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 : Novell (NOVL) dirt cheap, good buy? -- Ignore unavailable to you. Want to Upgrade?

To: Steve Bannister who wrote (26425)	4/2/1999 1:47:00 PM
From: Steve Bannister	Read Replies (1) \| Respond to of 42771

I just looked at the Caching Workshop agenda, and they are scheduled to discuss the bake-off results at 1 PST. That would be the very earliest that results are posted.

To: Steve Bannister who wrote (26425)	4/2/1999 2:31:00 PM
From: David O'Berry	Read Replies (1) \| Respond to of 42771

Agreed, it is the speed at which you can fill the pipe that determines the overall aggregate caching device speed. I have not looked at the specifications for evaluation yet but Novell has always been able to push more along the bus than NT. Solaris smokes along the bus about the same level if not a little better than Novell. I think the number were something like 800Mbps for Solaris, 725 for Novell, while NT was dragging ass in the 400 range. NT made some modifications that helped out but I do not remember if it was better with normal ethernet frame size of if they used jumbo frames to achieve the boost. -SIDEBAR-I have found that trunking is virtually as effective as gig but its just not as pretty. Has anyone had experience with I2O in the gig world and if so what are the effective performance boosts? Back to the bake-off question, the caching vendors would use gig if they thought they could fill the pipe. Currently, in caching appliances, it is usually unnecessary to use gig due to size of the pipe from the isp. As WAN bandwidth scales up you will see multiple connections bonded together to produce larger amounts of incoming and outgoing connections. It is not inconceivable to scale to those upper ranges but it is unlikely in the near future. My motto, in the WAN world, is usually more is better anyway. Also, if Novell uses this appliance on corporate intranets to offload overburdened Web servers from clients internally connected via 10/100 connections then the gig benefit will be noticed immediately. I will shut up now. Sorry to be so longwinded, David