|
Authored by: jesse on Monday, April 08 2013 @ 05:06 AM EDT |
http://en.wikipedia.org/wiki/Fibre_Channel
Even I remember arguments over HIPPI and Fibre channel.
All three patents appear to be the same, and just talking about different ways
to detect and correct errors.[ Reply to This | Parent | # ]
|
|
Authored by: JamesK on Monday, April 08 2013 @ 07:51 AM EDT |
"I was working on HP9000 systems with redundant everything.
Maybe there too I will remember the application and the
architecture. One of the applications was for a NOC [Network
Operation Center] to be used to monitor multiple external
clients. The other was probably a telecom company billing
system."
I wasn't involved with that system, but the telecom company I used to work for
had an HP system used in the Network Management Control Centre. I worked on a
Data General system, that sat beside it and controlled a huge display board
showing network status.
---
The following program contains immature subject matter.
Viewer discretion is advised.[ Reply to This | Parent | # ]
|
|
Authored by: Anonymous on Monday, April 08 2013 @ 09:02 AM EDT |
Perhaps you were thinking either of McData or QLogic, both
companies that were in the fiber channel switching business
quite a while, along with Brocade. Cisco got into it a bit
late, and I'm not sure when ATTO came along. QLogic
supplied components to other companies for their SAN units,
such as IBM, and to some of the switch manufacturers. And
they, like Brocade, have plenty of patents. McData probably
does, too.
What I'm seeing different in the patents does tend to
reflect the industry changes in switching. At first,
typical network "fabrics" as in the first patent ('662) were
composed of groups of switches, and each had its own
management system built into it. Error handling is done by
the switch to which the memory is attached.
In short order, management systems were developed. These
were often software packages on general purpose computers,
which managed the switches of the fabric, adjusting for
transmission problems. And that's exactly the kind of
language that shows up in the second patent ('177). Error
handling, including route management, is determined by the
managing system, not the local one.
These days, these management systems can be built right into
other switches. These switch managers were switches
themselves, but also able to update routing and other
parameters throughout the fabric. Error handling, including
route management, is determined by the managing switch.
('388)
It probably also is broad enough to be interpreted to
covering devices operating iSCSI (SCSI over IP), SoE (SCSI
over Ethernet), and SoFC (over Fibre Channel).
But the transmission technology is still broad enough to
allow this to cover newer technology as well, and it's just
peripheral. Claim 1 is an environment claim, imo. It's
describing the connection environment, what the network of
devices is capable of doing.
The real core of the patent is focused on a general
description of handling combined storage systems and
failures of those systems. The patent refers to CDAs, SSDs,
and other high-speed disks and storage devices, and how to
manage them under common program control to handle large
high-speed storage needs. Caching on the fastest devices,
It's about Storage Area Networks (SANs) and/or Network
Attached Storage (NAS) devices.
I believe companies like EMC, Hitachi, and IBM have the most
issues with these patents. EMC is the Cisco of the SAN
world. They've been in this business, and they have a lot
of technology related to these patents.
I also don't see well-defined algorithms or protocols for
handling these errors, just rough descriptions that could be
implemented in many ways. Very broad strokes. Perhaps so
broad as to be judged inadequate?[ Reply to This | Parent | # ]
|
|
Authored by: celtic_hackr on Tuesday, April 09 2013 @ 12:45 AM EDT |
PROM chips have used "switch fabric" or "crossbar switching" for a very long
time, also NUMA has always used crossbar switching and that was also in the
1990s.
Here, patent
4942516
filed 1988, published in 1990, with a priority date of 1970!
How's
that for prior art?[ Reply to This | Parent | # ]
|
|
|
|
|