It compresses (not just reading) about 1G GDS data per minute into our external memory system (this is simply snapshot of internal memory to hard drive and reliability is same as internal memory system). This one time reading is machine and data dependent (4200RPM notebook hard drive will be much slower than 7200RPM normal desktop hard drive, SATA2 will be faster than IDE, etc).
OASIS file takes about a minute per 200-250M in the first time reading.
Once OASIS/GDS file is read into our external memory system, loading and displaying library from external memory system is instant, and hundred users can access the same library at the same time. And it only takes a few seconds, plus loading time to physical memory, for each display. Even full chip detailed display including 1nm geometry takes a few seconds on less than 1G GDS which is over 15X faster than others. On some complete chip with 100 layers, we observed over 120X speed up in full chip 0 pixel detailed display.
You can carry our external memory system files instead, and convert or extract portion to GDS/OASIS any time. Entire external memory system to GDS takes about a minute per 2G. Run time for external memory system to OASIS is the fastest because it is OASIS compliant. It outputs original 1G GDS to OASIS in a few seconds.
We beat anyone in the display speed of detailed full chip view and we welcome benchmarking. |