Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Back then we had four primary rate connections 2.048MBit/s each (30 ISDN B channel) to transmit data to several subsidiaries. This was quite a lot of bandwidth at the time. Our litte on demand network started at 02:00 in the night when tariff was lower (paid by connection time, not by volume :D) and used Ascend and Cisco routers Ascend was always much quicker setting up MPPP channel and therefor preferred, but Ciscos where a bit better utilizing larger 30 channel MPPP connections (and some destinations preferred Cisco)

It was quite 'fun', because starting with morning when tariff changed everything had to be transmitted and if something didn't work out right (to slow buildup of the MPPP channel, delay in the batch jobs delivering the data) someone had to monitor the process in the next night to prevent further backlog. For a long time there was only a 64kBit (later 128kBit) connection over the day regular connectivity.

Eight MBit carefully distributed over 120 channel. With all the support equipment a server rack, filled by 2/3. Ridiculously little today, even over air. It sounds as if one talks about the ancient Egyptians, but it is only 25 ago



I thought primary rate interfaces were 24 channels (23 bearer channels plus one data channel) for a total throughput rate of 1.5 Mb/s, but apparently that’s only the case in the US, Canada, and Japan. European PRIs (E1 vs T1) are as you described. TIL.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: