Sadly, I am no longer employed by IBM, so this blog which started as a team blog for the IBM Business Partner Technical Strategy Enablement (IBPTSE) Telecom team no longer represents IBM in any shape or form. As I became the Chief Telecom Architect for IBM for the WebSphere brand worldwide, I continued to write posts. The WebSphere brand merged into the new Cloud brand in 2015 and I retained the same role working with Telcos all around the world to design software solutions to solve their business problems. Now as an independent Telecom architect, my hope is to continue/resurrect this blog on Telecom business issues and technology.
Thanks for visiting. Please comment on posts and leave your thoughts.
The Industry Business Partner Technical Strategy Enablement (IBPTSE) team is focused on helping IBM's business partners in their technical strategy with respect to IBM Software - specifically WebSphere software. We focus on three undustries:
Media & Entertainment
Energy & Utilities
I guess you're wondering what to expect form this blog. Over the next few posts, I plan to talk about resources that are available to IBM Business Partners - specifically in our team's focus industries, some of the latest issues we are seeing in our industries, some discussion of IBM software technology and some thoughts on IBM's Smarter Planet initiatives.
In terms of introductions, our team has people around the world, with coverage of North America, Europe and Asia Pacific. I live in Australia and have team mates in China, Taiwan and India and between us, we cover the whole of Asia Pacific.
It occurred to me the other day, when talking to a customer about Web2.0 - the participatory web, that I have a great example at home - I don't need to talk about Youtube, Flickr or Wikipedia - my five year old son is a great example of Web 2.0 in action. . My son Max - like many boys of his age is a big fan of the cartoon character Ben Ten. At home, he watches Cartoon Network to get his Ben Ten fix. In conjunction with the TV show, Cartoon Network have a number of games available on their web site - for them it is all about encouraging their viewers to keep watching and the way they do that is to offer games based on their shows on their web site to encourage more intense interest in the shows. They have recently launched a game creator which allows their web site users to build their own Ben Ten games. Max (my son) loves the game creator. It enables him to build his own games using a shockwave interface, then share that game with other users of the Cartoon Network web site. . That is the perfect example of the Web 2.0 concept of the participatory web. Max has ok computer skills for his age, but he still has a long way to go, yet he is able to and really enjoys creating his own games. More than that, he loves sharing his creations with others. That sort of participation, sharing and creation is exactly what Web 2.0 is all about. . I am such a proud dad! . If you want to try it our for yourself, this game is available at http://gamecreator.cartoonnetwork.com.au/?id=141081 .
. Some important things to note: The games are rated and stats are recorded on the number of attempts and time played - clicking on the "share this game" link add to the share count - Max doesn't (yet) have many friends with email accounts, so that's not a big deal to him, but older kids (and me!) find that a useful capability to share original games around... . What can we learn form this lesson as related to Telcos? Well, here are some things that I've learnt from my son:
Make it easy - in the telco space the closest Web 2.0 equivalent we have is the Mashup Center. Frankly, I think it is pretty easy to use, so I think we're doing ok on that score
Provide a rating capability - Max loves it when his games get blue balls (the rating visual that Game Creator uses) - likewise, the Widget library in the Mashup Center has this capability
Provide usage stats - I think it's really interesting to see which of Max's Games are getting played (let alone being voted for). I am not sure if Mashup Center or the Widget Library does this or not. If not, I think it would be a good addition.
Relate the participation back to your business - For Cartoon Network, that's all about getting web users to watch the show through getting players excited about the characters. For Telco's the Mashup environment should also encourage users to use Telco services and think of that telco as more than just their carrier, but their technology partner for the future...
So, I've found a real world example that I can now use in my Web 2.0 for Telco presentations... :-) .
This article describes how you can develop an offline charging application using the Rf interface in IBM® WebSphere® IP Multimedia Subsystem (IMS) Connector V6.2, presents a sample asynchronous offline charging adapter to enable multi-threaded throughput of the Rf client, and discusses performance tuning based on the Rf interface. .
WebSphere IP Multimedia Subsystem Connector V6.2 (hereafter referred to
as WebSphere IMS Connector) is an important component of the IBM
Service Delivery Platform for Telecommunications. In the IP Multimedia
Subsystem (IMS) architecture, the WebSphere IMS Connector connects SIP
applications with IMS core elements and provides functions of offline
charging (through the Rf interface), online charging (Ro interface),
and subscriber profile management (Sh interface).
article describes how to develop an offline charging application,
leveraging the Rf interface in WebSphere IMS Connector V6.2. An
asynchronous offline charging adapter that implements an asynchronous
callback interface to enable multi-threaded throughput of the Rf client
is then presented. This discussion concludes with a look at performance
tuning with WebSphere IMS Connector.
This article assumes a basic understanding of the IP Multimedia Subsystem,
Diameter protocol, Java™ programming, and Web services standards.
It struck me today that many of our business partners, at least the ones I deal with, don't have the foggiest what IBM offers to them in terms of online resources, assistance (paid or free), demonstrations available publicly etc. As I was preparing for a presentation on the subject in the Global and Medium System Integrator Telecom trianing that we ran in Malaysia a couple of months ago, I thought it would also be worthwhile sharing it with a larger Business Partner audience as well... . I've uploaded a file to my collection on Lotuslive - it is somewhat biased toward AP in terms of listed resources in the area, maps on slides etc, but could be easily made more specific to another geography pretty easily... . . .This is the basic agenda and flow:
Partner Programme Management
Partnerworld Industry Networks
Virtual Innovation Centre
IBM Developer Relations
Logging a call – types of calls etc
Business Partner Technical Strategy Enablement team
I was at a workshop with a customer in Manila recently when they started to talk about compression over a client link (especially from
Nokia S60 Mobile Phones) - a key value proposition of Lotus Mobile
Connect. Not since I was in a Pervasive Technical Sales in Australia / New Zealand had I
seen an opportunity for a hosted Lotus Mobile Connect (LMC) deployment. For those of you
that weren't aware that LMC supported a hosted deployment - it does.
If you have the Mobility Client installed (The client for Lotus Mobile Connect - on any platform) you will notice a field labelled "Organizational Unit" - ever wondered what that is for? It's simple really. It is there so than in a hosted deployment, the LMC authentication mechanism is able to distinguish between "John Smith" at Company A and "John Smith" at Company B.
Typically, you would use Tivoli Directory Integrator (TDI) to enable a federated directory model so that the individual client companies can manage their own internal directory - and because TDI uses LDAP to communicate with those directories, it doesn't matter what those client directories are (Domino, MS Active Directory, Sun Directory Server, Novell Groupwise Directory, openLDAP etc) - as long as they support LDAP V3.
Basically, there are two deployment topologies that enable LMC to be
deployed in a hosted environment... (I have deliberately left TDI out
of the diagram since the purpose of the following diagrams are intended
to illustrate the Client options and Encryption break or end-to-end)
Secondly, there is a lower security (and cheaper to deploy) topology
that still gives the end users the advantages of LMC, but without
end-to-end encryption - this model requires that the client companies
trust the Telco since there is a break in encryption at the Telco's
hosting centre. This model would not be suitable in high
security/privacy industries such as Finance, Health, Government,
military or Emergency Services.
A Telco might offer the lower security model as their standard product
and the end-to-end model as the premium service with a price premium...
This is a potential salable product to a Telco's enterprise customers in it's own right, but if we look at the offering that also come from LotusLive ([particularly LotusLive Notes and LotusLive iNotes). In a market like the Philippines or many others across Asia, I suspect there is a business to be made by offering Domino capabilities or even just plain old hosted email but kept separate from the masses of a Telco's standard data customers who all get an email address like email@example.com. Using LotusLive Notes or iNotes would allow a small business to maintain their own virtual email system and keep their own email domain, internal email addresses but without the headaches of looking after their own servers. If we think about the LotusLive offerings in a Telco - where the LotusLive products are rebranded to suit the telco, they could easily go along with a hosted LMC offering. This would provide secure access by remote or mobile users to their own network and their own virtual email environment.
I had hoped that for the LotusLive deployment of Domino
in LotusLive notes that some code changes had been made to make Domino
work in a multi-realm environment - alas, no. Consequently, there is a
minimum customer deployment size of 1000 users - way bigger than most
Telcos would be looking for and way too big for the Philippine market.
As it stands, LotusLive iNotes is not much better at 500 users, but it
so far looks like that is an IBM decision and that if the Telco is to
take on the level one support, then it would be up to the Telco to
decide what the minimum customer size is to be. Indeed, Some legacy Outblaze (from
whom we bought assets from to deploy LotusLive iNotes) customers have some
ISP/ASP customers that resell their service to end customers with 5,10
or 20 users.
Perhaps a diagram is in order to explain who it might all come together. I have refined my diagrams that illustrate the hosted deployment of LMC
with LotusLive iNotes (or ANY LotusLive product for that matter -
Engage, Connections, Meeting etc). First, the Premium offering:
Or in a slightly less secure deployment (with a break in Encryption at the Telco - probably not acceptable for a Bank or Government department, but fine for may smaller businesses) :
As I see it, a Telco offering this type of service could charge a premium for the end to end encryption model while the second model might be a cheaper service.
As an adjunct to the LMC and LotusLive iNotes offering, a Telco might
also offer Lotus Foundations for an on-premise offering to SMBs. I am
not sure if Foundations will interest every Telco, but we already have
some success with Telco sold Foundations in Singapore.
]If you are interested in understanding this hosted model for Lotus Mobile Connect (LMC) or LotusLive iNotes, please let us know... It could make for an interesting series of blog posts
I am working with a number of IBM business partners and I found a need to explain to them how our Software licensing works. I found that many of our sales staff don't fully understand it either, so I figured I would post the explanation I wrote for the business partners to try and explain it so more people "get it". The other thing that struck me in speaking with some partners was that - despite some of them them partnering with Oracle more often than they have with us in the past - they had a simplistic view of Oracle's licensing thinking that it was simply CPU based. Oracle's licensing scheme is similar to our own PVU scheme in weighting different multi-core CPUs differently for licensing purposes.
First - IBM's PVU scheme
The majority of the IBM runtime components are priced per PVU. The Processor Value Unit or PVU is an arbitrary notion that IBM came up with to cater for multi-core CPUs and the fact that some platforms offered more processing power per CPU core than other platforms. Different brand processors cores are considered equivalent to PVU counts from 30 PVUs to 120 PVUs per core.
For example, an Intel single-core CPU is 100 PVU. Intel multi core CPUs are considered to be equivalent to 50 PVUs per processor core (or 70 PVUs per core for the newer Intel chips), so a dual core CPU would be 100 or 140 PVU and a quad core CPU would be 200 or 280 PVU. Prior to the latest generation of Intel multi-core CPUs, Intel multi-core architecture was such that a single dual core CPU offers similar processing power to a single core CPU, so to be fair to customers that use Intel multi core CPUs, IBM only rates each core at 50 PVUs. The latest chips have improved their processing power per core over previous generations of chip and they are now rated at 70 PVUs per core as a result.
IBM PowerPC chips are more efficient and therefore the PVU rating per CPU core is 80 PVU per core for Power 6 blades although other PowerPC CPUs are rated at 50, 100 or 120 PVUs per core.
"Processor: shall be defined as all processors where the Oracle programs are installed and/or running. Programs licensed on a processor basis may be accessed by your internal users (including agents and contractors) and by your third party users. The number of required licenses shall be determined by multiplying the total number of cores of the processor by a core processor licensing factor specified on the Oracle Processor Core Factor Table which can be accessed at http://oracle.com/contracts. All cores on all multicore chips for each licensed program are to be aggregated before multiplying by the appropriate core processor licensing factor and all fractions of a number are to be rounded up to the next whole number. When licensing Oracle programs with Standard Edition One or Standard Edition in the product name, a processor is counted equivalent to an occupied socket; however, in the case of multi-chip modules, each chip in the multi-chip module is counted as one occupied socket.."
This basically means that for Intel quad core CPUs, they are priced at twice the price of an Intel Single core CPU (a multiplier of .50 per core) - exactly the same as IBM pricing for Intel Quad core CPUs. Likewise, for PowerPC (Po dual core CPUs, they apply an factor of 0.75 since they do not differentiate between the processing power from other manufacturers other than Intel, AMD or Sun and just apply a generic multiplier of 0.75. Oracle have introduced a more comprehensive factor table to calculate their per CPU licensing price (introduced in March this year I think) where they added multipliers of 0.5 and 1.0 to their table. Oracle's core factor table is available at http://www.oracle.com/corporate/contracts/library/processor-core-factor-table.pdf
To illustrate, if the Oracle product license cost is $100 per CPU and the IBM price is $1 per PVU, then the following table illustrates how Oracle and IBM pricing will change depending on the processor that software is deployed on.
Assuming the base software price is $100/CPU (Oracle) or $1 per PVU (IBM)
Oracle Cost calculation Price x RoundUp(CPU cores x multiplier)
Oracle Extended software cost
IBM PVU rating (PVUxCPU-cores)
single core CPU (any)
100 x 1
Intel/AMD Quad Core(older)
100 x RoundUp(4 x 0.5) = 100 x 2
Intel/AMD Quad Core(new)
100 x RoundUp(4 x 0.5) = 100 x 2
Sun UltraSparc T1 Hexa-core(1.0 or 1.2 Ghz)
100 x RoundUp(6 x 0.25) = 100 x 2
Sun UltraSparc T1 Hexa-core(1.4 Ghz or higher)
100 x RoundUp(6 x 0.5) = 100 x 3
Sun UltraSparc T2 Hexa-core
100 x RoundUp(6 x 0.75) = 100 x 5
IBM PowerPC Dual Core POWER6 (520, JS12, JS22 servers)
100 x RoundUp(2 x 1.0) = 100 x 2
IBM PowerPC Dual Core POWER6 (550,560,570, 575, 595 svrs)
100 x RoundUp(2 x 1.0) = 100 x 2
IBM Power5 Quad Core
100 x RoundUp(4 x 0.75) = 100 x 3
This illustrates that both IBM and Oracle understand that not all multi-core CPUs are created equally - some are more like multiple single core CPUs just placed on a single die. It also shows that Oracle and IBM both understand that CPU architectures such as the SunSparc and Intel/AMD x86 offer less processing power per CPU core that IBM PowerPC architecture.
Lets dispel the myth that Oracle price per CPU only - their multipliers provide a similar pricing strategy to IBM's PVU based pricing - sometimes IBM has the price advantage, sometimes Oracle has the price advantage. Oracle first introduced this type of multi-core licensing back in 2005 although back then the multipliers were set at a generic 0.75 per CPU core for all processor types - regardless of CPU processing power.
Note - as both Oracle and IBM have the right to change their pricing at any time, I can only vouch for the accuracy of this post at the time it was originally posted (Nov09).
For the second year in a row, IBM AIX UNIX running on the Power or “P” series
servers, scored the highest reliability ratings among 15 different server
operating system platforms – including Linux, Mac OS X, UNIX and Windows.
Those are the results of the ITIC 2009 Global Server Hardware and Server OS
Reliability Survey which polled C-level executives and IT managers at 400
corporations from 20 countries worldwide. The results indicate that the IBM AIX
operating system whether running on Big Blue’s Power servers (System p5s) is
the clear winner, offering rock solid reliability. The IBM servers running AIX
consistently score at least 99.99% or just 15 minutes of unplanned per server,
per annum downtime.
A South African information technology company proved it was faster for them to transmit data with a carrier pigeon than to send it using Telkom, the country's leading internet service provider.
Internet speed and connectivity in Africa's largest economy are poor because of a bandwidth shortage. It is also expensive.
Local news agency SAPA reported the 11-month-old pigeon, Winston, took one hour and eight minutes to fly the 80 km from Unlimited IT's offices near Pietermaritzburg to the coastal city of Durban with a data card was strapped to his leg.
Including downloading, the transfer took two hours, six minutes and 57 seconds -- the time it took for only four percent of the data to be transferred using a Telkom line.
Okay, it was a bit of a stunt. I am sure if I posted a 32Gb SD Card to the Sydney (standard mail service- often next day delivery, but sometimes the day after that), it would arrive faster than I could transfer that content from my home office. What does that prove in terms of available bandwidth? Not much really - SD cards can hold an incredible amount of information these days. I have worked with customers in the past who shipped hard-drives around when they needed to transfer large amounts of data - even today - on most networks, it would be faster to courier a 1Tb HDD anywhere in the world than to transfer that much data over the wire.
The article did get me thinking though. I travel quite a bit around Asia and have experienced first hand the speed of networks in many countries. I've seen networks slower than a dial up modems (in Vietnam IBM Office) - in fact I reckon that my mobile phone as a modem over an EDGE connection (3G in Vietnam is very patchy) would have been faster than the IBM office network connection. This is not a unique situation - in many countries I visit, the network speed is faster in my hotel than it is at the local IBM office.
How does this effect the way we behave? Lets look at a specific example. Last year, I was doing a lot of work for the Globe Telecom SDP project that we eventually won with NSN in the Philippines. I was using Cattail (an IBM Research project for sharing files - similar functionality to the Lotus Connections Files capabilty that we now have in MyDeveloperworks) to upload files so that the local IBMPH IBM team could get to them rather than clog up their mail boxes. Smart - or so I thought. With Cattail, you are able to see who is downloading your files - often quite interesting as it was in this case. I noticed that only one person in the Philippines was downloading the files, despite notifying about 12 people that they each needed to look at the content. After a while I asked this one person why no one else was downloading the files from Cattail - he told me that because the network was so slow, most people were unable to even load the Cattail page to begin the download, so he went through the pain for everyone, then emailed the files around the local team! So much for not clogging up their mail files.
I am constantly frustrated by the US centric assumption that the whole world has the same bandwidth available to them as they do. Even in Australia, I am paying AU$68 per month for 12Gb of traffic - typically around 2 Mbps actual (10Mbps claimed capacity) downstream and 250 kbps actual upstream. By US standards, that must seem slow, but by the standards of developing nations in ASEAN, that's pretty darn good. There is still a huge digital divide between the haves (the US) and the have-nots (developing nations) - while some countries will have fibre to the home deployed (or being deployed) over the next few years - Singapore will be done very quickly I anticipate - I wont have that sort of speed available to me until 2012 the Australian federal government claims (I expect it will be more like 2020 though as I do not live in the inner suburbs of Melbourne)
So, what point was I trying to make? I am not sure. I am frustrated at my bandwidth sometimes (usually not) but in countries that I visit, the whole nation must feel frustrated. I often see web pages sizes in excess of 500kb - a ridiculously large size and unusable in most of Asia. Application designers need to be mindful of the bandwidth availability if they hope to be successful in Asia. If you have thoughts, please comment...
Yes, our team is focused on SDP, but this article was interesting because it is in our part of the world (I live about 45kms from it) in AP and Telstra make extensive use of Netcool in their Network Operations Centre. I wonder if the folks with only two screens suffer from 'screen envy' when so many others have four screens?
Verizon Wireless, AT&T and several major international carriers and vendors threw their support behind an IMS-based approach to delivering voice and SMS services over LTE networks. The level of operator support--the approach also is supported by Orange, Telefonica, TeliaSonera and Vodafone--sits in sharp contrast to another approach, called Voice over LTE via Generic Access, or VoLGA, which is supported by T-Mobile International.
Vendors including Alcatel-Lucent, Ericsson, Nokia Siemens Networks, Nokia, Samsung Electronics and Sony Ericsson also voiced their support for the initiative, dubbed One Voice. The companies said they concluded that an IMS-based approach "is the most applicable approach to meeting the consumers' expectations for service quality, reliability and availability when moving from existing circuit-switched telephony services to IP-based LTE services. This approach will also open the path to service convergence, as IMS is able to simultaneously serve broadband wireline and LTE wireless networks."
The companies said that the purpose of the initiative is to create the largest LTE ecosystem possible, and to avoid fragmentation of technical solutions.
Interestingly, both Alcatel-Lucent and Ericsson also support the VoLGA approach, and Nokia Siemens has supported its own solution, called Fast Track Voice, which proposes having mobile switching center servers handle VoIP traffic over LTE networks. VoLGA proponents argue that their approach should be used as an interim solution. All three vendors said they do not see a conflict in supporting the different approaches.
Is it just me or when you read "VoLGA" do you think "Vulgar" - I think the ALu acronym police need to get out from behind their desks and make an arrest for that one!
I've drawn up a representation of the situation as I see it. NSN going it alone with Fast Track Voice, almost everyone else supporting VoLGA and planning to move to OneVoice.
The thing that I find really interesting is the inference by ALu that they will continue to stand by VoLGA as well as support it as an interim step to OneVoice while NSN seem to be saying that Fast Track Voice is only an interim step on the path to OneVoice. It's also interesting to note that the VoLGA consortium seems to be mainly Network Equipment Providers (NEPS) while OneVoice is bot NEPs and Telcos.... I suppose the most appropriate message is "watch this space"...
PS. On rereading this post, I imaging some of you are going 'Huh?' I
apologise for the Telco jargon. Let me take a moment to try and
explain some of the terms that appear in this post.
IMS - IP Multimedia Subsystem (not
IBM's mainframe database that helped put man on the moon). This is a
specification controlled by 3GPP (a Telco standards body)
to describe a next generation IP based telephony environment. Most
telcos today still run a legacy switched environment based on very
specialised protocols such as SS7 and Sigtran. These protocols are not
IP based and as such require very specialised (read expensive) skills
to work with them. The other thing is that they are not really
standardised - each NEP has their own version of the SS7 protocols.
IMS promises to bringmuch cheaper skills and shorter development cycles to the Telcos core platform - something they have not had before. IBM has a number of products that are targeted at telco's IMS infrastructure (WebSphere IMS Connector, WebSphere Presence Server and WebSphere XML Document Management Server)
LTE - Long Term Evolution is seen by most NEPS as the next logical evolution step for carriers with GSM networks.
That evolutionaty path goes something like this:
GSM->GPRS->EDGE->UMTS->HSDP->LTE. LTE promises to
deliver high bandwidth mobile connections. The main rival to LTE is
WiMax which you may have heard of before.
ALu - Alcatel Lucent (a very common abbreviation for the joint company)
I had a request on the other week to create a number of topology diagrams that showed how a Telco might start small and grow their environment to add new capabilities and services. This was specifically for a telco in Vietnam, but I figured it would make sense to generalise the presentation and the images to make it usable for other opportunities. We've had a similar request from other telcos recently as well. The presentation step through 11 phases from a pilot/trial environment through to a full blown system. Each slide has speaker notes explaining what is being added at each phase in terms of products and capabilities. This presentation is not meant to make any recommendations on how to evolve form a small system to a more complex and capable one. What it is supposed to illustrate one possible evolution... Note that it focuses only on the IBM components and some other components would also be required for some phases (such as a transcoding engine in the media extension phase).
Below are three of the diagrams - Phase 1, Phase 6 and Phase 11 and the speaker notes that go along with that phase - to give you a feel for the flow...
Phase 1 - Test Environment
At this first stage, an initial deployment might be considered a proof of concept or a trial – which could become the test and or ISV environment, The functions that this could offer are:
Composite applications that bring together functions provided by the network. For instance an application that consumes SMS messaging and integrates the location of the handset into an app.
WSRR will get them down the path of SOA Governance – it is important to get this in early to ensure that the governance model is maintained and the Telco will now need to rework services that are created at this stage.
Complex workflows and business processes can be built which include human tasks (such as prototype processes for the production implementation )
Phase 6 - Developer Ecosystem including Web 2.0
Phase 6 introduces the Developer Ecosystem components such as :
Idea Factory for Telecom – which will help make a dispersed group of developers into a community. It enable the sharing of ideas and a framework for the Telco to manage the evolution of the ideas that are generated within the community. It also provides a rapid prototyping capability via...
IBM Mashup Canter which allows users to drag widgets onto a workspace and simply wire them together. It is both the development and the runtime architecture. This means that developers don't need deep development skills in order to build new applications.
WebSphere sMash which provides a PHP and Groovy scripting environment (both development using the Dojo toolkit and the runtime environment)
This combined with the web services exposure deployed in phase 4 means that the developer ecosystem can now cater for all levels of developers – those with no skills can use the drag and drop mashup environment, script developers can use sMash and more advanced developers can use the web services interface. In the backup slides there is an illustration of this.
For advanced developers the Telco can support developers across a range of IDEs ranging from Rational and Eclipse (where we have Telecom Toolkits available for free) to other IDEs (such as Microsoft Visual Studio or Sun Netbeans) where the IDE has tools to assist developers with consuming web services. In all the IDEs, developers will consume the Web Services Description Language (WSDL) file from a UDDI directory in the DMZ. The UDDI directory (part of WPS) is populated from the WSRR internal services repository.
Phase 11 - IMS integration and extension
When the Telco goes down the IP Mulitmedia Subsystem (IMS) path, the software deployed already has IMS enablement, but at this point we can also add WebSphere Presence Server (PS) and WebSphere XML document Management Server (XDMS – formerly WebSphere Grouplist Manager) which provides IMS services for the IMS services plane. The core infrastructure that was deployed way back in phases 1 and 2 are critical to the IMS Services plane.
It is important to understand that the phases I have split them down into are purely arbitrary and are not necessarily what would happen in a real telco. Which function occurs at what point and in combination with other functions is something that must be driven by the business requirements of the telco. The intent is to illustrate how a telco could start small and add function incrementally building on the previous investments. Still want it? Great - feel free to download it from MyDeveloperworks files. Please let me know what you think.
@Alex, regardless of why Oracle chose to compare the RISC and CISC
platforms, the result is not meaningless. Java App Servers are
licensed on cores that they run on. For most companies, bang for
buck is a key measure. As the comparison is JEE compliant Java App
Servers, we can legitimately compare app servers running on
different underlying hardware architectures and come up with a
legitimate measure of bang for buck.
Let's look at IBM's and Oracle's licensing for this benchmark; the
WebLogic instance would require 24 CPU equivalent*** (48 x 0.5)
licenses of Weblogic App Server (Enterprise Edition) while the
Power system will require 480 Processor Value Units (4x120) of
WebSphere Application Server Network Deployment. To compare the two
models if we take the list price for WebLogic Enterprise Edition
(US$30,500* per CPU equivalent) and IBM's WebSphere App Server ND
(US$174/PVU*) then we see that the Oracle WebLogic App
Server license (US$732,000) will cost 776% more than the IBM
WebSphere App Server license (US$83,520).
Oracle are claiming "...nearly 7 times..." the performance despite
the fact that 9,455.17/1,197.51 = 7.90 (to 2 decimal places) which
in my book is nearly 8 times the performance, not nearly 7 times. I
think their marketing people got their percentages mixed up - 7.90
times the performance of the IBM score is a 690% improvement on the
So, let's give them the benefit of the doubt and assume the Oracle
marketing folks made a mistake and their benchmark system can
deliver 7.9 times the performance of the IBM
benchmark system, they are doing it for 8.7 times the
price of the IBM system in terms of app server licensing,
that is not looking like the spectacular win that Oracle are
claiming it to be... In the bang for buck war (at least in
software licenses), IBM still wins.
In response to the presentation I built (and blogged about - see Telecom Systems Evolution) a number of IBMers asked for costings for each phase. While that would be possible at the early stages where you don't have to take into account scaling, later in the phases, that really does become impossible. For instance lets say phase 4 was US$1m for Telecom New Zealand (with about 2 million subscribers and a heavy post paid mix of subscribers) and compare that with Bharti-Aitel in India with 120 million subscribers and a heavy pre-paid mix. Because of the variations in telcos around the world, there is no way any pricing that was added to the presentation would be in any way applicable - that is something that has to happen on a per telco basis.
That said, we are working on a "starting point" or "lite' implementation and wrapping some costs around that - we think this would be a good starting point for many SDP projects - kind of a standard phase 1. The architecture would be built with expansion in mind even though at phase 1, the components would not be excessively sized. So far, we have a near final Bill of Materials and a number of diagrams. I thought I would share those diagrams with you here. We still need to document all the assumptions and limitations that this "SDP Lite" phase 1 architecture would entail.
I would appreciate feedback on these diagrams, so please comment.
The IBM Telco specialists should recognise this - it is the Service Provider Delivery Environment Version (SPDE) 3 - IBM's industry framework for Telecoms. I included it here for reference and comparison with the follow diagram that illustrates the areas of impact that SDP lite will have on SPDE 3.0.
Now, see the areas of impact that SDP Lite has on SPDE 3.0 - the Orange shades indicate areas of impact. Iif we map the IBM products in the Bill of Materials over the top of that, we get...
That's it with the logical views of SDP Lite. The next one is a marketecture diagram to help explain the key principles and functions of SDP lite. I am only showing the production environment in this diagram, but you could also have a separate environment that is a duplicate of this that could be dedicated to both test and ISVs.
For those that want to understand the Deployment units and their basic layout, I have a deployment unit diagram
Finally, an illustration to show how the SDP Lite infrastructure could support a developer ecosystem - the shaded components could be added in subsequent phases - for now, it is all about secure exposure of web services and REST interfaces to a developer ecosystem that is using conventional Integrated Development Environments (IDE). Note also that the hosting servers for the developer applications are out of scope and could be hosted anywhere on the Internet. . Keep in mind that this is supposed to provide a starting point - or initial deployment. Infrastructure that could grow, but could be used immediately for smaller trials. With that in mind, in estimating the performance, this is what we get:
This SDP Lite infrastructure would allow a telco to begin offering a range of new services and products as well as form the basis for a larger and more functionally rich topology later on. Some of the use cases that come to mind immediately include:
Service exposure to 3rd party ISVs - Web Services and/or REST. This represents a new revenue opportunity for many telcos.
Build composite applications that consume and combine existing services to automate processes like retailer commissions
Marketing campaigns based around bundled products. For instance: top up your prepaid mobile broadband and get unlimited text messages for 24 hours on your mobile - Globe Telecom in the Philippines are doing this exactly - check out the following Globe TV commercial:
Begin to build a developer ecosystem around the exposed services
Build composite applications that consume and combine existing services to offer new products to subscribers such as
Family Finder - Allow parents to see where their kids are and where they have been
Meet on Click - Friends can see where each other is, what is on locally and send invitations to catch up via SMS, MMS or PushWAP
Emergency notifications - Notify everyone in a specified geographical area of some danger
Location based marketing - Send SMS/MMS/PushWAP messages to subscribers who have oped in when they get within 500m of a retail outlet.
lots of others.....
50% CPU Maximum
JS12 Blades (one for TWSS Access Gateway, One for TWSS Service Platform)
SMSM via SMPP
PushWAP via SMPP
MMS via MM7
LBS via MLP
Parlay via CORBA
Presence via SIP
IMS/VoIP via SIP
Expected max transaction rates
100 TPS* for SendSMS via SMPP
10 TPS* for SendMMS via MM7
Other transactions could be added into the mix, but would lower the SMS or MMS transaction rate.
50% CPU Maximum
80% Shortlived transactions
20% Long Lived transactions
Expected max transaction rate
20 TPS* of the above transaction split
50% CPU Maximum
JS12 Blades (one for WebSeal, One for TAM Policy Manager)
Expected max transaction rate
DB2 Enterprise Edition
50% CPU Maximum
Minimum of 12 Disks in RAID 1+0 Array
So, assuming that both internal and external users/systems were using the system concurrently, we could support up to 40TPS from external developers (limited by TAMeb) which if we assume they are using 90% TWSS services and 10% composite services will mean that internal systems would have 74TPS of TWSS services and 16TPS of WPS services available if the developers are consuming 50% of the TAM CPU.
for Internal Use
(32.7 TPS of SendSMS + 3.2TPS of SendMMS)
(67.3 TPS of Send SMS + 6.8 TPS of SendMMS)
(3.2 shortlived + 0.8 longlived TPS)
(12.8 TPS shortlived + 3.2 longlived TPS)
These are really rough numbers and I would like to add some more assumptions around then. Of course, if your Telco won't have a developer programme, then 100% of the transactions would be available for internal consumption.
What do you think? Are we on the right track for your Telco customer? Would you like to see some changes?
Wow! Optus in Australia are blocking their own Android subscribers from buying apps from Google's App Store - only permitting subscribers to buy apps from Optus's own app store. Imagine the uproar if they did the same thing for the Apple iPhone and Apple's app store!