Path: utzoo!attcan!uunet!samsung!think!snorkelwacker!paperboy!osf.org!collins From: collins@osf.org (Jeff Collins) Newsgroups: comp.sys.alliant Subject: Re: non-zero ipl Message-ID: <4869@paperboy.OSF.ORG> Date: 12 Mar 90 16:47:43 GMT References: <3723@alliant.Alliant.COM> <1990Mar6.222236.7098@uokmax.uucp> Sender: news@OSF.ORG Reply-To: collins@osf.org (Jeff Collins) Organization: Open Software Foundation Lines: 30 In article <3723@alliant.Alliant.COM>, neray@Alliant.COM (Phil Neray) writes: > Concentrix 5.5 has some very useful features for systems that have > lots of users (and need high throughput), and/or systems with the > requirement for lots of aggregate I/O. Good words on the benefits of Concentrix 5.5 deleted for brevity... > The aggregate I/O with disk striping (TFS) is about 2x compared with > Concentrix 5.0, and the aggregate network I/O is also about 2x (this > is for smaller packet sizes which are representative of terminal and > X11 network traffic). "Lock-wait" has been virtually eliminated. > Actually this understates the improvements that have been made in the I/O throughput. A more accurate statement would be that "aggregate I/O now scales with the number of I/O channels". Previous to Concentrix 5.5, the aggregate I/O topped out fairly early. With the complete multithreading available in 5.5, this no longer happens. I am sure that someone at Alliant can give very explicit numbers comparing 5.5 to version 5.0 (what about it Jeff?), but I know that it is very impressive. > My apologies if this sounds too much like a commercial, but the > Concentrix developers (led by Jeff Jones and Tony Anzelmo) have done > some really fine work here and I thought you should know. > I have to agree, the Concentrix 5.5 developers did a great job. Jeffery A. Collins Phone: (617) 621-8958 Open Software Foundation FAX: (617) 225-2782 11 Cambridge Center Email: collins@osf.org Cambridge MA 02142 uunet!osf.org!collins