Use Case: Experimentation in Big Data

Our Focus

Sigma Systems, a pioneer in the field of big data. It regularly works with datasets consisting of upwards of 100 million rows.

Background Brief

Challenges

Sigma Systems needs to conduct extensive, in-depth analytics across a large range of datasets, each of various sizes but all using different data models and consisting of enormous raw quantities of data.

The company currently has only a loose general idea of the relevance of each dataset, necessitating significant experimentation. The quantity of data is such that all processing must occur in the cloud as the compute resource requirements simply do not exist elsewhere. 

The analyses will be performed using traditional methods initially, but Sigma is eyeing the use of AI & ML at a later stage. Due to the enormity of the data it’s agreed within the company that the analytics should be performed without copying it back and forth to the degree it’s reasonably possible.

With The conventional Cloud Solution

Initial market research concludes that all public cloud providers will be able to deliver the necessary compute resources. However upon closer study it becomes apparent that using any of them will necessitate copying all of the data to the provider’s storage.

In addition there is a steep learning curve to understand how to properly configure and manage all the relevant cloud accounts and resources.

Cost projections indicate that the use case would furthermore be impossible to find within any reasonable budget.

With a Cloudshift21 Cloud CapsuleTM

The Cloudshift21 Capsulized CloudTM was originally conceived as a solution for this very scenario—within the highly demanding telco space where data protection is a critical necessity and volumes are enormous.

Utilizing their Cloud CapsuleTM Sigma Systems is able to blueprint a viable solution for their use case, leveraging the following key features:

  1.  With a Cloud CapsuleTM, Sigma’s actual program code is able to reside within a shared instance, minimizing costs. One user account is likely enough to do all the initial programming and testing—only requiring additional resources once the full datasets are ready to be analyzed.

  2. Critically, all of Sigma’s data can now reside on-premise as their Cloud CapsuleTM has robust extensibility via APIs. With that, Sigma can process the data they keep locally.

  3. The extensive learning curve Sigma identified as unavoidable with the major cloud providers is no longer a factor once they spin up their Cloud CapsuleTM. The Capsulized CloudTM  comes fully pre-configured by default. Work can begin immediately after registering for the service and signing into their user account.

  4. Every instance of the Cloudshift21 Capsulized CloudTM comes with the powerful Jupyter notebook manager which runs on top of the exceptionally capable Apache Spark engine. This enables Sigma’s developers to use a variety of different languages, in addition to leveraging AI & ML libraries.

  5. If access to their enormous datasets turns out to be too cumbersome to manage over time, Sigma now has the ability to migrate their entire Cloud CapsuleTM—containing their complete cloud stack—to their own self-managed infrastructure. This can be done without any software or service rewrites, as the platform is fully plug-and-play.

Ready to see for yourself?

Sign up to stay
up-to-date

Register for the Capsulized CloudTM open beta

Register for the Capsulized CloudTM
open beta

Beta users get 2 weeks free PLUS early access pricing that's up to 67% off our already radically affordable rates. Leave your info below to secure your spot!

Download an exclusive white paper analyzing the issues within the cloud industry & learn about the technological innovations behind the

Capsulized CloudTM

Capsulized CloudTM