Mike Gualtieri, Principal Analyst with Forrester Research joined us for a webinar titled Productionizing Hadoop: Seven Architectural Best Practices. Following the webinar, Mike answered a number of questions from participants, including a question about productionizing Hadoop.
Q: How do you move from the test phase to productionizing Hadoop? How do you put it into practice?
MG: There are several issues to think about when putting Hadoop into production. In essence, it comes down to implementing the proper architecture. There are some basic universal objectives that you have to be able to achieve to put Hadoop into production. Make sure you understand the following 7 Architectural Best Practices and have an solution for each one of these:
- Experience: Users' perception of the usefulness, usability, and desirability of the application. How easy is it to operationalize and manage Hadoop in a production environment?
- Availability: The readiness of the service or application to perform its functions when needed. High-availability strategy and architecture are often overlooked in proof-of-concept projects.
- Performance: The speed to perform functions to meet business and user expectations. You need a cluster that performs well across disparate workloads, and is able to handle multiple workloads simultaneously.
- Scalability: The ability to handle increasing or decreasing volumes of transactions, services and data.
- Adaptability: The ease with which an application or service can be changed or extended. How quickly can you reconfigure or provision new clusters or new data within existing clusters?
- Security: The ability to support the security properties of confidentiality, integrity, authentication, authorization, and nonrepudiation.
- Economy: Being able to minimize the cost to build, operate, and change an application or service without compromising its business value. Every architectural decision has an impact on the ROI for Big Data analytics platforms.