Case Studies

Case Study: The Slowed-Down Startup

After enjoying initial success, a mid-sized tech company in the San Francisco Bay Area began faltering in the face of bold competition and an increasingly crowded market. The company’s R&D was encumbered by numerous shortcuts taken to patch up issues and could no longer deliver value as rapidly as their customers demanded.

In our work with the CEO we pinpointed the essential elements for his operation to regain its innovative spark. We realigned the priorities of the business, shifting investment away from underperforming areas and toward a streamlined R&D process. We helped the client take the unprecedented step of ceasing new R&D efforts for a period of six months in order to clean house. Along the way, we ensured that short-term thinking did not imperil these investments. After this period we restored normal operation.

As a result, within ten months the company quickly regained its position as an innovator in the market. The company achieved a 90% reduction in time to market and began delivering new value to customers at a pace unrivaled by their competitors. As a result of its improved reputation as an innovator the company was able to increase sales dramatically and attract new talent.

Case Studies

Case Study: Cloud Services at KT

In late 2010, KT, South Korea’s second-largest mobile phone network operator, was in the process of launching three IaaS clouds and transforming itself into the first cloud service provider in Korea. Executive management, middle management, and line staff were unfamiliar with the ramifications of this transformation and sought the advice of a seasoned consultant and cloud expert.

KT engaged the services of Orchestratus, led by Shlomo Swidler. Shlomo acted as IaaS Product Manager and Cloud Application Architect. Shlomo diagnosed these specific challenges:

  • An unfamiliar target market. IaaS is consumed directly by system administrators and developers – a very different audience than KT’s usual customer, mobile phone consumers and corporate IT departments. KT needed to ensure the cloud services would appeal to its potential customers, and to ensure it could effectively sell to this audience.
  • Appropriate tools, processes, and skills. Operating IaaS demands updated tools, processes, and skills in order provide reliable service – which KT lacked.
  • Technical expertise. KT needed to migrate existing applications into its clouds in order to benefit from the cloud’s elasticity and agility and in order to develop their own PaaS offerings, but lacked the expertise to do so.

In order to ensure the service would appeal to its potential customers Shlomo shifted R&D’s priorities to focus on the areas with greatest customer impact. He educated KT’s sales and pre-sales engineering teams to understand the cloud customer’s concerns. To internal and external developers, Shlomo evangelized cloud-appropriate application architecture and solved application-specific design issues for application scalability and elasticity. He mentored R&D teams to guide the design and development of PaaS services. Shlomo also set guidelines for the most reliable service level guarantees KT could provide.

With Shlomo’s help, the KT Cloud Service Business Unit successfully released its IaaS services on schedule and won significant customer adoption. Hundreds of employees were trained to operate the service, to support customers, and to sell effectively. High-profile applications were successfully reengineered to use the cloud service, saving millions of dollars in hardware costs. And KT’s reputation as the leader in Korean cloud was firmly established.

Cloud Developer Tips

Fragment of heretofore unknown Tractate of Babylonian Talmud discovered

Ancient wisdom apparently has much to offer modern cloud application architects. This fragment was discovered in a shadowy basement in the Tel Aviv area of Israel.

MasechetDBKammaSee a PDF of the fragment

This finding clearly shows that ancient cloud application architects in the great talmudic academies of Babylon struggled with the transition away from classic databases. At the time, apparently, a widely used solution was known as Urim veTumim (“oracle”). Yet this database was unsuited for reliable use in cloud applications, and the text explores the reasons behind that unsuitability.

Okay, here’s the real story: I created this for a client in 2011, and I was delighted to find it on my computer serendipitously today. It reflects the state of the art at the time. Translation into plain English:

1. Oracle RAC does not run on EC2

2. Achieving Oracle high availability on EC2 is a problem: there is no shared device, and relying on NFS is problematic.

3. The cloud frameworks (enStratus, etc.) do not currently support Oracle.