Author |
Message
|
fitzcaraldo |
Posted: Mon Jul 18, 2011 6:53 pm Post subject: Broker PVU versus WESB PVU |
|
|
Voyager
Joined: 05 May 2003 Posts: 98
|
Interesting question.
We do a lot of broker development and one of our customers is considering migrating all their flows to Websphere ESB. Most of their stuff is XML so technically it appears possible.
They look at the PVU pricing and a Message Broker PVU is about 3 times that of a WESB PVU. The question they have asked us is: can 1 WESB PVU do the same amount of work as 1 Broker PVU? If you need 3 WESB PVUs to do the same amount of work as 1 broker PVU then there's no point - right?
Any thoughts? |
|
Back to top |
|
|
elvis_gn |
Posted: Mon Jul 18, 2011 10:23 pm Post subject: |
|
|
Padawan
Joined: 08 Oct 2004 Posts: 1905 Location: Dubai
|
Hi fitzcaraldo,
Lots of thoughts but which cannot be shared on a public forum, unfortunately
I'll just say that the decision to move to another ESB should not be based on software cost...but on technology refresh (and I don't mean broker is old technology ), skills, current issues, functionality etc. Basically what's the pain ?
And the migration will not be as straightforward as moving XML, would be wise to compare functionality & implementation of flows to how it'll be mapped into WESB.
Regards. |
|
Back to top |
|
|
fitzcaraldo |
Posted: Mon Jul 18, 2011 10:39 pm Post subject: |
|
|
Voyager
Joined: 05 May 2003 Posts: 98
|
Thanks elvis
What's the pain? Well the situation is they are running V5 WMB (yes, I know) and before upgrading to V7, they want to know if there's a viable alternative. Understand that they will effectively need to rewrite everything from scratch.
But putting aside the specific customer, do you think if we built a simple XML transformation in broker and WESB and then ran each on the same resource profile they would get comparable throughput?
Love to hear your other thoughts out of band. |
|
Back to top |
|
|
elvis_gn |
Posted: Mon Jul 18, 2011 11:42 pm Post subject: |
|
|
Padawan
Joined: 08 Oct 2004 Posts: 1905 Location: Dubai
|
Hi fitzcaraldo,
fitzcaraldo wrote: |
What's the pain? Well the situation is they are running V5 WMB (yes, I know) and before upgrading to V7, they want to know if there's a viable alternative. Understand that they will effectively need to rewrite everything from scratch. |
Rewrite everything from scratch to migrate ? Why ? I'd assume you'd build a better framework with all the features in v7, but why everything from scratch ?
If you move to WESB (or any other ESB), you'll definitely need to re-write due to the ESQL to any other language translation.
fitzcaraldo wrote: |
But putting aside the specific customer, do you think if we built a simple XML transformation in broker and WESB and then ran each on the same resource profile they would get comparable throughput? |
I feel for simple things the responses will be in milliseconds, no real performance difference...it's only when you scope your entire requirements that you'll see the difference. Best way to assess would be to provide all inputs to IBM and they can give you a sizing estimate on PVUs required for each product.
In general opinion, I'd say that a C/C++ engine will surely give you more performance than a J2EE engine.
Regards. |
|
Back to top |
|
|
rbicheno |
Posted: Tue Jul 19, 2011 4:12 am Post subject: |
|
|
Apprentice
Joined: 07 Jul 2009 Posts: 43
|
In my opinion they would be crazy to look to migrate to WESB when they are on WMB already. In my experience it has leading performance and has more connectivity options than WESB. Sure it costs more but you pay for what you get, looking just at the ticket price is a bit short sighted.
I would suggest they maybe look at tuning existing flows and reassess capacity planning to see if they can reduce costs by fitting WMB on smaller hardware or even virtualisation. If its a small implementation then they could even look at the starter edition:
http://www-01.ibm.com/software/integration/wbimessagebroker/starter/ |
|
Back to top |
|
|
lancelotlinc |
Posted: Tue Jul 19, 2011 4:51 am Post subject: |
|
|
Jedi Knight
Joined: 22 Mar 2010 Posts: 4941 Location: Bloomington, IL USA
|
When was the last time you went to McDonalds to buy the cheapest steak dinner? If I want a steak, I don't mind paying US$12 at Texas Roadhouse which is twice as much as US$6 for a Big Mac.
If the only criteria someone is looking at for their ESB is PVU cost, then they are rookies at running a business. PVU cost is only one element in the total cost of a system.
The return on investment for WMB is three years. The cost per transaction on a Power7 box for WMB is one-quarter of one cent (US) for a banking application. If your management doesn't understand rate-of-return and return-on-investment, they have no business making decisions.
Reminds me of Sheila Jackson Lee's statement about the US winning the Vietnam war and both Vietnams now live peacefully with each other. Do you want Sheila Jackson Lee making decisions on how to build the next Space Shuttle? Or, Al Gore designing your next toilet? or next TCP/IP stack?
Vitor says: Power7 ...! _________________ http://leanpub.com/IIB_Tips_and_Tricks
Save $20: Coupon Code: MQSERIES_READER |
|
Back to top |
|
|
smdavies99 |
Posted: Tue Jul 19, 2011 4:59 am Post subject: |
|
|
Jedi Council
Joined: 10 Feb 2003 Posts: 6076 Location: Somewhere over the Rainbow this side of Never-never land.
|
lancelotlinc wrote: |
The return on investment for WMB is three years. The cost per transaction on a Power7 box for WMB is one-quarter of one cent (US) for a banking application. If your management doesn't understand rate-of-return and return-on-investment, they have no business making decisions.
|
These facts are well known but a PHB somewhere has been told to reduce costs. They will (possibly) replace Broker with WESB and indeed reduce a cost.
The PHB gets his big fat bonus despite the fact that the cost on another line in the P&L have risen dramatically.
I have seen a situation where the company were total cheapskates. They had broker installed with a Capacity limited licence and then because broker was running flat out in it castrated form they threw it out as not being suitable for the job. They are still struggling to get its replacemnt talknig to SAP properly. _________________ WMQ User since 1999
MQSI/WBI/WMB/'Thingy' User since 2002
Linux user since 1995
Every time you reinvent the wheel the more square it gets (anon). If in doubt think and investigate before you ask silly questions. |
|
Back to top |
|
|
Vitor |
Posted: Tue Jul 19, 2011 5:12 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005 Posts: 26093 Location: Texas, USA
|
lancelotlinc wrote: |
If the only criteria someone is looking at for their ESB is PVU cost, then they are rookies at running a business. |
Or they're in the mainstream of management thinking. Who would sooner buy a Big Mac than a steak, claim credit for the saving and worry about being hungry later.
lancelotlinc wrote: |
PVU cost is only one element in the total cost of a system. |
Yes, but it's a $ amount and actually gets paid. This means it's a big, bright blip on the management radar.
Accepting that your management appear not to have such radar. How's the funishing of the ivory tower coming? _________________ Honesty is the best policy.
Insanity is the best defence. |
|
Back to top |
|
|
Vitor |
Posted: Tue Jul 19, 2011 5:20 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005 Posts: 26093 Location: Texas, USA
|
Vitor wrote: |
lancelotlinc wrote: |
If the only criteria someone is looking at for their ESB is PVU cost, then they are rookies at running a business. |
Or they're in the mainstream of management thinking. Who would sooner buy a Big Mac than a steak, claim credit for the saving and worry about being hungry later. |
Case in point - I just got toasted to a golden brown by a collection of middle & senior management complaining that the slow pace of delivery and large number of defects is (and I quote directly), "costing the company money". I rebutted that given the WMB development team is myself, an ex-COBOL mainframe developer and a .NET coder we were doing quite well and if they wanted things to improve they should get the others the training I've requested.
The Director of IT looked me in the eye and said, "we can't justify that kind of expenditure at the moment", and started talking about QA proceedures and code checking. I pointed out that if I had to QA & check the work of 2 people things were likely to slow down even more as I'd have no time for my own work.
Things spiraled down a bit after that. But it illustrates the disconnect between "cost" and "spending money" I think. _________________ Honesty is the best policy.
Insanity is the best defence. |
|
Back to top |
|
|
PeterPotkay |
Posted: Tue Jul 19, 2011 9:05 am Post subject: |
|
|
Poobah
Joined: 15 May 2001 Posts: 7719
|
IBM should just bump the price of WMB a couple of % points and have "free" annual or bi annual training including with each license. I think it would be in IBM's interest to have every user of its software be as effective as possible with the software. Less PMRs, better perception of WMB and what it can do, therefore a better perception of IBM, more IBM products sold due to the halo effect, etc. _________________ Peter Potkay
Keep Calm and MQ On |
|
Back to top |
|
|
Vitor |
Posted: Tue Jul 19, 2011 9:41 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005 Posts: 26093 Location: Texas, USA
|
PeterPotkay wrote: |
IBM should just bump the price of WMB a couple of % points and have "free" annual or bi annual training including with each license. I think it would be in IBM's interest to have every user of its software be as effective as possible with the software. Less PMRs, better perception of WMB and what it can do, therefore a better perception of IBM, more IBM products sold due to the halo effect, etc. |
_________________ Honesty is the best policy.
Insanity is the best defence. |
|
Back to top |
|
|
|