”

Virtual I/O Provides Universal Connectivity

Virtual I/O alleviates that issue in two ways:

1) I/O modules can be added as needed: The I/O modules in the I/O Director effectively become the I/O for the servers. They replace the Ethernet and FC ports that would otherwise be found in the servers themselves. When a new type of I/O is needed (such as 8G FC or 10G Ethernet), you simply add that module to the I/O Director. That I/O is immediately accessible to all attached servers. The same is true if you need new ports for new isolated networks (such as VMware vMotion or FT networks). Or if you simply need more uplink bandwidth.

2) Connectivity isolation: Each network and storage port on the Xsigo I/O Director is an isolated connection. When a virtual NIC is assigned to a port, that vNIC alone has access to that network, exactly as if you had pulled a cable between that server and that network. (Of course you can also assign multiple vNICs across multiple servers to a single port.)

As a result, this “insurance policy” works two ways. It ensures that the I/O technology is evergreen. You can always add new physical I/O when needed. And it ensures that any server can be connected at any time.

This combination removes several significant nightmares from the data center planning process and from the ongoing maintenance/updating process. And isn’t getting a better night’s sleep the whole point of insurance?

VMworld is all about the cloud. And what you’ll find in that cloud is Xsigo.

Behind the scenes at VMworld, now happening in San Francisco, there are 14 Xsigo I/O Directors connected to hundreds of physical hosts. This gear, which is running hands-on labs, booth demos, and classes, is located at several locations around the country. I’m told that one is on the east coast. The idea of course is to help validate the cloud concept. You don’t care where the server is, just that a server is always available and accessible.

Click to the next page for a look behind the scenes. (more…)