Re: State of the Mainframe - News Article
On Tue, 24 Jan 2006 12:16:58 -0600, Tom Schmidt [EMAIL PROTECTED] wrote: -snip- I think it is myopic to reduce the mainframe community to the viewpoint of 2 males, one from Pennsylvania and the other from Texas, and pass it off as the state of mainframes. (What of the females? What of the rest of the world?) Just so you know, I'm not from Texas, and very happy to be able to say that. :) Mark Post -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: State of the Mainframe - News Article
On Tue, 24 Jan 2006 15:42:58 -0500, Kreiter, Chuck [EMAIL PROTECTED] wrote: I can't say I understand these articles that talk about escalating mainframe costs. My reading of it wasn't that it was about _escalating_ costs, but rather that the costs are still too high. At least, that was my perspective when I spoke with the author. Mark Post -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: State of the Mainframe - News Article
On Tue, 24 Jan 2006 16:02:45 -0600, Sebastian Welton [EMAIL PROTECTED] wrote: I do know that one of the persons interviewed who stated that his employer would never use FLEX-ES has got his facts wrong slightly. I know that his employer has contacted resellers of FLEX-ES regarding the purchase of such a system a number of times. However, this was not in the USA. I have to disagree that I got it wrong. I was talking of the people in my company that set strategy, standards, and direction, not the actions of any particular account out in the field. I tried to get them to at least look at the technology, and they flatly refused. Mark Post -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: State of the Mainframe - News Article
Undoubtedly to the choir. But they can buy small/medium servers and software from petty cash. After some dozen(s), it's noticed that the support admins are overworked and stressed, and they pull a position from mainframe and hire another PC jockey. By the time the TCO arguments really make a difference, the commitment to Micro$oft is as entrenched as the mainframe was 15 years ago. In fact, we've seen a number of studies *directly from customers* that show that software costs are frequently higher on distributed servers than on the mainframe. Customers tend to forget that they must buy many copies of software to accomplish the same thing that one copy does on the mainframe. Not only must they buy more software copies, but they must also dedicate more bodies to administering that infrastructure. But I'm preaching to the choir here, right? I won't go into detail about where IBM sees issues in software pricing, as it would undoubtedly cause consternation from some who frequent this list. Suffice it to say that IBM knows about the problem and is working to address it. That's precisely why you see the various flavors of workload-based software pricing and the specialty engines that do not count towards increased MSUs on the mainframe. --- Bill Seubert zSeries Software I/T Architect IBM Corporation [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: State of the Mainframe - News Article
Undoubtedly to the choir. But they can buy small/medium servers and software from petty cash. After some dozen(s), it's noticed that the support admins are overworked and stressed, and they pull a position from mainframe and hire another PC jockey. By the time the TCO arguments really make a difference, the commitment to Micro$oft is as entrenched as the mainframe was 15 years ago. Maybe. But then, if you continue the narrative, one of two things happens: 1. The too-high costs result in the company's lack of market competitiveness, and the company goes out of business (figuratively or literally). How fast that happens depends on the degree of market competition, but the marketplace seems to be getting more brutal with each passing year. (That's great for consumers, I guess.) 2. The company outsources their IT so they can get their costs back under control. It's the single biggest reason for outsourcing: the business-side managers throw up their hands and say, Enough. We have a business to run. We're tired of paying for your IT toys, and we just want to give our salesforce better information, run the production line more efficiently, etc., etc. By the way, I don't think anybody is arguing that every computer should be a mainframe or that every computer should be a PC. There's going to be a mix. But if you get the mix wrong then it's door #1 or door #2. :-) - - - - - Timothy F. Sipples Consulting Enterprise Software Architect, z9/zSeries IBM Japan Ltd. E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: State of the Mainframe - News Article
The user had to pay for IMS based on the size of the processor Yes, in the late 1980s. Fast forward to the 2000s and there's subcapacity workload-based licensing, for IMS and just about everything else. Now I'm not arguing that it's the perfect answer in every situation, but it's relatively new and much more business-friendly, IMHO. - - - - - Timothy F. Sipples Consulting Enterprise Software Architect, z9/zSeries IBM Japan Ltd. E-Mail: [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
State of the Mainframe - News Article
Yes, in the late 1980s. Fast forward to the 2000s and there's subcapacity workload-based licensing, for IMS and just about everything else. Now I'm not arguing that it's the perfect answer in every situation, but it's relatively new and much more business-friendly, IMHO. It doesn't even come close to solving the problem. In fact, after two and a half decades of attrition, it has almost no effect at all except partially helping to contain increases due to extra MIPS capacity with succeeding generations. -- Phil Payne http://www.isham-research.co.uk +44 7833 654 800 -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Tape is Far More Expensive the DASD? (Was: State of the Mainframe - News Article)
Ron, Some corrections about Hitachi boxes: 7700C could be expanded to 1,5TB. 5 cabinets (2L,Controller,2R). It wasn't described in initial documentation, but Im' pretty sure it is possible, because I used such configuration. 15 classics sounds horribly, but the same capacity can be easily achived using *one* 7700E. Again, it is worth to know that 7700E supports 47GB drives (AFAIR not mentioned in the initial documentation). Why don't you mention a little younger machines ? I suspect, it's because they don't fit to your figure g Even 9900V are available on the stock, or latest Symmetrix (pre-DMZ). Prices are really attractive. Regarding the licenses: *Some* of them are optional, but if you don't have the licenses then you don't have all the features you described. BTW: some disk vendors tie the license to the box, so it is possible and legal to buy the box with the license for PPRC. HDS don't do it. Last but not least: Minus n generation can be price effective, even with the licenses. Regarding the solution: I didn't say the soultion based on cheap (non-USP) DASD is state of the art. No. It has reduced functionality and limited scalability. However you get 90% of features (IMHO the most important ones) for fraction of the price. It depends, as usually. Regards -- Radoslaw Skorupka Lodz, Poland Ron and Jenny Hawkins wrote: Radoslaw, All of the software is you mentioned is optional. The only software that is required for External Storage on the USP or NSC is the Universal Volume Manager (UVM) Software. Also, I don't think it is correct to say that all of the software is all charged at Enterprise price across the board. Some software that delivers Enterprise Storage functions for cheap disk is charged at Enterprise price (e.g. TrueCopy), but there is other Software that uses a Tiered Pricing Model (e.g. Tuning Manager). I think of software like TrueCopy in the same way as a good quality spanner - the tool costs the same price whether you use it on a Mercedes or lawnmower. I'm not sure how buying minus n boxes is going to make cheaper replacement for TrueCopy and Shadowimage compared to using UVM, as you are still going to have to buy the licensed software for those boxes in order to have those features. For example, if you want to do something like FlashCopy from your current storage to some minus n boxes you will have to buy FDR/PAS or TDMF, or buy a Remote Copy licence if both boxes are from the same vendor. I'm just picturing how much floor space and power it would take to store 10TB of data in a 10 year old EMC 5500 or HDS 7700C, compared to a few drawers of SATA storage using 400GB drives. My fading memory seems to recall that a 7700 classic could hold about 720GB, so you would need 15 fully configured boxes to hold 10TB. I mention the 7700 classic because I seem to recall that you are using 7700 and 7700E for ML2 data. If you had a USP or NSC, then these could be replaced with some cheap SATA drives, and the only software you would need is UVM, where UVM is based on External TB only. In most cases there would be some TCO savings in floor space, environmentals and Maintenance costs, but we know this varies from site to site and country to country - YMMV. BTW, how does TrueCopy pricing compare with VTS P2P, including the CX1 units in both sites? Ron SATA drives attached to mainframe looks vry good ...unless you see the prices... I like very much the idea of HDS TagmaStore USP. You can attach almost any DASD box you have (in fact, CKD boxes are not supported) and use it as mainframe/or FBA storage. You can mirror it using PPRC, you can FlashCopy (Shadowimage) it. Everything is OK, except the prices. You pay license per TB of external storage attached to USP, you pay per TB fees for PPRC and Shadowimage. At the end you'll find out the SATA storage is cheap, but the solution is not. For relatively small to average capacities it is more effective to buy previous generation technology DASD boxes (or second hand - in that case price is dramatically lower), for large capacities tape is still cost-effective. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Tape is Far More Expensive the DASD? (Was: State of the Mainframe - News Article)
Radoslaw, Some corrections about Hitachi boxes: 7700C could be expanded to 1,5TB. 5 cabinets (2L,Controller,2R). It wasn't described in initial documentation, but Im' pretty sure it is possible, because I used such configuration. That's correct using 9GB drives. The 7700s I've worked with used the original 4.5GB drives 15 classics sounds horribly, but the same capacity can be easily achived using *one* 7700E. Again, it is worth to know that 7700E supports 47GB drives (AFAIR not mentioned in the initial documentation). The 7700E supports up to 73GB drives, with a total of 18TB with all ACP and cabinets installed. 10TB would need three cabinets (Controller plus two Array Cabinets), which takes up about the same space as three refrigerators or a Symmetrix 5500, 87nn, 89nn, DMX2K, DMX3K and DMX3. This is still significantly larger than two SATA drawers in a 19 rack. Why don't you mention a little younger machines ? I suspect, it's because they don't fit to your figure g Even 9900V are available on the stock, or latest Symmetrix (pre-DMZ). Prices are really attractive. Yes, I chose the oldest unit that I recall you are using (I checked your post of Jan1 last year and you did say HDS7700) to dress up the argument. The Maintenance on 10TB of DASD more than two or three years old would still be a damn site more expensive than 10TB of SATA disk. Of course if your not paying for maintenance that's a different matter. Regarding the licenses: *Some* of them are optional, but if you don't have the licenses then you don't have all the features you described. BTW: some disk vendors tie the license to the box, so it is possible and legal to buy the box with the license for PPRC. HDS don't do it. Last but not least: Minus n generation can be price effective, even with the licenses. The software is licensed and runs on the USP or NSC, not the SATA storage. No-one is asking IBM or Innovation for a discount when you use their utilities to copy from disk to tape, so why do you want it for SATA. A 10TB licence for Shadowimage allows you to copy to and from any internal and external drives as long as the sessions don't exceed 10TB. What you're asking is to have three different prices for Shadowimage: 1) Internal == internal 2) Internal == SATA 3) SATA == SATA What about if you are using a 7700E for external storage - should there be another price structure? Anyway, I'll sit on this until DSS and FDR have a cheap Disk to Tape licence. Regarding the solution: I didn't say the soultion based on cheap (non-USP) DASD is state of the art. No. It has reduced functionality and limited scalability. However you get 90% of features (IMHO the most important ones) for fraction of the price. It depends, as usually. Maybe we've missed the point. The whole idea of hooking up old and/or cheap kit to the USP or NSC is to provide that kit with functionality and scalability that it doesn't have. Ron -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Tape is Far More Expensive the DASD? (Was: State of the Mainframe - News Article)
Ron and Jenny Hawkins wrote: Yes, I chose the oldest unit that I recall you are using (I checked your post of Jan1 last year and you did say HDS7700) to dress up the argument. The Maintenance on 10TB of DASD more than two or three years old would still be a damn site more expensive than 10TB of SATA disk. Of course if your not paying for maintenance that's a different matter. Otherwise I wouldn't take second hand units. BTW: It is possible to get support from broker, much cheaper. Regarding the licenses: *Some* of them are optional, but if you don't have the licenses then you don't have all the features you described. BTW: some disk vendors tie the license to the box, so it is possible and legal to buy the box with the license for PPRC. HDS don't do it. Last but not least: Minus n generation can be price effective, even with the licenses. The software is licensed and runs on the USP or NSC, not the SATA storage. No-one is asking IBM or Innovation for a discount when you use their utilities to copy from disk to tape, so why do you want it for SATA. A 10TB licence for Shadowimage allows you to copy to and from any internal and external drives as long as the sessions don't exceed 10TB. What you're asking is to have three different prices for Shadowimage: 1) Internal == internal 2) Internal == SATA 3) SATA == SATA What about if you are using a 7700E for external storage - should there be another price structure? Anyway, I'll sit on this until DSS and FDR have a cheap Disk to Tape licence. I'm definitely not asking for any changes in licensing models, neither expect anything from HDS. The only thing I did is economical analysis of the solution and comparison to other solutions of the same problem/need. The other solutions are worse, but they fulfill the basic needs at lower price. Mercedes is good car, but Hyundai is sometimes enough for simple transportation. Regards -- Radoslaw Skorupka Lodz, Poland -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: State of the Mainframe - News Article
On Thu, 26 Jan 2006 13:23:15 +0100, Phil Payne [EMAIL PROTECTED] wrote: It doesn't even come close to solving the problem. In fact, after two and a half decades of attrition, it has almost no effect at all except partially helping to contain increases due to extra MIPS capacity with succeeding generations. Phil, please help me understand what part of the equation you believe is still a problem. I suspect you already know this, but the cost of ownership of server systems in general is very much skewed towards administrative (people) cost. HW SW is a relatively small part of the equation, and getting smaller as server proliferation continues unchecked. In fact, we've seen a number of studies *directly from customers* that show that software costs are frequently higher on distributed servers than on the mainframe. Customers tend to forget that they must buy many copies of software to accomplish the same thing that one copy does on the mainframe. Not only must they buy more software copies, but they must also dedicate more bodies to administering that infrastructure. But I'm preaching to the choir here, right? I won't go into detail about where IBM sees issues in software pricing, as it would undoubtedly cause consternation from some who frequent this list. Suffice it to say that IBM knows about the problem and is working to address it. That's precisely why you see the various flavors of workload-based software pricing and the specialty engines that do not count towards increased MSUs on the mainframe. --- Bill Seubert zSeries Software I/T Architect IBM Corporation [EMAIL PROTECTED] -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: State of the Mainframe - News Article
-Original Message- From: IBM Mainframe Discussion List On Behalf Of Ted MacNEIL On the same system, an IMS DB/DC system managing customer service - 6 to 8 seats, paying in absolute terms actually more than the CICS/DL/I system. It worked out at something like $23,000/butt/month. Guess what? The customer moved it to Prime, and the mainframe lost another application. You have to take all $ per all butts. You cannot cherry-pick. As always, it depends. My reading of Phil's example demands cherry-picking: But for the 6- to 8-seat application, there was no need for an IMS license. In that example, licensing IMS for an application with so little usage was much like hiring a container ship to move one box of paperclips around. You have to rent the whole boat (with crew) regardless how little of its capacity you want to use. -jc- -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Tape is Far More Expensive the DASD? (Was: State of the Mainframe - News Article)
Ron and Jenny Hawkins wrote: Edward, I wouldn't say far more expensive, but in terms of TCO, SATA DISK is quite competitive with Tape, with all the value added DASD stuff like Remote Copy, FCV1 and FCV2, Zero Mount-Position-Rewind time, PAV, multi-user file access, etc. Come to think of it, I would much rather DEFRAG a disk based ML2 volume with FCV2 than recycle Tape based ML2 volume :) SATA drives attached to mainframe looks vry good ...unless you see the prices... I like very much the idea of HDS TagmaStore USP. You can attach almost any DASD box you have (in fact, CKD boxes are not supported) and use it as mainframe/or FBA storage. You can mirror it using PPRC, you can FlashCopy (Shadowimage) it. Everything is OK, except the prices. You pay license per TB of external storage attached to USP, you pay per TB fees for PPRC and Shadowimage. At the end you'll find out the SATA storage is cheap, but the solution is not. For relatively small to average capacities it is more effective to buy previous generation technology DASD boxes (or second hand - in that case price is dramatically lower), for large capacities tape is still cost-effective. -- Radoslaw Skorupka Lodz, Poland -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: State of the Mainframe - News Article
That's his point. The user had to pay for IMS based on the size of the processor, which made it prohibitive to keep the application on the mainframe. Had there been a way to pay a reasonable fee based on the number of seats using the software, the app would never have had to be moved off to the PR1ME box. IOW, making the customer rent the whole boat forced them to move one more app off. Jon snip As always, it depends. My reading of Phil's example demands cherry-picking: But for the 6- to 8-seat application, there was no need for an IMS license. In that example, licensing IMS for an application with so little usage was much like hiring a container ship to move one box of paperclips around. You have to rent the whole boat (with crew) regardless how little of its capacity you want to use. /snip -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: State of the Mainframe - News Article
Amen to that. It has made the mainframe's great strength into a great weakness. snip The fundamental problem is the whole concept of charging applications by system size in a multi-application system. Period. It's nonsense, and it's been nonsense since 1989. /snip This has made life more and more difficult for us mainframers as the years have progressed. snip Charging for applications by the size of the system they just happen to run on eventually disqualifies all applications from multi-user systems. And that's the definition of a mainframe. /snip Jon -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: State of the Mainframe - News Article
-Original Message- From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of Phil Payne Sent: Tuesday January 24 2006 15:05 To: IBM-MAIN@BAMA.UA.EDU Subject: State of the Mainframe - News Article On the same system, an IMS DB/DC system managing customer service - 6 to 8 seats, paying in absolute terms actually more than the CICS/DL/I system. It worked out at something like $23,000/butt/month. Guess what? The customer moved it to Prime, and the mainframe lost another application. Having worked on Prime (and they're still one of my favorite systems, architectures, etc.), based on the size of the application, I'll have to agree with Jon B. with the container ship/box o' paperclips analogy. But I feel for this company because not too long after the stated date of 1989 is when Prime started to go south, big time. So they probably had to turn around and buy some even-crappier Windoze-based crap. Later, Ray -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Tape is Far More Expensive the DASD? (Was: State of the Mainframe - News Article)
We did the math and found the TCO of storing data on tape is far more expensive than DASD. So we use only DASD for data storage. We use tape only for backups. We have about 2tb of DASD and about 1,000 tapes. We consume less than 90 person minutes per day for *all* tape related activities. There are plans to reduce that time to near zero. This is part of a total 'lights out' strategy from day one (in 1999). We are up to a total of six 'operators' that spend most of their time managing telecom issues. Long ago, we realized that humans were expensive and tended to make mistakes. Keeping the human involvement minimal really pays off. -Original Message- From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of Edward E. Jaffe Sent: Tuesday, January 24, 2006 5:25 PM To: IBM-MAIN@BAMA.UA.EDU Subject: Tape is Far More Expensive the DASD? (Was: State of the Mainframe - News Article) Hal Merritt wrote: Hardware costs have been plummeting for a while now. So much so that it makes a serious difference in how we manage resources. We now consider tape to be far more expensive than DASD. YMMV. Tape far more expensive than DASD? I find this difficult to believe. Could you elaborate? -- -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Tape is Far More Expensive the DASD? (Was: State of the Mainframe - News Article)
I'd be interested in seeing some of that math .. So would I! We found once we went to libraries, the cost was less (or better). If you handle tapes without automation, it's expensive. Once you get into a VTS solution, it's cheap. - -teD I’m an enthusiastic proselytiser of the universal panacea I believe in! -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
State of the Mainframe - News Article
Folks: More food for thought http://www.esj.com/news/article.aspx?EditorialsID=1597. First of a 3-part series on the future of the mainframe. Also, near and dear to my heart, they will explore the state of mainframe training in the future articles. I'd be interested in your thoughts, comments, etc. Cheers, John Wynton Themis Training 800-756-3000 Toll Free 910-673-1427 Direct 908-723-5064 Cell [EMAIL PROTECTED] mailto:[EMAIL PROTECTED] http://www.themisinc.com http://www.themisinc.com/ -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: State of the Mainframe - News Article
On Tue, 24 Jan 2006 12:52:08 -0500, John Wynton wrote: Folks: More food for thought http://www.esj.com/news/article.aspx?EditorialsID=1597. First of a 3-part series on the future of the mainframe. Also, near and dear to my heart, they will explore the state of mainframe training in the future articles. I'd be interested in your thoughts, comments, etc. The article is an interview with two persons, one is the major non-IBM zLinux point person and the other seems to be getting himself interviewed in several publications just this month (see http://www.techworld.com/networking/features/index.cfm?FeatureID=2163 for a different interview that includes Joe Poole, where it identifies his company). Curiously, he is listed as a mainframe systems programmer in your article whereas he is listed as Tech Support Manager in other publications or presentations on the web. It doesn't look as if he was demoted recently -- it looks more like the article's author got some facts wrong; I wonder how many. I think it is myopic to reduce the mainframe community to the viewpoint of 2 males, one from Pennsylvania and the other from Texas, and pass it off as the state of mainframes. (What of the females? What of the rest of the world?) -- Tom Schmidt Madison, WI -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: State of the Mainframe - News Article
Scanned it. Saw nothing of interest to me. I did find myself being a little thankful the ESJ is not an airline magazine :-) -Original Message- From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of John Wynton Sent: Tuesday, January 24, 2006 11:52 AM To: IBM-MAIN@BAMA.UA.EDU Subject: State of the Mainframe - News Article Folks: More food for thought http://www.esj.com/news/article.aspx?EditorialsID=1597. First of a 3-part series on the future of the mainframe. Also, near and dear to my heart, they will explore the state of mainframe training in the future articles. I'd be interested in your thoughts, comments, etc. Cheers, John Wynton Themis Training 800-756-3000 Toll Free 910-673-1427 Direct 908-723-5064 Cell [EMAIL PROTECTED] mailto:[EMAIL PROTECTED] http://www.themisinc.com http://www.themisinc.com/ -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: State of the Mainframe - News Article
I can't say I understand these articles that talk about escalating mainframe costs. At my company, we consistently make deals where we end up with better storage, faster processors, etc and our costs actually go down. Whereas, our counterparts in open systems consistently see their hardware, software and support personnel costs go significantly. I don't know if my boss is just a good negotiator or if everyone sees this trend. Vendors who have their hand out at every CPU upgrade are generally shown the door. Anyone else notice this or are we just unique? * This message was scanned by State Auto's mail server for viruses and objectionable content. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: State of the Mainframe - News Article
I have to agree with Tom and Chuck, looks like a couple of guys with there own opinion and they are very much entitled with it. Ironically I think Chuck and I have the same negotiator working for both our companies, because our costs keep going down while the open systems costs go up. You do not get the cool GUI interface like you do on the open systems. The only thing MF can offer is scalability and reliability, none of the cool stuff you see in the sky mall magazine. ;-) Thanks, Fletch -Original Message- From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of Kreiter, Chuck Sent: Tuesday, January 24, 2006 3:43 PM To: IBM-MAIN@BAMA.UA.EDU Subject: Re: State of the Mainframe - News Article I can't say I understand these articles that talk about escalating mainframe costs. At my company, we consistently make deals where we end up with better storage, faster processors, etc and our costs actually go down. Whereas, our counterparts in open systems consistently see their hardware, software and support personnel costs go significantly. I don't know if my boss is just a good negotiator or if everyone sees this trend. Vendors who have their hand out at every CPU upgrade are generally shown the door. Anyone else notice this or are we just unique? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: State of the Mainframe - News Article
Anyone else notice this or are we just unique? The big problem is the mainframe costs are all up front. The alternate platforms cost sneak up on you, and are sometimes missed. Also, nobody does the $/butt arithmetic. I have, for many years, pointed this out. But, reality is 99% perception. And, the remaining 1% is ignored. - -teD I’m an enthusiastic proselytiser of the universal panacea I believe in! -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: State of the Mainframe - News Article
we have the same situation here and we are doing a cpu upgrade from our 2064 because like you said will save us money we are also looking at alternatives to those software products that have such large cost increase's to the new cpu we may be dropping one vendors monitoring software because of it and switching to anothers Kreiter, Chuck [EMAIL PROTECTED]To: IBM-MAIN@BAMA.UA.EDU TEAUTO.COM cc: Sent by: IBM Subject: Re: State of the Mainframe - News Article Mainframe Discussion List [EMAIL PROTECTED] EDU 01/24/2006 03:42 PM Please respond to IBM Mainframe Discussion List I can't say I understand these articles that talk about escalating mainframe costs. At my company, we consistently make deals where we end up with better storage, faster processors, etc and our costs actually go down. Whereas, our counterparts in open systems consistently see their hardware, software and support personnel costs go significantly. I don't know if my boss is just a good negotiator or if everyone sees this trend. Vendors who have their hand out at every CPU upgrade are generally shown the door. Anyone else notice this or are we just unique? * This message was scanned by State Auto's mail server for viruses and objectionable content. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: State of the Mainframe - News Article
I do know that one of the persons interviewed who stated that his employer would never use FLEX-ES has got his facts wrong slightly. I know that his employer has contacted resellers of FLEX-ES regarding the purchase of such a system a number of times. However, this was not in the USA. quote He says his employer flatly rules out using FLEX/ES hardwareeven if a customer doesnt qualify for Big Blues smallest mainframe systems. I know my company won't even consider using FLEX/ES, even if we have clients that want a dedicated machine, but don't come anywhere near needing an entry-level z800. /quote And then again, FLEX-ES is not hardware, unless you count the channel cards! This, along with a number of other articles I see about mainframes is why I take them with a pinch of salt. Seb. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: State of the Mainframe - News Article
ADDRESS OPINION X = RANTMODE(ACTIVATE) Nope. You are in good company. IBM continues to offer ways to cut software costs. ISV's are under a lot of pressure from both IBM and us. More and more shops are finding serious business cases to fire uncooperative vendors. And those shops are reporting no regrets after doing so. Hardware costs have been plummeting for a while now. So much so that it makes a serious difference in how we manage resources. We now consider tape to be far more expensive than DASD. YMMV. I suppose it might be true that some old, large shops continue to see escalating costs. But I think that is more a cultural and management issue. Which brings me back to my point: there was nothing of interest to me in the article so far. I can only assume the punch line will be a discussion of 'alternatives'. I'll bet a virtual beverage that Sun will be mentioned. X = RANTMODE(DEACTIVATE) EXIT RC=0.02 -Original Message- From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of Kreiter, Chuck Sent: Tuesday, January 24, 2006 2:43 PM To: IBM-MAIN@BAMA.UA.EDU Subject: Re: State of the Mainframe - News Article I can't say I understand these articles that talk about escalating mainframe costs. At my company, we consistently make deals where we end up with better storage, faster processors, etc and our costs actually go down. Whereas, our counterparts in open systems consistently see their hardware, software and support personnel costs go significantly. I don't know if my boss is just a good negotiator or if everyone sees this trend. Vendors who have their hand out at every CPU upgrade are generally shown the door. Anyone else notice this or are we just unique? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
State of the Mainframe - News Article
The big problem is the mainframe costs are all up front. The alternate platforms cost sneak up on you, and are sometimes missed. Also, nobody does the $/butt arithmetic. I can't let that pass. The fundamental problem is the whole concept of charging applications by system size in a multi-application system. Period. It's nonsense, and it's been nonsense since 1989. $/butt? Yeah we did the arithmetic. It was about 1989, and I've used the example many times. A customer of mine had a massive 20,000 seat CICS/DL/I system that ran his company. Very happy with $/butt. On the same system, an IMS DB/DC system managing customer service - 6 to 8 seats, paying in absolute terms actually more than the CICS/DL/I system. It worked out at something like $23,000/butt/month. Guess what? The customer moved it to Prime, and the mainframe lost another application. Charging for applications by the size of the system they just happen to run on eventually disqualifies all applications from multi-user systems. And that's the definition of a mainframe. In every other sphere of commercial activity, it make sense to share a common large facility. You don't find user departments with their own dining rooms, air conditioning, electricity generation, first aid, elevators, etc., because the whole purpose of consolidating functionality is to benefit from economies of scale. Imagine, for a moment or two, that a laboratory wants to use UV light in a 2m3 space. The building janitor says: Fine, but you have to pay for the building's entire 80,000m3 space, even though 79,998m3 won't use it. PSLC. -- Phil Payne http://www.isham-research.co.uk +44 7833 654 800 -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Tape is Far More Expensive the DASD? (Was: State of the Mainframe - News Article)
Hal Merritt wrote: Hardware costs have been plummeting for a while now. So much so that it makes a serious difference in how we manage resources. We now consider tape to be far more expensive than DASD. YMMV. Tape far more expensive than DASD? I find this difficult to believe. Could you elaborate? -- - | Edward E. Jaffe|| | Mgr, Research Development| [EMAIL PROTECTED]| | Phoenix Software International | Tel: (310) 338-0400 x318 | | 5200 W Century Blvd, Suite 800 | Fax: (310) 338-0801| | Los Angeles, CA 90045 | http://www.phoenixsoftware.com | - -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: State of the Mainframe - News Article
On the same system, an IMS DB/DC system managing customer service - 6 to 8 seats, paying in absolute terms actually more than the CICS/DL/I system. It worked out at something like $23,000/butt/month. Guess what? The customer moved it to Prime, and the mainframe lost another application. You have to take all $ per all butts. You cannot cherry-pick. If you start saying that CICS is cheap, IMS is expensive, you are slitting your own throat. It has to be all $ for all butts on a specific platform. All else is apples vs oranges. - -teD I’m an enthusiastic proselytiser of the universal panacea I believe in! -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: State of the Mainframe - News Article
On Jan 24, 2006, at 2:42 PM, Kreiter, Chuck wrote: I can't say I understand these articles that talk about escalating mainframe costs. At my company, we consistently make deals where we end up with better storage, faster processors, etc and our costs actually go down. Whereas, our counterparts in open systems consistently see their hardware, software and support personnel costs go significantly. I don't know if my boss is just a good negotiator or if everyone sees this trend. Vendors who have their hand out at every CPU upgrade are generally shown the door. Anyone else notice this or are we just unique? Hardware is one thing software is another. A certain vendor may offer you a cheap price but when it comes time to upgrade get out the wheel barrels of cash. Ed -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Tape is Far More Expensive the DASD? (Was: State of the Mainframe - News Article)
Edward, I wouldn't say far more expensive, but in terms of TCO, SATA DISK is quite competitive with Tape, with all the value added DASD stuff like Remote Copy, FCV1 and FCV2, Zero Mount-Position-Rewind time, PAV, multi-user file access, etc. Come to think of it, I would much rather DEFRAG a disk based ML2 volume with FCV2 than recycle Tape based ML2 volume :) Ron -Original Message- From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of Edward E. Jaffe Sent: Wednesday, 25 January 2006 7:25 AM To: IBM-MAIN@BAMA.UA.EDU Subject: Tape is Far More Expensive the DASD? (Was: State of the Mainframe - News Article) Hal Merritt wrote: Hardware costs have been plummeting for a while now. So much so that it makes a serious difference in how we manage resources. We now consider tape to be far more expensive than DASD. YMMV. Tape far more expensive than DASD? I find this difficult to believe. Could you elaborate? -- - | Edward E. Jaffe|| | Mgr, Research Development| [EMAIL PROTECTED]| | Phoenix Software International | Tel: (310) 338-0400 x318 | | 5200 W Century Blvd, Suite 800 | Fax: (310) 338-0801| | Los Angeles, CA 90045 | http://www.phoenixsoftware.com | - -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
