Karaoke in the Cloud, IT Infrastructure and the Next Generation Network

August 14, 2009


–  Sung by Ishikawa Sayuri

In this karaoke song Ishikawa Sayuri conveys the essence of the Japanese heart.  The spring, lying-down under nocturnal cherry blossoms, hanami zake, and that special ‘anata’ indeed explain why she feels being born in Japan is so good!  The hanami (cherry blossom viewing) fever sweeps through Japan from the southern part to the northern part during the period from January to May.  The cherry blossoms last only a short period of time and the delicate flowers are viewed as a metaphor for life itself depicting its fleeting and ephemeral nature and are used as an excuse for drinking sake and partying.  The collective nature of Japanese society becomes highly visible during hanami parties where the parks are decorated with temporary paper lanterns to celebrate yozakura late into the night and groups gather under the cherry blossoms with drinks and dango (dumplings).

Karaoke is as Japanese as hanami and captures another aspect of cultural essence where technology, business, social context and human element are closely interwoven.  During the heyday of Japanese business leadership, Karaoke ranked at the top with 43 percent, followed by traditional hobbies like tea ceremony, flower arrangement and horticulture with 32 percent, and creative arts such as drawing, photography and sculpture with 24 percent. (Japan Times, 4 November 1993) [1]. A wide variety of venues are available for Karaoke making it accessible to mass consumer market.  A part of “mizu shobai” (unstable business dealing with unpredictable customers, such as bars), Karaoke bar has become a component of Japanese corporate culture.  Japanese management cultivates special karaoke bars for entertaining their clients and creating a “safe” place where business is conducted.  It is usual for a buchou or a shachou to establish a special relationship with a Mama San and conduct his business for both entertaining clients and to discuss with his group in a relaxed atmosphere.  One Japanese manager of a large corporation warned me “Rao San, be careful on what you say to Mama San tonight.  She has e-mail connection with the shachou”.  A senior buchou uses karaoke bar to train his junior members (senpai and kohai relationship), listens to their complaints without judgment and gathers information from colleagues.

In the past, Karaoke and Love Hotels in Japan provided high technology impetus to continually improve the “experience” for their customers.  The ubiquity and demand from all strata of society provide a mass consumer market for the karaoke service and Karaoke service could be a natural candidate for the emerging cloud computing technology.  A network centric virtual karaoke bar could extend the reach globally.

“The service in this example enables an individual to throw a network-based karaoke party, in which the invitees can participate from their own homes. The host issues electronic invitations and reserves the time for the party. Before the appointed time, the host will choose the service providers, browse the clips and songs, and choose to create a customized medley of songs and sceneries for the party. The service is billed on the basis of actual time used and allows suitable discounts for longer periods and a larger number of participants. It also charges for optional services, such as the ability to add personalized digital images or video clips to the sceneries. It also allows discounts if the service providers are allowed to inject commercial messages or clips into the party.  At the appointed time, invitees can join the karaoke party, and their images are mixed, in real time, with the chosen backgrounds in a smooth fashion. The singers can be chosen and the songs selected with intuitive user interactions. When a song begins, the singers and the sceneries can be edited and blended in real time to give a seamless image that transcends the boundaries of space and time for a short period. This simple service brings out key requirements for the grand convergence, in which computing, storage, and networking must blend in a way that provides a thrilling experience for the participants.  Obviously, the same requirements will make a host of other services, such as business videoconferencing, data sharing, and so on, possible.”  This service was described by the authors [2] in the year 2002 during the heyday of Internet service optimism and before the Internet bubble burst.  Many advances have occurred but the technologies are still far away from implementing the real-time blending of singers and the scenery alluded to in this article and required telecom grade service features.  The closest service that has recently been introduced by Animoto and is successfully duplicated by Microsoft is the service that blends pictures and music to create a video.

Obviously this service has broader applications such as providing massively scalable and globally interoperable virtual Karaoke bars, interactive video conferences and entertainment industry infrastructure.  However recent economic down turn and the graying of Japanese society (where 21.5 percent of its population is 65 or over) are having a major impact on the future of Karaoke:

1. On one hand, the dwindling profit margins and tight corporate budgets are taking a toll on Karaoke bars by limiting the clientele to physicians, independently wealthy folks and a few high level business executives.
2. On the other hand, a new set of “day-time” Karaoke bars are springing up in every town in Japan catering to grandmothers and grandfathers who pay $10 for half a day while enjoying singing and conversation.  My friend says that his mother, already 84 years, enjoys this kind of Karaoke two or three times per week in a small town.

Be that as it may, the Karaoke in the cloud or for that matter, any other service in the cloud should provide end-to-end, generic requirements for the computing cloud infrastructure, service assurance, service creation and service delivery just as the simple voice service provided the requirements for AT&T Bell Labs researchers who defined Local Switching System Generic Requirements (LSSGR), Operation Support System Generic Requirements (OSSGR) and Total Network Operations Processes (TNOP).  Whole industries were created with these requirements in the past.

Any generic cloud service requirements must address four stake holders:

1. Infrastructure developers who provide the computing, network and storage resources that can be dynamically controlled to meet massively scalable and globally interoperable service networks with varying workloads and business priorities.  The infrastructure will be used by both service creators who develop the services and also the end users who utilize these services.  This is very similar to switching, transmission and access equipment vendors incorporating service enabling features and management interface in their equipment. Current storage and computing servers lack dynamic dial-up and dial-down features and automated management capabilities which eliminate human latency along with a patchwork of legacy management systems.

2. Service providers who assure both the service developers and the service users that the resources will be available on demand, they can pay based on usage and the resources will be managed to provide Service Level Assurance to meet availability, performance and security required by each service.  The managed resources that the service provider provides are confined to CPU, memory, network bandwidth, storage IOPs, throughput and capacity that can be dynamically dialed up or down on demand with specific requests from the resource users.  This constitutes a computing services dial tone with availability, performance and security Service Level Agreements (SLAs).  In addition the service provider must provide both visibility and control to end users so that they can manage their computing resources to create logical servers with virtual storage to meet their service latency and business requirements.  The service provider manages the application to computing, network and storage resource connection with appropriate SLAs.  This is a totally different model from most cloud computing solutions that are nothing more than hosted infrastructure or applications accessed over the Internet.

3. Service Developers of cloud based services will use   the management services API to configure, monitor and manage service resource allocation, availability, utilization, Performance and security of their applications in real-time.  Service management and service delivery are integrated in the application development so that the application developers will be able to specify run time SLAs.

4. The end users of cloud based services demand choice, mobility and interactivity with intuitive user interfaces [2].  The managed resources allow the service developers to create virtual services using logical servers that meet appropriate SLAs and provide services to end-users and charge them based on usage.   The service providers must also assure service levels for end users.

Figure 1 shows the reference model [3] that shows three stake holders involved in computing clouds:

Computing Clouds Reference Model

Computing Clouds Reference Model

These requirements truly bring the convergence of the Next Generation Network and the next generation computing infrastructure.   Today’s IT infrastructure with its server, network and storage silos, myriad management systems and resulting human latency associated with end-to-end service management does not meet the requirements for the cloud services.  What is missing is the Service Provider assuring availability, performance and security SLAs and providing end-to-end visibility and control to meet the requirements of both the service developers and end users.  The service developers will then be able to develop logical servers with virtual resources and develop truly distributed managed services fabrics [4, 5, and 6].  A technology with profile based dynamic connection management between consumers and suppliers of resources using a distributed computing model has applicability beyond fixing the IT infrastructure management.

Putting lipstick on current IT infrastructure with incremental approaches will not yield leadership either.  Convergence of Next generation Computing Clouds and the Next Generation Network must undergo an architectural transformation to gain 10X productivity improvement. Transformation from COBOL programming to Object Oriented computing, analog to digital switching and landline to wireless communications are just such examples that occurred within the past two decades since I wrote my thesis on Japanese management.

Cloud computing might just be the driver for the Japanese companies to regain their lost momentum in becoming Number 1 providers of IT infrastructure as Sekizawa had dreamed [7].  It also might be a driver that will radically improve the performance of the Kasumi ga Seki cloud and the Japanese government may well go back to their strength in orchestrating nationwide innovation by betting on the right technologies with long term vision as opposed to the West’s teaching of letting the free market short term greed from wall street gamblers do its own thing (which has not worked out so well as shown by recent developments).  It is ironic to note that a pure technology company Sony has been transformed into a movie gambling company in the name of hardware and software convergence through western influence while Steve Jobs by sticking to his knitting, brought technology to Hollywood through Pixar.  Steve Jobs also has demonstrated with Apple that long term vision and no-compromise innovation do succeed in the market place proving many Wall Street pundits wrong on their predictions.  People who remember Newton and NeXT computer will appreciate Iphone.

During the past two decades, while the Japanese Management surrendered their leadership in telecommunications and their ambitions for IT dominance, they have managed to preserve their R&D labs. The innovations required for creating next generation computing clouds are telecom grade hardware with dynamic provisioning, mass market consumer devices and large scale coordination of different groups (Kyozon Kyoei discussed in my thesis [8]) to create infrastructure fabric and the management services fabric – all of which were strong suits for Japanese management at least in the past.  Vertical integration (still preserved in Japan) of chip to system to solution development may well be an advantage for quick-to-market prototypes if the uchi and soto barrier can be overcome.

Is the current Japanese management bold enough for creating new computing cloud infrastructure leadership?  Who will be the next Sekizawa or Sekimoto who will lead the next generation computing clouds?  Will the mighty MITI gain its leadership again?  If the current generation of Japanese management is not up to the challenge, will the old masters come out of their retirement to restore the old glory?  Will the “Japan Inc” of the eighties rise again or will it become a mere footnote in history’s pages?  I am looking for input, comments and collaboration and invite anyone with appropriate insight to participate in an update to my thesis of almost twenty years ago.

[1] Bill Kelly, “Japan’s Empty Orchestras: Echoes of Japanese Culture in the performance of Karaoke” in “The worlds of Japanese popular culture”, Cambridge University Press, edited by Dolores P. Martinez, 1998. p76

[2] Mikkilineni, R and Pugmire, G, “The connection between profit and services in the next-generation network”, Annual Review of Communications V 54, 521, 2002

[3] Mikkilineni, R and Sarathy, V “Cloud Computing and Lessons from the past”, 18th IEEE International Workshops on Enabling Technologies: Infrastructures for Collaborative Enterprises, WETICE 2009, Groningen, The Netherlands, 29 June – 1 July 2009, Proceedings. IEEE Computer Society 2009, p 28-32

[4] P. Goyal, “The Virtual Business Services Fabric: an integrated abstraction of Services and Computing Infrastructure,” in Proceedings of WETICE 2009: 18th IEEE International Workshops on Enabling Technologies: Infrastructures for Collaborative Enterprises, 33-38 (2009).

[5] P. Goyal, R. Mikkilineni, M. Ganti, “Manageability and Operability in the Business Services Fabric,” in Proceedings of WETICE 2009: 18th IEEE International Workshops on Enabling Technologies: Infrastructures for Collaborative Enterprises, 39-44 (2009).

[6] P. Goyal, R. Mikkilineni, M. Ganti, “FCAPS in the Business Services Fabric Model,”  in Proceedings of WETICE 2009: 18th IEEE International Workshops on Enabling Technologies: Infrastructures for Collaborative Enterprises, 45-51 (2009).

[7] Lyne, Jack. “Fujitsu PC Corp. CEO Akio Hanada: Coming to America . . .And Just about Everywhere Else” 1996,  http://www.developmentalliance.com/docu/pdf/43354.pdf

[8] Rao Mikkilineni, “Current Issues in Japanese Management – Is Japanese Software Thrust As Powerful As Their Hardware Thrust?”, Thesis submitted to Japan American Institute of Management Science, Honolulu and Sophia University, Tokyo, 1990


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: