Sunday, September 24, 2006

Converging on the converged network


Several bits of news coming down today (Monday, Sept. 25) all point to the ongoing effort to craft a Swiss Army Knife of data center networking. I’ll update this posting with links to full stories on each item as soon as they become available on eetimes.com.

The Storage Bridge Bay alliance has released its 1.0 spec and announced IBM and Network Appliance have signed on. The latter partially answers one blogger’s criticism back in March when SBB debuted that the effort only has partners Dell and EMC driving it.

What’s key here is the Dell/EMC duo and their new partners have essentially put out to the industry a core spec for a mainstream modular storage array that can be adapted to many uses. That will lower OEM costs and time-to-market while providing some added ease of use. Time will tell by how much that will raise the heat on the big competitors still doing their own thing—Hewlett-Packard and Hitachi Data Systems.

And there’s plenty more shoes to fall. As I said in postings earlier this month both IBM and NEC have new proprietary products for modular storage arrays in the works.

Separately the drive to merge networking, clustering and storage on to Ethernet takes three small steps forward today. HP announced it will use the 10Gbit chip from NetXen. Chelsio rolled out its third-generation architecture and NetEffect said it will show at the Intel Developer Forum this week its chip running the Open Fabric Alliance software for accelerated Ethernet.

This accelerated Ethernet movement has needed some forward progress after a few years of slogging out specs and shifting from Gbit to 10Gbit as the market target. As Dwight Barron, an HP server technologist told me, 2007 will be the year when the rubber finally meets the road on converged Ethernet.

It is not without competition. Today the Infiniband Trade Association announces it is supporting iSCSI. IB is already getting significant traction in clustering, and now the push is on in storage. Ethernet backers say their 10Gbit parts will mow over reigning Fibre Channel which is at 4Gbit, and IB says look out for its 20Gbit parts just starting to hit the market. Their Achilles’ heel: The IB software is still a long way from being as mature as that of the competing interconnects.

The lesson? Convergence in an open market is a messy free-for-all. --rbm

2 comments:

Anonymous said...

Storage Bay alliance is not about innovation but about establishing control points to slow innovation. Customers may save some short-term dollars but in the end, they will pay a higher price.

IBM joining this effort is no different than their blade strategy. Their product mix combined with a service-centric revenue stream cannot support their hardware cost structures. They, like EMC, must shift the cost of innovation to others and rally around the banner of open standards because they can't compete at the necessary level of innovation.

As with blades, once people examine the power and thermal issues, the need to improve the management integration, etc. etc. people find that these open standards are so incomplete that aside from providing a bit of marketing benefit, the adopter ends up spending considerable dollars to fill in the gaps and even more adjusting for the endless compromises and slow standardization process.

Innovation should be open and standardized where it makes sense. But for now, these standards are so deficient from a solution's point of view to be laughable. So, please everyone, rush to standards like the Storage Bay but keep in mind most customers have more important things on their minds than just an enclosure and a small fraction of the management ecosystem. That is what keeps customers up at night fretting about how to do more with less.

Anonymous said...

The IBTA supports iSCSI follows the iSCSI that already exists on Ethernet. Further, it is a recognition that SRP while a reasonble basic data mover simply lacked the rest of the technology definition to deliver a real solution sans acting as the wire protocol for Fibre Channel over IB.

Fibre Channel is moving to 8 Gbps technoloogy which many believe may be its last volume speed bump. It is cost effective and will meet SAN storage needs through 2012 quite handily. The question then becomes will the storage providers implement enough of the iSCSI specification suite to enable the glacial-moving enterprise customer base to migrate to either Ethernet or InfiniBand.

Some companies will not wait as they are moving to InfiniBand attached storage. Conceptually a nice story but one fraught with many issues - questionable software quality / robustness, incomplete InfiniBand storage management, even incomplete InfiniBand management - can't seem to get a robust QoS or multi-path software package to market as yet, and so forth. InfiniBand holds a lot of promise but it isn't a panacea no matter how much its proponents praise their efforts (talk about marketing at its finest given how small the real dollars are being spent on InfiniBand in production environments outside of HPC customers).

Ethernet by no means is a clear winner either. It lacks the multi-path and QoS to translate the converged fabric promise into a reality. The IEEE has been busy filling in the gaps but it will take some time just like InfiniBand to translate what amounts to just some NICS, some low-latency switches, a bit of management software, and a lot of marketing into a reality. What Ethernet has going for it is customer mind share and a large, established management suite that is much easier to evolve the glacial than force feed a transition.

As for InfiniBand having a speed advantage, it's all true. Unfortunately, driving an Indy car is cool but does little if governed by 55 MPH speed limit. InfiniBand proponents are filling in the functional gaps but the challenge to deliver an enterprise solution - simply ignore the marketing drivel about there being an enterprise distribution - is more daunting and will leave many customers with a bad taste if positioned as anything more than a good starting evaluation point for the enterprise market.

 
interconnects