Bigdata research

the first please check attached file what doc about Bigdata platform.

I need your advice and suggestion considering items as follows.

We don't want to handle cluster design and/or administration. (Only considering SaaS & Managed-style Platforms).

We will give up real-time processing. (Instead, we'll calculate via batch processing with data taken from summary tables in a RDBMS-type system, such as MySQL.)

Since we're moving with batch processing as a prerequisite, using ad-hoc clusters is perfectly fine. (Simply, we only want to use resources at the time of the batch process. After that, the server can lapse with no issue.)

We would like to remove full-text search from the scope of this project. (Focus purely on log file analysis.)

We would like to reduce running costs as much as possible.

Simple implementation. (Currently, we use fluentd to stream our log files into S3. Please keep this in mind for compatibility's sake.)

In consideration of the above, we would like you to limit the suitable services of this project down to around 3.

(Moreover, it would be very helpful if you could give a simple breakdown of your reasoning.)

After you've limited this down, please propose the points of comparison that you'll be considering.

In the determind stage above,we will request the work to you valuation and verification

about BigData Platform again.

regards.

Big Data Sales Entrada de datos Procesamiento de datos Investigación

Nº del proyecto: #6854530

Sobre el proyecto