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 : Concurrent Computer (CCUR) -- Ignore unavailable to you. Want to Upgrade?

To: jeffbas who wrote (5915)	12/6/1998 9:03:00 AM
From: Nimbus	Respond to of 21143

Trade-offs The cost per stream is heavily driven by the hardware in most cases. The way you get cost out of hardware is to use as much commodity stuff as possible, and "simple" stuff at that. This is not a stupid approach ... and it may have been the key to get an affordable enough equation to allow a mass market to emerge where it might have been 2 more years otherwise ... DIVA and NCUBE have custom hardware, and use higher-end disk schemes. SEAC and CCUR buy commodity hardware, .. workstation class. The problem with the commodity approach is that there is only so much you can do to make it "GREAT" in combination, where DIVA and NCUBE designed in features that commodity items will never have, like on-board battery-backed memory, or Reed-Solomon Error Correcting busses. These custom features are critical to providing years of uninterrupted IVOD service (never an outage), which in some cases (they hope) will be a major descriminator in the selection process. As for the today's MediaHawk, I can't go into a lot of detail here, but in order to get the cost down IN MY VIEW, the MediaHawk design traded-off the following major items in decreasing level of criticality: 1. Fault Tolerance 2. Scalability; go above 1200 streams and it hurts big-time 3. # streams/file; There is a limit to number of streams per file (a little know issue). 4. Footprint 5. Latency 6. SAN Now before everyone jumps all over me with "CCUR SAYS THEIR MACHINE IS FAULT TOLERANT and scalable ! " ... just remember that there are DEGREES of all of these things. The DIVA trade-off list is significantly different than the CCUR list, and SEACs list is also different. That is the nature of competition. The real test will be when we get 1000 interactive users on one of these puppies, and see if it can run for a month without an outage or too many complaints. At $5/stream, each outage instantly costs the cable operator $6000 (he can't charge active viewers for cut-off PPV) and he loses about $3000 per hour of revenue while the server is down. It doesn't take many outages and the associated avalanche of complaints, to consider the more fault-tolerant offerings. The MH and SEAC machines will be under the microscope during the coming trials and we will see if any of these affordable solutions are reliable and scalable enough to roll-out. I hope so. Having been in several system trials in my career, I am only cautiously optimistic about the speed this will all happen, leading one day to the PRODUCTION of servers (not build of multiple trial systems). AGAIN ... this is one man's opinion. Do your own research.