Silicon Investor (SI) -- The First Internet Community

STOCKTALK

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.

Technology Stocks : The New QLogic (ANCR) -- Ignore unavailable to you. Want to Upgrade?

To: KJ. Moy who wrote (25202)	12/12/1999 2:33:00 AM
From: Kerry Lee	Read Replies (1) \| Respond to of 29386

For those who can't be bothered to sort thru the clutter/profanity on Yahoo message board , here's a worthwhile repost: Inifiband based networking gear by: viangio 12/11/99 11:08 pm Msg: 22009 of 22016 The LAN networking types are there because Infiniband is "Infinite bandwidth" for I/O. It's not just I/O bus replacement in servers. It's classic bus replacement anywhere. The IP router folks all have big wide buses in their IP routers. The Inifiniband idea is throw the bus away and replace it with a fabric. The fabric can scale as much as you want simply by adding links and little switching chips. So in a big IP router for example you could have line cards that externalise to standard interfaces such as Ethernet, ATM etc. But internally, the LAN packet router is sending packets over a fabric of Infiniband links. If I want to increase the capacity of my IP router I can add more line cards while also scaling the internal fabric to match. I cannot do that with today's LAN routers and the reason is the internal bus won't scale. Cisco and others have all played with fabric based router architectures at one time or another. They are common in parallel super computers. But now, the Inifiniband effort is set to make this a high volume off the shelf technology. I'll be able to build application specific hardware - whether it be a new kind of server (CPUs, memory and disks) or a new kind of IP router (one CPU and loads of LAN interface line cards) or even a big graphics engine (lots of pixel renderers and distributed frame buffers). So the answer is Nortel, Lucent, 3Com and Cisco are all there because Infiniband is an architecture that allows them to innovate, design and build next generation communication equipment. Dell, Compaq, HP, IBM and Sun are all there because Infiniband is an architecture that allows them to build application specific scalable servers for less than today's models. Infiniband is a totally revolutionary architecture. It's been coming for a long time and a number of prior experiments had to fail to get this far. Classic servers and what they attach to will change in drastic ways. One can imagine for example; an Inifiniband based server and LAN IP router in the same cabinet - with all the neurons attached to the fabric. Hardware for an integrated application engine - where the application is serving Web pages, video feeds, voice mail, eCommerce credit card transactions etc. etc. It's an architecture for a totally connected planet. Phew. The Ancor deal cannot be understated - Ancor's switch ASICs could show up in Cisco equipment. Can you imagine.... BTW, business school CEOs don't get this. Technology leaders do. Maybe this explains some of the rhetoric lately. Posted as a reply to: Msg 22005 by jfieb View Replies to this Message

To: KJ. Moy who wrote (25202)	12/13/1999 9:48:00 AM
From: The Phoenix	Respond to of 29386

KJ, I've not completed my due diligence. But the phyical limitations of PCI are being solved on the WAN using native interfaces. Now I do realize that the WAN characteristics are altogether different. Generally latencies are higher...but what I'm finding difficult to rationalize - and perhaps I haven't dug deep enough yet - is why latency is an important factor in the SAN....even more so than in the WAN it would appear. Seems to me that the bigger issue should be time to move a file. If that's the case 10G (OC192) would seem to be a good alternative. What am I missing? OG

To: KJ. Moy who wrote (25202)	12/20/1999 3:46:00 PM
From: Greg Hull	Read Replies (2) \| Respond to of 29386

KJ, On yahoo (http://messages.yahoo.com/bbs?action=m&board=4687096&tid=ancr&sid=4687096&mid=22785) you posted: "Ancor's closing price on 12/31/99 was $4 if I am not mistaken. I don't have the data with me at the moment. At 1,891%, Ancor's price was about 75.75. Does anyone know what PUMA's closing price was at 12/31/99? Ancor may end up #1 yet at 12/31/99. That would be awesome. " I show that PUMA closed at 3 3/8 on 12/31/98 and ANCR closed at 4. As of today they are both about 84. Neck-n-neck going into the finish line. More importantly, I wonder how the top performing stocks of 1998 did in 1999. How many of them fell into an abyss? How many still had stellar returns? I'll see if I can get the list. Greg