INFO-VAX Tue, 01 Apr 2008 Volume 2008 : Issue 183 Contents: Re: Does POWER_OFF really work ? Re: Does POWER_OFF really work ? Re: Does POWER_OFF really work ? Re: Does POWER_OFF really work ? Re: Does POWER_OFF really work ? Re: Does POWER_OFF really work ? Send your questions to Paul Otelini Re: System Programming Resources for Alpha Architecture Re: System Programming Resources for Alpha Architecture Re: zeroing out errors? ---------------------------------------------------------------------- Date: Mon, 31 Mar 2008 17:17:15 -0400 From: JF Mezei Subject: Re: Does POWER_OFF really work ? Message-ID: <47f15559$0$3857$c3e8da3@news.astraweb.com> VAXman- @SendSpamHere.ORG wrote: > 'Twould be fine with me too but the power-hour wasn't about light pollution, > it was a gesture about conserving energy. Power companies generate electric > based on average demands. I really doubt that they made any fuel consumption > changes based on a minor brief hour dip on the grid. Not sure how authoritative, but I heard the city of Toronto noticed an 8% drop in electrical consumption during that hour. ------------------------------ Date: Mon, 31 Mar 2008 15:37:38 -0700 From: Fred Bach Subject: Re: Does POWER_OFF really work ? Message-ID: JF Mezei wrote: > VAXman- @SendSpamHere.ORG wrote: > >> 'Twould be fine with me too but the power-hour wasn't about light pollution, >> it was a gesture about conserving energy. Power companies generate electric >> based on average demands. I really doubt that they made any fuel consumption >> changes based on a minor brief hour dip on the grid. > > Not sure how authoritative, but I heard the city of Toronto noticed an > 8% drop in electrical consumption during that hour. I heard on the radio it was about 7% out here Vancouver way. . fred . ------------------------------ Date: Mon, 31 Mar 2008 15:54:28 -0700 (PDT) From: AEF Subject: Re: Does POWER_OFF really work ? Message-ID: On Mar 31, 11:09 am, VAXman- @SendSpamHere.ORG wrote: > In article <1ea8e2fb-040d-4f8f-ad2f-40a84aabd...@e39g2000hsf.googlegroups.com>, AEF writes: > > > > >On Mar 30, 7:46 am, VAXman- @SendSpamHere.ORG wrote: > >> In article , hel...@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes: > > >> >In article <47ef152b$0$23915$c3e8...@news.astraweb.com>, JF Mezei > >> > writes: > > >> >> Figured I would shut down one of my alphas for EARTH-HOUR . > > >> >What is EARTH-HOUR? > > >> Just a way to make people concerned about the environment feel good about > >> themselves and nothing more. > > >>http://www.earthhour.org/ > > >> -- > >> VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM > > >> "Well my son, life is like a beanstalk, isn't it?" > > >>http://tmesis.com/drat.html > > >Anything that makes the sky darker, even for only an hour, and/or has > >the net result of improving lighthing by increasing efficiency (esp. > >by decreasing wasted light that goes up into the sky, helping no one), > >and reducing glare (which actually INCREASES nighttime visibility > >making a safer and more pleasant nighttime environment), is fine by > >me. Bravo! > > >Seehttp://www.darksky.org/ > > 'Twould be fine with me too but the power-hour wasn't about light pollution, > it was a gesture about conserving energy. Power companies generate electric > based on average demands. I really doubt that they made any fuel consumption > changes based on a minor brief hour dip on the grid. > > -- > VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM > > "Well my son, life is like a beanstalk, isn't it?" > > http://tmesis.com/drat.html I don't care about the energy consumption as much as I do about dark skies and reduced glare and improved visibility and safety! At least it was darker for an hour and maybe people realized that it's not so bad and maybe even enjoyed seeing more stars and blah blah blah. It's a step in the right direction and with the NYC metro area being a major light pollution source, it's just nice to hear of something going right somewhere, even if only for an hour. And if there are any dark-sky enthusiasts in one of these cities it might help them with their message! http://www.darksky.org/ AEF ------------------------------ Date: 31 Mar 2008 22:56:09 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: Does POWER_OFF really work ? Message-ID: <47f16c09$0$5615$607ed4bc@cv.net> In article <47f15559$0$3857$c3e8da3@news.astraweb.com>, JF Mezei writes: >VAXman- @SendSpamHere.ORG wrote: > >> 'Twould be fine with me too but the power-hour wasn't about light pollution, >> it was a gesture about conserving energy. Power companies generate electric >> based on average demands. I really doubt that they made any fuel consumption >> changes based on a minor brief hour dip on the grid. > >Not sure how authoritative, but I heard the city of Toronto noticed an >8% drop in electrical consumption during that hour. Could be but the generators were still turning! -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM "Well my son, life is like a beanstalk, isn't it?" http://tmesis.com/drat.html ------------------------------ Date: Mon, 31 Mar 2008 16:40:31 -0700 (PDT) From: johnwallace4@gmail.com Subject: Re: Does POWER_OFF really work ? Message-ID: <84178d7d-3ef1-4277-b019-802510649b27@e6g2000prf.googlegroups.com> On Mar 31, 5:09 pm, VAXman- @SendSpamHere.ORG wrote: > In article <1ea8e2fb-040d-4f8f-ad2f-40a84aabd...@e39g2000hsf.googlegroups.com>, AEF writes: > > > > >On Mar 30, 7:46 am, VAXman- @SendSpamHere.ORG wrote: > >> In article , hel...@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes: > > >> >In article <47ef152b$0$23915$c3e8...@news.astraweb.com>, JF Mezei > >> > writes: > > >> >> Figured I would shut down one of my alphas for EARTH-HOUR . > > >> >What is EARTH-HOUR? > > >> Just a way to make people concerned about the environment feel good about > >> themselves and nothing more. > > >>http://www.earthhour.org/ > > >> -- > >> VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM > > >> "Well my son, life is like a beanstalk, isn't it?" > > >>http://tmesis.com/drat.html > > >Anything that makes the sky darker, even for only an hour, and/or has > >the net result of improving lighthing by increasing efficiency (esp. > >by decreasing wasted light that goes up into the sky, helping no one), > >and reducing glare (which actually INCREASES nighttime visibility > >making a safer and more pleasant nighttime environment), is fine by > >me. Bravo! > > >Seehttp://www.darksky.org/ > > 'Twould be fine with me too but the power-hour wasn't about light pollution, > it was a gesture about conserving energy. Power companies generate electric > based on average demands. I really doubt that they made any fuel consumption > changes based on a minor brief hour dip on the grid. > > -- > VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM > > "Well my son, life is like a beanstalk, isn't it?" > > http://tmesis.com/drat.html "Power companies generate electric based on average demands." is a bit of an oversimplification. They're supposed to install and operate capacity based on some realistic view of peak demand; some magic market connection between demand and supply is supposed to make that happen (hello California). Once that's done, peak demand is likely significantly greater than average, so at any given time some capacity sits idle by definition. It makes sense to have the most expensive-to-run capacity be idle at any given time, but you have to balance this against the fact that sometimes you need power which can be switched on (or switched off) relatively quickly, and that the quickest to respond isn't necessarily going to be the cheapest to run. At the "slowest of the slow" end in responsiveness terms is nuclear power. Nuclear costs are largely impenetrable, it depends on whose accounting principles you prefer, but the timescales to start up and for a planned shut down are undeniably long, so nuclear is first on, last off. Coal has somewhat clearer costs and somewhat quicker response times, but still far too slow to handle a one-hour "blip". In the UK and elsewhere, electricity privatisation led to a "dash for gas" on the mainland where gas was cheap (Northern Ireland was different, no cheap gas). Large numbers of gas-fired pre-packaged relatively cheap relatively small relatively fast combined cycle gas turbine generators (CCGT) were built. In 1989 there were no CCGTs, in 2004 half the UK's fossil-fuel electricity came from CCGTs. CCGTs have a response time in minutes not hours and could be switched in and out relatively quickly in a cost effective manner (if you thought that using scarce natural gas resources for thermal electricity generation was a long term sensible cost effective use of a limited and valuable resource, which of course it wasn't, but who could have ever imagined us Europeans would soon end up dependent on Russia, Libya, Iran and Iraq [1] for our gas supplies). At the "fastest of the fast" in this space are the miracles of civil and electrical engineering known as "pumped storage" systems, such as Dinorwig in north Wales (www.fhc.co.uk). Dinorwig can go from 0 to 1.8GW output in less than 20 seconds when ready and waiting, or just over a minute from a no-notice all-stopped condition. And then it can keep going for a few hours at maximum capacity. Just what you need to cope with a million or so homes switching on their 3kW kettles all at the same time, or a lot of people switching lots of things off at the same time, or to fill the gap for a few minutes/hours till gas and coal generators catch up after an unplanned shutdown of a nuclear station. The total UK demand is probably 60GW or so, I forget, so Dinorwig's 1.8GW is a reasonably sizeable chunk, around half the size of the UK's largest coal-fired stations (say 4GW). So, average doesn't really cover it. One of the more interesting VMS sites in this sector was the national company which did the remote meter reading for large users. Large electricity users have their meters read automatically very frequently (every few minutes) and are charged not just by how many MWh they use, but also by their "maximum demand" (maximum MW) in an interval. This national meter reading systems was a VMS cluster, and rightly so, there was serious money at stake. Regards John [1] http://news.bbc.co.uk/1/hi/uk_politics/3581637.stm ------------------------------ Date: Tue, 01 Apr 2008 00:40:33 GMT From: BobH Subject: Re: Does POWER_OFF really work ? Message-ID: VAXman- @SendSpamHere.ORG wrote in news:47f16c09$0$5615$607ed4bc@cv.net: > In article <47f15559$0$3857$c3e8da3@news.astraweb.com>, JF Mezei > writes: >>VAXman- @SendSpamHere.ORG wrote: >> >>> 'Twould be fine with me too but the power-hour wasn't about light >>> pollution, it was a gesture about conserving energy. Power >>> companies generate electric based on average demands. I really >>> doubt that they made any fuel consumption changes based on a minor >>> brief hour dip on the grid. >> >>Not sure how authoritative, but I heard the city of Toronto noticed an >>8% drop in electrical consumption during that hour. > > Could be but the generators were still turning! They may well be spinning, but the amount of energy they use is proportional to the amount of power they are outputting. The wheels on a car may turn at the same speed down hill and uphill, but downhill the engine may be near idle, while uphill it is consuming gasoline at a much greater rate, even theough it is turning, along with the wheels, at the same speed. ------------------------------ Date: Mon, 31 Mar 2008 23:31:53 -0400 From: JF Mezei Subject: Send your questions to Paul Otelini Message-ID: <47f1ad24$0$3867$c3e8da3@news.astraweb.com> The BBC is organising an "Have your say" event with Intel's Paul Otellini this coming friday. You can submit your questions at: http://news.bbc.co.uk/2/hi/technology/7322205.stm If BBC deems the question to be important, it will send them to Intel for an answer. The more people send question about IA64's future, the more chances one of them will be sent to Otellini for answer. ------------------------------ Date: Mon, 31 Mar 2008 13:47:50 -0700 (PDT) From: The-Grue@hotmail.com Subject: Re: System Programming Resources for Alpha Architecture Message-ID: <2d74da48-b51b-414d-aeae-98f470d8f397@s8g2000prg.googlegroups.com> On Mar 30, 3:26=A0am, "John Wallace" wrote: > "Maciej W. Rozycki" wrote in messagenews:alpine.SOC= .1.00.0803300006500.26418@piorun... > > > Well, I think "System Programming / OS Development / OS Porting" makes i= t > >pretty much clear it is not application programming that is in question > >here. =A0I have yet to see an operating system that does not require any > >platform-specific code, including handwritten assembly. =A0And then even > >higher-level language OS code may require knowledge of processor-specific= > >details, like architecture of the register file or the TLB. =A0Resources > >mentioned elsewhere in this thread should be good enough though. > > =A0> Maciej > > Quite. You and I might think that but for some reason we seem to be in a > minority here. Some clarification from the original contributor would be > welcome in due course, but in the absence of that, his reference to Minix = on > his Myspace page gives a possible hint that perhaps we're not necessarily > looking at traditional application porting here... Hi! Thanks for all of the replies. I was looking at porting Minix to another platform for my Master's project. I guess I should have stated that up front as some folks went off on the application porting tangent. I am looking into the links and resources that some of you posted. I have several choices of platform and a bit of time to make the decision, so nothing is firm about Alpha yet or even if an OS port would be an acceptable project. Thanks again! James T. Sprinkle (The Grue) http://www.myspace.com/jamestsprinkle ------------------------------ Date: Mon, 31 Mar 2008 15:27:04 -0700 (PDT) From: johnwallace4@gmail.com Subject: Re: System Programming Resources for Alpha Architecture Message-ID: On Mar 31, 9:47 pm, The-G...@hotmail.com wrote: > On Mar 30, 3:26 am, "John Wallace" > wrote: > > > > > "Maciej W. Rozycki" wrote in messagenews:alpine.SOC.1.00.0803300006500.26418@piorun... > > > > Well, I think "System Programming / OS Development / OS Porting" makes it > > >pretty much clear it is not application programming that is in question > > >here. I have yet to see an operating system that does not require any > > >platform-specific code, including handwritten assembly. And then even > > >higher-level language OS code may require knowledge of processor-specific > > >details, like architecture of the register file or the TLB. Resources > > >mentioned elsewhere in this thread should be good enough though. > > > > Maciej > > > Quite. You and I might think that but for some reason we seem to be in a > > minority here. Some clarification from the original contributor would be > > welcome in due course, but in the absence of that, his reference to Minix on > > his Myspace page gives a possible hint that perhaps we're not necessarily > > looking at traditional application porting here... > > Hi! Thanks for all of the replies. I was looking at porting Minix to > another platform for my Master's project. I guess I should have > stated that up front as some folks went off on the application porting > tangent. I am looking into the links and resources that some of you > posted. I have several choices of platform and a bit of time to make > the decision, so nothing is firm about Alpha yet or even if an OS port > would be an acceptable project. Thanks again! > > James T. Sprinkle (The Grue)http://www.myspace.com/jamestsprinkle Thanks for the update. Now I know what you're potentially up to, I can fill in some accidental omissions from my first reply, apologies if the gapfilling is now unnecessary, and good luck with the Minix->Alpha port if you do go ahead! I mentioned chips without going into enough detail for you to reasonably easily find more info; it's out there, but you may need a few pointers. The first commercially available Alpha chip was the 21064 aka EV4. The next generation was the 21164 aka EV5, then 21264 aka EV6. The support chips had a similar numbering sequence - 21071, 21171, etc - the support chips might not be of interest to you if you are prepared to "be inspired by" code from BSD or Linux rather than work out from hardware docs. These chips were available on the "commodity" semiconductor market and have the level of published documentation required for someone building systems at that level. You can find many of the relevant earlier docs at ftp://ftp.netbsd.org/pub/NetBSD/misc/dec-docs/index.html - unfortunately, later ones (eg for EV6/21264) will take a bit more finding if you're interested (which you may not be, as these didn't really hit what might be called the "commodity" market in any worthwhile manner, even though it would have been technically quite possible, as anyone who's seen the innards of a DS10 will know). A 30-page April 2000 Compaq-badged technical overview of EV6 and EV6- based systems ("Exploring Alpha Power For Technical Computing") can be found at http://www.cs.virginia.edu/~dp8x/alpha21264/overview/wp_alpha_tech_apr00.pdf There are/were more recent Alpha chips (EV7, and a bit of EV8) but they're probably not all that relevant to your exercise and didn't come with the same level of published hardware documentation. One other that you might want to look at depending on your interests is DIGITAL's 1999 "Alpha vs IA64 performance and parallelism" whitepaper which offers an interesting comparison of performance based on out of order execution on Alpha (and as still used by today's x86/ AMD64) vs performance based on EPIC/VLIW on IA64. Obviously it's no longer available from DEC/CPQ/HP, but it's still out there on the Interweb, e.g. at http://www.cs.trinity.edu/~mlewis/CSCI3294-F01/Papers/alpha_ia64.pdf (and hopefully elsewhere). Have fun with the project anyway, John ------------------------------ Date: Mon, 31 Mar 2008 15:43:33 -0700 From: Fred Bach Subject: Re: zeroing out errors? Message-ID: Richard B. Gilbert wrote: > tadamsmar wrote: >> On Mar 27, 10:14 am, koeh...@eisner.nospam.encompasserve.org (Bob >> Koehler) wrote: >> >>> In article >>> , >>> tadamsmar writes: >>> >>> >>>> Is there an easy way to zero out the error count you see with SHOW DEV >>>> D >>>> without rebooting? >>> >>> Yes. But it has no benefits so it may not be maintained. If you >>> look around there is a kernel mode hack that does this. >> >> >> Use SET/DEV/RESET. >> >> Only benefit is that its easier to keep track of when I have new >> errors. I don't have to remember that it was "42" or whatever the >> last time I checked. > > I once wrote some DCL to check the error count and notify me if the > count had increased. It was part of a larger procedure that I ran every > workday morning to check for problems that needed my attention. I may > still have a copy somewhere if anyone cares. > Sounds useful! I'd like it, please. Thanks in advance. . fred . music at triumf dot c a ------------------------------ End of INFO-VAX 2008.183 ************************