home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky vmsnet.alpha:128 comp.sys.dec:7063
- Newsgroups: vmsnet.alpha,comp.sys.dec
- Path: sparky!uunet!bcstec!silverm
- From: silverm@bcstec.ca.boeing.com (Jeff Silverman)
- Subject: Re: No 64-bit OpenVMS soon?
- Message-ID: <C18sLv.JwM@bcstec.ca.boeing.com>
- Organization: Boeing Computer Services
- References: <1993Jan15.194312.3438@sol.UVic.CA> <1993Jan18.172458.24506@eco.twg.com> <1993Jan18.211506.27601@oracle.pnl.gov> <1993Jan18.234713.1225@cmkrnl.com>
- Date: Fri, 22 Jan 1993 06:27:30 GMT
- Lines: 17
-
- Actually, I was disappointed to read that the Alpha is never going to support
- true 64 bit addressing. Depending on the page size the most you can expect
- is 55 bits.
- Why does this make a difference? I for one, was excited about the possibility
- of building an in-RAM relational database. I had been doing some calculations
- on the subject and some design studies for a sharable multiport RAM which
- would hold the database. Like a solid state disk, it would be fast, with
- average access times of on the order of microseconds. Unlike a solid state
- disk, which looks like a disk drive, this gizmo could be accessed without
- intervention (or with minimal intervention) from the OS.
- For another, I was intrigued with the idea of building a minimal microprocessor
- system using the address lines in lieu of address decoding for chip select logic.
- Several people have pointed out that anybody interested in building something
- around an Alpha obviously isn't building a minimal microprocessor.
-
- Jeff Silverman, Boeing Commercial Airplanes.
-
-