Industry perspective on Cloud and big data research
0 likes71 views
The document outlines key research areas in cloud and big data, emphasizing the need for applied research to meet business requirements in emerging economies. It discusses the integration of big data with cloud technologies, highlighting challenges in managing data infrastructure and potential research focuses such as virtualized performance, SD-WAN applications, federated machine learning, and streaming algorithms. The author encourages contributions to open-source projects to maximize the impact of research findings.
Industry perspective on Cloud and big data research
- 1. Cloud and Big Data Research Syed Affan Ahmed • Dec 2019 bigC 2019 A view from the Industry
- 2. About myself PhD. in CS, from USC, Associate Prof. at FAST-ISB Director Engineering at PLUMgrid, Head of Engineering at AN10 Senior Consultant at Emumba Member National Steering Committee for NCBC, Heading the OpenStack Pakistan User group
- 3. Overview - Soap-box! - Cloud and Big Data - Research areas to focus
- 4. Funding Academic Research, in “emerging economies”
- 5. Industry chooses the research to meet a business need always a lag between innovation and the business case
- 6. applied research … is to gain knowledge or understanding necessary for determining the means by which a recognized need may be met basic research …is to gain more complete knowledge .. of observable facts, without … products in mind. Development is systematic use of the knowledge …. from research, directed toward the production …. and development of prototypes…. It excludes quality control, routine product testing, and production.
- 7. Cloud and Big Data
- 8. Why discuss together? Big Data can solve some problems ….. even better identify some unknown ones Managing the Big Data Infrastructure is … not easy
- 10. Areas for possible research focus Virtualized Performance SD-WAN and its Applications Federated Machine Learning Streaming Algorithms … that should have quick impact
- 11. For greatest impact, contribute research to opensource projects!
- 12. Virtualized Performance Performance optimization: full-stack hardware, network stack, software High packet throughput, low jitter DPDK, SRIOV, XDP, Smart-NICs DCN fabric and optimized protocols Neutral/rigorous measurement studies https://www.netmagicsolutions.com/blog/datacenter-virtualization
- 13. SD-WAN and Applications over SD-WAN https://www.cdnetworks.com/enterprise-applications/sdwan/ Allows innovation in the WAN space Opportunity: No Opensource or standard!
- 14. Federated Learning (edge ML)Security Compliance Privacy Performance
- 15. Streaming Algorithms Makes processing large volume data possible New and efficient implementation of algorithms Handle real-world scenario (loss, corruption)
- 16. It's a never ending story… but I will stop for now!