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: Joe Wagner who wrote (13784)	1/21/1998 10:43:00 PM
From: Eleder2020	Read Replies (1) \| Respond to of 29386

The Bollinger Bands only this narrow 4 other times in the last 6 months.The trend was up twice and down twice after this narrowing. The up trends were over 50% and the down trends were about 50%.If the Bands continue to narrow in the next four or five trading days and if recent history holds the price of Ancor should be on the move in 5 to 8 trading days. Alan Aranoff- How far off the mark am I?Seems like bouncing off the lower band would be nice? No?Good at looking at it,but could use some help in the interpretation. Doing this for entertainment while ANCR sitting. Thanks , Ed

To: Joe Wagner who wrote (13784)	1/21/1998 11:15:00 PM
From: Eleder2020	Respond to of 29386

Came across some recent back issues of VAR Business and sure enough their was a section where they ranked VAR's and their was Litton in the number 8 spot for Top 10 Vertical Market VAR's. In the top 10 technologies where a VAR listed technologies as one of their primary technologies Vertical applications ranked 3rd behind LAN/WAN and client-server.Internet and Unix were 4 and 5. Top 4 of the top 10 vertical markets 1)Manufacturing 2)Governement 3)Health care 4)Finance Ed

To: Joe Wagner who wrote (13784)	1/21/1998 11:35:00 PM
From: George Dawson	Read Replies (1) \| Respond to of 29386

Joe, My read of the article is that the I/O system software rather than the processor speed is the limiting factor. A good reference on this is "Understanding I/O Subsystems" by W. David Schwaderer and Andrew W. Wilson. Chapter 10 is called "PC I/O System Software" and it reviews the change in performance between 16 bit and 32 bit (Windows 95 and NT) applications - 40 - 50% faster. From your link you can see the approach by various GE switch makers is to make their own drivers. Schwaderer and Wilson give an example of how drivers are layered in Windows NT, and how additional drivers need to be added to enhance functionality (figure 10-5). You can get their book from Adaptec. It was a real eye opener for me, since I spent the summer thinking about latency in the hardware. Maybe some of the engineers on the thread can comment on the software factor. FC has to connect to storage servers in the SAN. Two closing thoughts: 1. GE and FC have the same physical layer. 2. FC has documented speeds greater than the speeds given in these articles, and in big systems the bus rather than the software has been the problem. George D.

To: Joe Wagner who wrote (13784)	1/22/1998 1:09:00 AM
From: Kerry Lee	Read Replies (2) \| Respond to of 29386

I think Craig and others are more qualified to answer your question from a technical point of view, however, I'll give my 2 cents anyways: I think that article refers mainly to Windows NT being the bottleneck. Also, I don't believe that FC actually "bypasses" the servers when in fact it directly connects many servers, disk arrays and nodes ( eg, workstations ) as opposed to utilizing routers ( I'm treading on very soft ground here, someone feel free to correct me )... There is too much Vendor/Systems integrator support and marketing muscle behind GE for it to fail. I personally don't think anything can/will stop Gigabit Ethernet from taking over the LAN, with the exceptions where QOS ( ATM, FC Class 1) and longer distances ( beyond 500 meters ) are requirements for the network. Also GE is specified at 1250 mbps vs FC at 1062 mbps, therefore theoretically a "Gigabit" of GE actually has a true Gig ( 1000 mbps ) of actual throughput vs FC at approx 800 mbps throughput. Can someone comment on the how the scaleability of FC compares to GE?

To: Joe Wagner who wrote (13784)	1/22/1998 9:14:00 AM
From: Craig Stevenson	Respond to of 29386

Joe, One of the major hurdles that Gigabit Ethernet (or Fibre Channel) must overcome are bandwidth limitations in the hardware and within the operating systems themselves. The 64-Bit PCI and I2O initiatives will help from a hardware standpoint, and operating systems will undoubtedly improve to support the higher bandwidth requirements of gigabit adapters. This quote from the article link you posted was especially interesting, given the recent ANCR trashing by a Microsoft employee. <g> <<Windows NT is about 800 megabits shy of a gigabit.>> Having taken a cheap shot at Ed <g>, I will say that Microsoft will attempt to address this issue in Windows NT 5.0. Microsoft will not concede the enterprise to Unix or NetWare. Craig

To: Joe Wagner who wrote (13784)	1/22/1998 10:05:00 AM
From: Craig Stevenson	Read Replies (1) \| Respond to of 29386

Joe, I neglected to respond to your SAN question. I think that Storage Area Networks and Networked Attached Storage will benefit Fibre Channel, but from a connectivity standpoint, more than from a performance standpoint. Basic connectivity only requires a loop, and/or hubs. This is the reason for the proliferation of hub companies, such as Gadzoox, Vixel, and G2. High-performance applications will demand switches, and presently that is done using servers. (That will undoubtedly change, but it will take some time.) A search of the internet for companies working on Network Attached Storage units reveals a surprising number. Many are implementing Fibre Channel now, or will be soon. If you haven't already, take a look at the "A Fibre Channel Future" thread. Neil has posted some interesting press releases from Emulex, QLogic, DEC, and others that detail how successful their Fibre Channel products have been and what their future strategies will be. Of particular interest to me was the QLogic release that talked about the strength of Fibre Channel in the Asia-Pacific region. (With the recent economic disaster over there, I wouldn't have expected that.) Craig