SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Strategies & Market Trends : Gorilla and King Portfolio Candidates -- Ignore unavailable to you. Want to Upgrade?


To: buck who wrote (17321)2/6/2000 12:43:00 PM
From: Uncle Frank  Read Replies (1) | Respond to of 54805
 
>> buck, who feels quite inadequate at this point...

Self image is rarely accurate. Karen Carpenter felt quite fat.

From an outside observer's point of view, change your description to,

Buck, who has impressed the hell out of the G&K thread, and will surely make us all a lot of money.

uf



To: buck who wrote (17321)2/6/2000 1:43:00 PM
From: buck  Read Replies (5) | Respond to of 54805
 
Buck's FC - Part B

Finally, to the router side of it. I've broken CRDS out, because I believe that it has the most potential to be a gorilla, or at least a mini-gorilla in a mini-tornado. Hence, I've put more time, thought and research into it. I hope that's apparent.

CRDS has a proprietary open architecture for FC routing (assuming I understand POA correctly). The current product set (in it's third generation) connects a SAN to SCSI devices, typically tape drives, but not limited to them. They also connect SCSI hosts to SANs. These hosts are either a) too old to have FC HBAs available, or b) the customer does not want to install FC HBAs in them (costly downtime, expensive OS upgrades, etc.). This host connection is a CRDS exclusive today. CRDS has demoed the server-free backup capability mentioned in Part I at Comdex and N+I, for the last two years.

CRDS also has announced (and demoed at Comdex99) a router that moves FC data across an ATM WAN or LAN link. This allows the SAN to be extended beyond the 10KM limit of pure FC. Initial applications will probably remote electronic vaulting for disaster recovery and data mirroring for remote-site data sharing.

CRDS is also involved in the InfiniBand game. Intel invested in them for IB routers, to attach IB-enabled servers to external non-IB devices and networks. The proof-of-concept router was demoed at the Intel Developers Forum in August of 99.
These things -- FC<>SCSI, FC<>ATM, IB<>who_knows, and server-free backup -- represent to me a strong proprietary open architecture.

CRDS had been an OEM-only mfr. until 99. They have announced OEM agreements with ADIC, ATL, Bull, CPQ, DELL, Exabyte, Fujitsu/Siemens, HP, Inrange, McDATA, and STK. They have since opened up channel sales agreements with a number of VARs and distributors. CRDS has also announced partnerships with software cos. Legato, Tivoli, Veritas, and Computer Associates. These ISVs will provide enabling applications that take advantage of server-free backup, as well as other data movement applications. They will also provide management capabilities for the SAN that do not exist today. These partnerships provide the beginnings of a strong value chain.

CRDS' barriers to entry seem to be low-ish to medium. SCSI is a standard, FC is a standard, ATM is a standard, and IB is a standard. However, the internal software is what seems key in this space. Server-free backup is done with this internal software. If you follow the Cisco model of routers, other applications can be added here, such as access control lists. SAN-specific apps could include data mirroring and snapshot copies of databases. Their follow-on products, ATM and IB, are sure to include some elements of these, as well.

My question is how do we value this firmware, and it's capabilities? I'm still learning how to do that.

Additionally, CRDS has a large list of interoperable configurations that they have made public. They don't guarantee these configurations, but they do state that they have verified interoperability with a huge amount of software and other SAN components. This amount of interoperability testing and verification seems to be BTE, but I'm not sure how it might be valued.

I believe that there are medium switching costs for end-user customers, when they get them. Since the value chain(?) is so long -- application, platform, HBA, fabric, router, tape library/RAID device -- I think it will be "not simple" to change from one router to another, because of the application support. CRDS' advantage here is being the firstest with the mostest. How do you value the importance of one router over another if you are someone like a Legato or or Tivoli or Veritas or CA?

CRDS has three competitors at this time, none of whom are publicly traded:

ATTO Technologies: produce a storage router; have a strong channel; well-positioned in the Apple market. They have one announced tier-three OEM, MicroNet.
attotech.com

Chapparal Technologies: an Adaptec spin-off; builds FC RAID controllers that they have adapted to storage routing. They have one announced tier-two OEM, Qualstar.
chaparraltech.com

Pathlight: newer company; directly competing with CRDS, as they have no other products except storage routers. They have one tier-one OEM (IBM!!!), and two tier-three OEMs (Glyph, Aviv), both of which are niche-market (bowling alley?) suppliers.
pathlight.com

Bottom Line

Crossroads can be a gorilla inside the FC tornado, according to my notes:
- They have a POA in their multi-protocol routing software.
- They have created new markets, in FC<>SCSI, FC<>ATM, and server-free data movement.
- They have an extensive value chain in their OEMs and application software partners. It seems to be strengthened by their interoperability testing.
- They have medium barriers to entry, in that the software in their box is relatively complex, and I/O engineers aren't growing on trees (they all want to do B2B Java, instead of dull, old I/O<g>) Again, interoperability testing COULD be a BTE.
- They have a product that is "worth more" because of their extensive interoperability testing, and configuration verification, and their announced partnerships.
- Switching costs are an unknown today, and the biggest one that would preclude gorilla-dom. Will the application providers (Tivoli, et al) get to market fast enough to make CRDS a de-facto standard with whatever CRDS enables? Will that "standard" survive assaults from others? Will others have de-facto standards that supersede CRDS' standards?
- They are first-movers (not a GG criteria, I know, but one I respect).
They can be found at crossroads.com Skip the intro unless you just love Flash animation and the sounds of a throbbing internal combustion engine.

Important Note: Absolutely none of this takes into account last week's announcement of their acquisition of Polaris Communications, a mainframe-focused company. Part of me says why bother, another part says "ENTERPRISE!" I dunno...must digest first.

So, there you have it. Shoot if full of holes. I know the financial side is missing, and I need to know what is important, particularly for these little bitty newly public cos. FWIW, I'm re-reading a LOT of TRFM. Tell me what else I'm missing, and I'll go get it. The lovely Mrs. Buck is out house-hunting, I've got both beer and cigarettes, and the NHL All-Star game is on later, so I've got the makings of a great Sunday afternoon (too bad it's not baseball season yet.)

buck, who feels like one of those ice skaters waiting for the judges to post their scores

PS Dang, this seems long!



To: buck who wrote (17321)2/6/2000 3:16:00 PM
From: freeus  Respond to of 54805
 
re Ancr and others
Thank you.
If you feel inadequate writing it, you can't imagine how a want-to-understand-technology writer/actress/teacher of small children feels reading it and attempting to know what she's reading!
I'll keep trying though!
Thanks again for your hard work.
Freeus



To: buck who wrote (17321)2/6/2000 3:22:00 PM
From: Greg Hull  Read Replies (1) | Respond to of 54805
 
RE: Fibre Channel

Buck,

Let me second UF's comment: I'm very impressed with your FC overview. Very well written and very informative. Since you are an industry insider I am interested in your thoughts on this point:

I do believe that the FC fabric market will produce either a Gorilla or a Very Strong King. Am I correct to infer that you believe Brocade can/will be a Gorilla in Fibre Channel switching? I realize that Brocade is the subject of another PH report, but I'm interested in your thoughts on the matter. Why would the Big Boys (EMC, IBM, Compaq, SUN) of storage tolerate the presence of a Gorilla in FC switching when they have it in their power to squash the possibility now? What is it about Brocade's products that would cause the storage vendors to fall in line to support their Gorilla status?

Thanks for taking the time to educate the thread on FC. It's also nice to find another hockey fan.

Greg



To: buck who wrote (17321)2/6/2000 10:25:00 PM
From: James Burke  Respond to of 54805
 
Buck,

I was fortunate to get into Ancor (ANCR) last year when it was low single digits. It rode up to 95 and has since settled in the 40s. I'm still holding and expect them to be a great performer during the next 2-3 years while the SAN market explodes. ANCR is not for the faint of heart. It's a very volatile stock.

Yes, ANCR has won SUN but what is bigger in my eyes is INTEL's recent investment in ANCR for future Infiniband development. This $14 million investment was at a price of $58. There are also strong indications (last week's earnings cc) that new OEMs will be announced soon. It is of the opinion of the SI and Yahoo ANCR boards that one of the OEMs will be EMC.

I strongly suggest any one who is considering an investment in ANCR read the SI ANCR board. Also check out ANCR's home page and the following site goancr.com

Though many believe Brocade is the Gorilla in the switch market, I think ANCR is where the real potential is. I believe many companies will eventually dual source their switches so I'm not sure either will be a true Gorilla. But if you take into account the INTEL investment, an investment in a relatively small company for something as important as the next generation IO Infiniband, which has backing from all the major tech companies, I think ANCR is well suited for aggressive investors.

BTW- Fidelity has recently upped their stake in ANCR to 11%.
I also hold CRDS. Their sales are exploding and I think they could eventually get bought out by some one like CSCO (just my opinion).

Best of Luck to all.

JB



To: buck who wrote (17321)2/7/2000 8:54:00 AM
From: DownSouth  Respond to of 54805
 
Buck, you have made a very important contribution. The whole FC/SAN set of concepts and relationships is much clearer to me now. Still a huge amount to learn. And for you to put it all in ggamer terminology is a real bonus.

Inadequate is not a word I would use to describe that post. "Mind expanding" is more appropriate.

Thank you.